A Literature Review of Inter-Organizational Sustainability Learning

Abstract

Sustainable development goals (SDGs) have become increasingly important for today’s firms as they build sustainability strategies that integrate SDGs into their core activities. Addressing these goals collaboratively, in line with SDG 17—partnerships for the goals, has gained momentum, hence the growing literature on sustainability-oriented partnerships. However, addressing SDGs through partnerships is not straightforward. For firms, contributing to SDGs through alliances and partnerships requires building environmental capabilities and embracing new value frames; in other words, going through the complex process of inter-organizational learning. This paper reviews the literature on sustainability-oriented partnerships with a focus on the inter-organizational learning process. As a result of the review, a model of inter-organizational sustainability learning is presented. This model captures the different levels and types of the inter-organizational learning process; partner and partnership characteristics that impact learning; the environmental conditions that set the conditions for learning to take place; the catalyst and inhibitors of learning; and finally outcomes of learning. This model expands and re-organizes the existing scholarly conversation about inter-organizational learning in the context of sustainability-oriented alliances and partnerships and offers a learning-based understanding of sustainability partnerships to practitioners. Based on the review, the paper proposes ideas for future research and contributes to the development of a future research agenda in the area of sustainability-oriented alliances and partnerships.

1. Introduction

Sustainable development goals (SDGs) “aim the combination of economic development, environmental sustainability and social inclusion” and they can only be addressed with the efforts of the private sector [1]. According to some scholars [2], these goals “present both a significant opportunity and a significant challenge [for the private sector]: an opportunity as it brings the benefits of additional finance, technology, skills and innovation from the business sector; and a challenge in that it bestows unprecedented power and expectations on business as a development agent purposely seeking to deliver sustainable development outcomes.” Overcoming these challenges requires the development of capabilities that address and integrate sustainable development into the core business and also a deeper engagement with value frames that promote sustainable development [3,4]. Partnerships can facilitate a platform to address complex and systemic issues highlighted in the different SDGs collectively [5,6].

SDG 17—partnerships for the goals, invites the private sector to implement SDG 1–16 through collaboration with other societal actors to create value for nature and society by sharing knowledge, expertise, technology and financial resources [7]. In their seminal work, Austin and Seitanidi [8] discuss four different types of value generated as a result of collaborative work: associational value, transferred resource value, interaction value and synergistic value. The authors define associational value as “a derived benefit accruing to another partner simply from having a collaborative relationship with the other organization”; transferred resource value as a “benefit derived by a partner from the receipt of a resource from the other partner”; synergistic value as “value that arises from the underlying premise of all collaborations that combining partners’ resources enables them to accomplish more together than they could have separately” [9]. While partnerships for SDGs are expected to create these four different types of value, this article specifically focuses on the interaction value which is defined as “the intangibles that derive from the processes of partners working together”. One of these intangibles is considered as inter-organizational learning, which is the focus of this article [9].

Collaborative partnerships trigger inter-organizational learning processes which lead to the development of new capabilities that would help businesses to address sustainability concerns internally [10]. Furthermore, they enable cognitive changes in the private sector to embed sustainability into the core through frameshifts that take place in such collaborations [11,12]. Indeed, Agarwal, Gneiting and Mhlanga [2] proposed that partnerships with various stakeholders can help firms “shift from using a narrow business case approach to aligning their core activities with broader societal values and interests.”

This paper positions itself in this area of inter-organizational sustainability learning that takes place in alliances and partnerships formed to tackle SDG-related challenges. The objective, herein, is to re-organize the pre-existing work on sustainability-oriented alliances and partnerships with a focus on the inter-organizational learning process, its antecedents, and its outcomes. The paper, therefore, uses a review of 122 academic articles to provide a comprehensive review of the field.

The review results in a model which includes the following categories: partner and partnership characteristics that impact the learning process; the environmental conditions that set the conditions for learning to take place; the catalyst and inhibitors that impact learning; and finally firm-level, partnership-level, and system-level outcomes of learning and partnerships. This model organizes the existing literature on inter-organizational sustainability learning. Furthermore, the model demonstrates how different theoretical approaches and concepts fit together, as opposed to competing, when it comes to explaining the process of inter-organizational sustainability learning.

This paper contributes to the literature on sustainability-oriented alliances and partnerships. This contribution is thanks to the model and the future research agenda built in the paper because of the synthesis of different theoretical approaches and concepts that help us explain the complex phenomenon of inter-organizational sustainability learning. Furthermore, the model also contributes to the work of sustainability practitioners who manage alliances and partnerships by demonstrating factors that help them enhance inter-organizational learning and partnership performance.

The remainder of this paper is structured as follows. The Theory section summarizes the literature on SDGs, partnerships for SDGs and inter-organizational learning in the business context. The following Methods section introduces the stages of the systematic review, as well as the role of the metaphor in re-organizing the existing literature. The Findings section introduces a model that shows the antecedents and outcomes of the inter-organizational learning process in the context of sustainability. The Future Research section outlines research gaps in the field based on the review and provides a path for further research. Finally, the Conclusions section draws on the contributions of this paper to theory and practice.

2. Theory

2.1. Sustainable Development Goals (SDGs) in the Business Context

Sustainable development goals (SDGs) define 17 global targets for all types of organizations to address sustainability issues ranging from healthcare to fighting inequalities to climate change [13,14]. SDGs are positioned to cover the triple bottom line of sustainability (economic, environmental, social) and address concerns to do with people’s well-being, planetary boundaries and an inclusive notion of prosperity [6,14], and equality [15,16].

SDG 17—partnerships for the goals, emphasizes that the first sixteen goals need to be addressed by different constituents of the system [17]. In other words, SDG 17 recommends partnerships as a tool to foster sustainable development, underlining the importance of interactions between different societal actors. Sachs, et al. [18] highlight that achieving SDGs require societal transformations that “engages a different subset of business and civil society, facilitating targeted problem-solving, clear communication and the mobilization of stakeholders.” Similarly, Horan [19] recognizes that various stakeholders would be involved in these partnerships with different perceptions, expectations and interests. Indeed, thanks to the cross-sector interaction space in partnerships, actors from across different sectors can generate different types of value. While this article acknowledges the potential of value creation for different societal actors in a partnership context, the article specifically focuses on the ‘private sector’ and interaction value of ‘learning’ that private sector actors can gain through partnerships.

The private sector has already played an essential role in determining what SDGs should be [2,20,21,22]. Moving forward, the private sector has a crucial role in addressing the goals, re-designing business models, developing capabilities, accommodating resources and shifting their mission from profits to the wellbeing of the planet and people [2,20,21,22]. SDGs already impact the private sector, as we see some firms’ innovative products and services address SDGs, while other firms are changing their business models to align their core business with the SDGs [20]. It is important to note that some SDGs are relatively easier for the private sector to address, such as SDG 8—sustainable and inclusive growth, while other SDGs may be too complex to be addressed by a single actor alone [6].

Building sustainability-oriented partnerships, SDG-17, only became a goal in 2015 when the agreement on the SDGs was made. However, research on sustainability-oriented partnerships has been growing since the 1990s, both in the social [23,24] and the environmental sustainability domains [25,26,27,28]. Recent studies on SDGs in the business context suggest that a “way to enhance the strategic relevance of the SDGs is to engage in a proper portfolio of cross-sector and intra-sectoral coalitions or partnerships” [6]. These portfolios consist of two distinct types of collaborations: inter-firm alliances and cross-sector partnerships [29,30].

2.2. Inter-Firm Alliances and Cross-Sector Partnerships

Inter-firm alliances are defined as “a form of organizational arrangement for ongoing cooperative relationships among firms” [31]. In other words, alliances are “voluntary arrangements [between firms] involving durable exchange, sharing, or co-development of new products and technologies” [32]. Alliances are viewed as a hybrid form of governance “between markets and hierarchy that occur when transaction costs associated with a specific exchange are too high for an arm’s-length market exchange but not high enough to mandate vertical integration” [32,33]. This explanation sets the transaction cost economics motivation for formations of inter-firm alliances [34,35]; however, this is not the only rationale for why alliances exist.

Others explain how alliances help firms access resources [36], acquire knowledge [37], and develop new capabilities [38,39], thanks to inter-firm interactions. Furthermore, alliances help firms comply with institutional norms, values and regulations, and legitimate their actions using alliance arrangements [40]. Inter-firm alliances can occur between suppliers and customers in the firm’s value chain [41,42], with competitors [43] or various other innovation partners from different industries [44]. They may take a form whereby the partners share equity, as in the case of joint ventures, or a non-equity form whereby partners have a legal agreement, such as a memorandum of understanding (MoU), which clarifies the roles of partners and the boundaries of the alliance [34,45].

Cross-sector partnerships are “vehicles to mediate the changing roles and perceived responsibilities of what are commonly referred to as the three primary institutional sectors of society: government, business, and the civil sector” [46]. Firms engage in cross-sector partnerships with similar motivations to that of inter-firm alliances. Existing literature also studies cross-sector partnerships through the transaction-cost economics lens [47], resource and knowledge-based views [48], capabilities [49], and finally, compliance with institutional norms, values and regulations, and legitimation of their actions [50].

This paper reviews the literature on both inter-firm alliances, and cross-sector partnerships in the context of sustainability since research shows that they both play a crucial role in firms addressing SDGs [6]. However, it is crucial to highlight that there are some differences in inter-firm alliances and cross-sector partnerships.

First and foremost, the partner type; in other words, the heterogeneity of partners’ organizational forms, resources and capabilities, are different in these two distinct categories of engagement [51]. Due to the embeddedness of both partners in the private sector, inter-firm partners are often conceptualized as constituting lesser diversity in comparison to cross-sector [11,51].

Second, it is often assumed that public, private and civil sectors are dominated by different logics [50,52,53,54,55,56]. The differences in dominant logics lead partners to focus on different value objectives and introduce further challenges for the partnership [57,58].

Third, other than the differences in organizational forms, resources, capabilities, institutional logics and value objectives, some cross-sector partnerships are often identified with altruistic partnership motivations, aiming to impact systemic and societal grand challenges that go beyond self-interest [11]. However, this may not apply to all cross-sector partnerships, as some may not focus on sustainability challenges [59]. Though there are differences between inter-firm and cross-sector partnerships, they both provide a room for learning, development, and change.

2.3. Inter-Organizational Learning in the Business Context

Organizational learning and learning organizations have long been discussed in the domain of organization studies [60,61,62,63,64,65,66,67,68]. This literature discusses how organizations store knowledge in their memory [69]. Organizations absorb knowledge from external knowledge sources and transfer such knowledge internally [70]. They learn to change their behaviors, develop new capabilities and even alter the embedded collective cognition and, as a result, they shift towards new mental models [71]. Organizations need to learn because they need to fit the external environment [72]; in other words, co-evolve with the institutional environment and settings [73].

The literature on organizational learning could be categorized into two realms depending on the knowledge or value sources [74]. If the learning takes places across different teams and functions within an organization, then this learning or knowledge transfer is often referred to as intra-organizational [74]. If the knowledge sources are external to the organization, such as in the case of networks, alliances, consultants, suppliers and customers, then this learning is often referred to as inter-organizational learning [74]. The focus of this paper is on the latter, inter-organizational learning, and there are several ways in which it differs from intra-organizational learning [75].

Inter-organizational learning is often described as a paradoxical process due to the competitive tensions between the knowledge partners [76,77,78]. If two inter-firm partners are engaged in an alliance, scholars observed firms engaging in a competition to outlearn their partner. Indeed, some argue that “creating a successful alliance learning environment is the exception rather than the rule” [79].

Inkpen [79] lists several “explanations for the failure to learn from a partner: the alliance knowledge was undervalued; the necessary knowledge connections’ were not put into place; the nature of the knowledge itself made learning difficult; the parent corporate culture did not support learning.” Some studies focus on other barriers of inter-organizational learning and highlight that the “fear of loss of ownership, fear of loss of control of knowledge, and fear of loss of competitive edge” negatively affect the creation of a learning environment in an inter-organizational setting [80]. In a similar vein, others highlight that “the dynamics of power, opportunism, suspicion, and asymmetric learning strategies can constitute processual barriers to collective knowledge development” [81]. Generally, explicit knowledge is easier to be acquired from a partner than tacit knowledge [78]. However, “if there is access to, and recombination of, diverse knowledge in a network, it might be difficult to establish barriers to protect the competencies that each network member has in various knowledge fields” [78]. Moreover, the power imbalance between engaged parties may also create difficulties in establishing “inter-firm routines and the sharing of knowledge” [78].

In response to the barrier of partner opportunism and the fear of losing a competitive edge, the literature proposes governance mechanisms to ensure the protection of parties and social capital to improve their relationship. Two distinct forms of governance are relational governance and contractual governance [82]. Scholars highlight that when partners have relational governance based on competence trust which refers to “the confidence in the abilities of the other party to perform its share of the workload in an exchange”; then they are more likely to be engaged in learning [82]. Besides, governance through formal contracting also aims to protect parties and outline the roles and responsibilities of each party involved; hence “formal written contracts accomplish learning objectives by specifying the obligations and expected duties of partners” [82].

Scholars have identified that social capital, defined as “the aggregate of resources embedded within, available through, and derived from the network of relationships possessed by an individual or organization” is also an essential factor in catalyzing the learning process [83]. Indeed, some propose that in networks, thanks to the availability of social capital between network members, inter-organizational learning is facilitated further in comparison to that of inter-firm alliances [83]. Scholars propose that trust impacts the commitment between partners positively [84].

Social capital becomes important in inter-organizational transfer, not only because of possible partner opportunism but also because access to knowledge is more difficult outside the firm boundaries. Outside the firm boundaries, there is an additional issue that will affect the learning performance: cognitive distance or proximity [85,86,87,88].

Cognition “denotes a broad range of mental activity, including proprioception, perception, sense-making, categorization, inference, value judgments, emotions, and feelings, which all build on each other” [87]. Differences between organizations in terms of cognition lead to cognitive distance amongst partners [87]. Nooteboom shows that there is an inverted U-shaped relationship between cognitive distance and innovation performance [87]. This means that there is an optimal cognitive distance and “the challenge then is to find partners at sufficient cognitive distance to tell something new, but not so distant as to preclude mutual understanding” [87].

Cognitive distance shows that firms learn something new if they already have an existing knowledge and value base that will allow them to learn, which is a concept called ‘absorptive capacity’ [70,77,89,90]. Absorptive capacity is a dynamic capability that consists of the following processes: recognizing the value of new, external information, assimilating it, and applying it to commercial ends [71]. As a dynamic capability, absorptive capacity helps to create, extend or modify a firm’s resource base and develops other organizational capabilities [71,90]. Therefore, there is “a recursive relationship between organizational learning and absorptive capacity, where increased learning in an area can enhance a firm’s knowledge base and help to build greater absorptive capacity, which in turn can improve learning” [91].

Absorptive capacity is necessary but not sufficient for inter-organizational learning to take place. For knowledge transfer to take place between partners, partners would also need to have a disseminative capacity, defined as the “ability of knowledge holders to convey knowledge in a way that a recipient can comprehend it and put it into practice” [92]. In other words, disseminative capacity is “a combination of the sender’s ability to codify and articulate knowledge, the sender’s willingness to share knowledge, and the sender’s propensity to create and use opportunities for knowledge acquisition by the receiver” [93].

Above, different conditions for learning to take place in an inter-organizational setting are outlined, and the impact of different factors are briefly summarized. Other than the above-summarized conditions for learning to take place in an inter-organizational setting, it is essential to highlight that not all learning can be conceptualized in the same way. Crucially, learning can be exploratory and exploitative learning [71,94,95,96,97], single or double-loop learning [98]; or higher-level and lower-level [66]. These different levels and types of learning are defined in

Table 1. Levels and types of learning.

Concepts	Definition
Exploitation vs. Exploration	“Exploration includes things captured by terms such as search, variation, risk-taking, experimentation, play, flexibility, discovery, innovation. Exploitation includes such things as refinement, choice, production, efficiency, selection, implementation, execution” [68]. In other words, exploration is “the pursuit of new knowledge, of things that might come to be known”, exploitation is “the use and development of things already known” [99].
Single-loop vs. Double-loop	While single-loop learning occurs “whenever an error is detected and corrected without questioning or altering the underlying values of the system”, double-loop learning occurs “when mismatches are corrected by first examining and altering the governing variables and then the actions” [100].
Lower-level learning vs. Higher-level learning	“Lower-level learning occurs within a given organizational structure; a given set of rules. It leads to the development of some rudimentary associations of behavior and outcomes, but these usually are of short duration and impact only part of what the organization does. It is a result of repetition and routine and involves association building.” “Higher-level learning, on the other hand, aims at adjusting overall rules and norms rather than specific activities or behaviors. The associations that result from higher-level learning have long-term effects and impacts on the organization as a whole. This type of learning occurs through the use of heuristics, skill development, and insights. It, therefore, is a more cognitive process than lower-level learning, which often is the result of repetitive behavior” [66].

.

Based on the learning levels and types summarized in Table 1, it is possible to conceptualize two distinct categories of learning outcomes: changes in capabilities, routines, and organizational behavior and changes in mental models, values and beliefs [101].

For the first category of outcomes, this paper focuses on the concept of organizational capabilities which can be defined as the “existing repertoire of the possible actions of the groups and organizations” [86]. Otherwise, the “routinized processes that are embedded in the organization” [102]. Depending on the level and type of learning, organizations can refine and leverage existing capabilities or develop new capabilities using the knowledge acquired from partners [103,104].

For the second category of outcomes, this paper focuses on the concept of ‘value frames’ which refers to perceptions of value that guide different sustainability organizational level interpretations and priorities in terms of economic, environmental and social value creation and preservation [4,12,52,105]. Through higher-level learning, changes or shifts in value frames are also expected [106,107].

The newly developed capabilities or shifted frames are expected to improve a firm’s performance in two ways. First, it may help firms develop combinative capabilities that help a firm’s general innovativeness and, as a result, its competitiveness [73]. Second, firms would also develop capabilities to manage alliances [38,108,109] and alliance portfolios [110], which would improve their partnership performance or success.

3. Methods

The objective of this study is to re-organize the pre-existing work on sustainability-oriented partnerships and alliances with a focus on the inter-organizational learning process, its antecedents, and outcomes. Therefore, this article takes a systematic approach to review the existing literature in this area.

A systematic literature review research helps to identify, evaluate and synthesize the existing body of completed and recorded work produced by scholars in a systematic way guided by a reproducible method [111]. The review is guided by the PRISMA checklist, which identifies the steps that researchers need to take when conducting systematic literature reviews [112,113]. The introduction section has identified the rationale behind the review, and the theory section developed the theoretical basis of the review. The focus of this section is to explain how the review is conducted.

The review includes the following stages: searching for academic articles in databases, screening the articles found based on an inclusion and exclusion criteria, processing the selected articles through qualitative content analysis and coding, synthesizing, and presenting the review findings.

Table 2. Summary of searching and screening.

Stages	Result
Search in the Web of Science	5688 articles
Inclusion Criteria	1655 articles
Exclusion Criteria	75 articles
Snowballing	47 articles
Total	122 articles

summarizes the searching and screening stages followed in the systematic review.

In the searching stage, the Web of Science database is selected to gather articles in the field. Web of Science provides access to a wide range of journal articles that are both within business, management and sustainability domains and is commonly used for systematic review purposes [114]. It is necessary to identify keywords that will make up the search string to conduct a thorough and comprehensive review of a field [115,116]. The critical keyword categories for the searching stage are identified as learning, partnerships, and sustainability. The keywords in each category are selected in line with previous studies. For instance, for learning keywords such as knowledge development, knowledge acquisition, knowledge absorption, are also searched in line with prior studies in the field [10,89]. For sustainability, keywords such as green and eco-friendly, or social responsibility are used to cover the broad literature in the field. Even though the terms have slightly different meanings, scholars’ use of these terms have been converging, and at times they have been used interchangeably [117]. Finally, for partnerships, keywords such as alliance, cooperation, partnering, and collaboration are also used [118,119]. As a result, the search string below is generated with AND/OR Boolean operators:

(“sustainability” OR “sustainable” OR “CSR” OR “corporate social responsibility” OR “green” OR “eco-friendly”) AND (“collaboration” OR “collaborative” OR “partnership” OR “partnerships” OR “partners” OR “partnering” OR “partner” OR “cooperation” OR “alliance” OR “alliances” OR “joint venture”) AND (“knowledge development” OR “knowledge absorption” OR “absorptive” OR “capability development” OR “frame shift” OR “knowledge transformation” OR “knowledge exploitation” OR “knowledge assimilation” OR “knowledge acquisition” OR “learning” OR “transformation”)

This string of keywords used to search the Web of Science database for academic articles included that this content be in the English language and that all years are available. This search yielded 5688 articles.

These articles were screened based on two sub-processes: inclusion of articles only from relevant research fields, and exclusion of articles which contain the search string but in a different context. First, as an inclusion criterion, the following four Science and Social Science Index categories are selected to provide coverage of journals that are both in the business and management and sustainability fields [10]: Green and Sustainable Science Technology, Environmental Studies, Management and Business. Application of this inclusion criterion yielded a sample of 1655 articles.

Second, as an exclusion criterion, the scope of the current study is used. Within the 1655 articles, some studies referred to ‘sustainability of partnerships’ or ‘financial sustainability of businesses’ within business and management studies but were not about environmental or social sustainability issues. Other studies focused on sustainability partnerships but did not consider partnerships whereby a business actor was involved. Instead, they focused on cities, local authorities, communities and NGOs and their sustainability partnerships with each other whereby business actors were not among the partners or the study did not provide learning opportunities for the business context, which is the focus of this paper. As a result of this screening of 1655 articles, 75 articles were identified from the Web of Science.

Furthermore, to further check if any relevant articles were missed in the searching and screening phases, recent reviews on the topic have been used for snowballing. One of these review articles was about capability development in the context of sustainability, which included a subset of articles that studied capability development through collaborations [10]. Another review article was about value frames, which included a subset of articles that studied frameshifts through collaborations [52]. Finally, one study was providing a general overview of environmental collaborations [120], and another was providing an overview of the role of stakeholder engagement for environmental innovations [4]. After cross-checking the references of these previous review papers, 47 other relevant articles were identified.

Overall, 122 articles formed the review database for this study. The review was conducted using qualitative content analysis and coding on NVivo 12 Plus. Inductive two-stage coding is used on NVivo to identify patterns within the review articles. In Figure 1, the coding process is demonstrated with some examples for the development of each category.

In addition to the coding mentioned above, articles are also coded in the following areas: theories and methods, types of partnerships (inter-firm vs. cross-sector) and SDGs, (see Table in the Appendix A). The coding concerning SDGs was conducted using a study which describes the role of businesses in addressing SDGs [6]. The theories and methods were coded based on the relevant sections of the papers. 36% of articles in the review explicitly referred to the resource-based view, 32% to absorptive capacity, 30% to dynamic capabilities, 17% to institutional theory, and 11% to stakeholder theory. Furthermore, more than half of the articles in the review were qualitative, and mostly case-based (64), followed by 43 quantitative studies and 13 studies that are review or theoretical works and only two studies which employed mixed methods.

Finally, the review was dominated by articles that studied cross-sector partnerships (55), followed by studies that studied both forms of partnerships with various stakeholders (36) and finally, inter-firm alliances (31). Studies that focused on inter-firm alliances were mostly from the context of sustainable supply chain relationships [121,122,123,124]. Only a few studies discussed inter-firm alliances that were not in the supply chain context [125]. Cross-sector partnership studies focused on engagements between firms and governments [126,127], firms and NGOs or non-profits [25,128,129], or universities and research institutions [130]. Furthermore, a few studies focused on firms’ engagement with several societal actors in the same initiative through multi-stakeholder partnerships [131,132]. Besides, recently some authors studied cross-sector partnerships between firms and social or environmental enterprises as cross-sector [133,134].

4. Findings

This section introduces a model based on the review findings as organized in Figure 2. This model includes the following categories: partner characteristics, partnership characteristics, environmental conditions, catalysts and inhibitors, inter-organizational learning process and its outcomes (See Table in the Appendix A to view the articles in the review that contribute to different categories in the model).

This model shows some differences when compared to the inter-organizational learning process that take place in the context of commercial business alliances, which is discussed in the Theory section. First, while the learning literature in the business context discusses both internal and external pressures that motivate actors to learn, in the sustainability context, ‘environmental conditions’ demonstrated mostly isomorphic pressures set by the external environment. Second, there are several feedback loops identified in this review (represented with a thinner curved line in Figure 2). Third, other than firm-level and partnership-level outcomes, in the context of sustainability-oriented alliances and partnerships, system-level outcomes have been identified.

4.1. Partner Characteristics

In his critical review, Wassmer calls partner characteristics as “focal firm-level antecedents” and identifies that existing resources and capabilities, a focal firm’s strategy and existing portfolio of partnerships fall under this category [120]. While these factors are focal-firm level antecedents that explain firms’ entrance into collaborations, based on the review, two critical partner characteristics appear as antecedents of inter-organizational learning: absorptive capacity and disseminative capacity.

Absorptive capacity is the capacity of learning at an organization; in other words, the ability of a firm to assimilate and apply new knowledge successfully to its goals [122]. In the context of sustainability, these goals are not only commercial but also social and environmental. Hence, a firm’s absorptive capacity in the sustainability context is to do with pre-existing alliances and partnerships for sustainability, its sustainability-related management systems and organizational capabilities [135,136].

A recent study expands the notion of absorptive capacity in the area of sustainability from the ability to absorb sustainability-related external knowledge to create economic value to incorporating societal values to create social/environmental value [137]. This study argues that absorptive capacity helps firms to go beyond the acquisition of essential environmental or social sustainability-related knowledge from external sources. At the same time, it explains why some firms are receptive to a broad understanding of value; they articulate consistently their willingness to engage in value creation with a responsive approach [137]. Studies emphasize that absorptive capacity explains how some firms develop sustainable product and service innovations [136,138,139,140,141], proactive sustainability strategies [142], and environmental and CSR practices and capabilities [143].

Absorptive capacity explains why a focal firm would be willing and open to learning in different areas [144,145], and how it can engage in inter-organizational learning using its specific ability to acquire knowledge based on its prior experience [146]. However, inter-organizational learning does not only depend on this focal firm. It also depends on their partner’s ability to teach and disseminate knowledge and values [147]. In the context of environmental collaborations with suppliers, “firms that have a high quality of environmental capabilities disseminate green knowledge to supply chain partners by means of diffusing new capabilities to achieve high efficiency in supply chain processes” [146]. Similarly, in multi-stakeholder platforms about climate change, the dissemination of “information about climate change challenges and opportunities to the participants” helped in “motivating them to start innovating new low-carbon products, services, and business models” [132].

Similarly, Lin [148] gives the example of the collaboration between “the Pew Center on Global Climate Change and the World Resource Institute” which “are working closely with firms to promote and disseminate environmental solutions/technologies.” Through the knowledge and value dissemination that takes place between the partners, she highlights that a partner may shift their mental models and develop sustainable business models that address complex environmental problems proactively. Others focus on the role broker organizations in disseminating knowledge in public private partnerships (PPPs). They highlight that broker “organizations frequently organize problem-related round tables and disseminate PPP best practice cases with an inspirational purpose’’ [126].

In sum, in addition to firms’ existing knowledge, resources, culture and strategies that motivate them into entering alliances and partnerships for sustainability, firms’ absorptive capacity and their partners’ disseminative capacity help explaining the extent of their learning from these collaborations. It should be highlighted that absorptive capacity can be enhanced as a result of engaging in collaborations, developing new capabilities, and gaining new perspectives and values, hence the feedback relationship in Figure 2 [145]. In a similar vein to the literature in alliances in the business context [92,93,149,150], the research on sustainability-oriented partnerships also focus much more on focal firm’s absorptive capacity, and yet much less attention has been given to partner firm’s ability to disseminate knowledge and values.

4.2. Partnership Characteristics

The alliance or partnership characteristics are to do with the bond between the focal firm and its partner. Wassmer refers to this as “partnership-level contingencies” [120]. The review shows that the essential partnership characteristic that impact learning is to do with governance.

Governance, herein, is defined as “the coordination that is characterized by organic or informal social systems” [151]. In other words, it describes “the facilitation and administrative routines” [152]. Governance determines “the structures and processes by which societies share power, shapes individual and collective actions” by introducing “laws, regulations, discursive debates, negotiation, mediation, conflict resolution, elections, public consultations, protests, and other decision-making processes” [153].

Different forms of governance may aim to impact at different levels. For instance, a PPP with local authorities may intend to create a local effect [154,155]. In contrast, initiatives such as the UNGC, the UN Principles for Responsible Investment (UNPRI), and the GRI aim for global impact [151]. Nevertheless, other initiatives such as Extractive Industry Transparency Initiative, Responsible Care, Forest Stewardship Council and Marine Stewardship Council aim impact in specific industrial fields [151,156].

Actors such as governments, firms or NGOs can engage in non-collaborative forms of governance to tackle environmental and social sustainability challenges. A study finds that “firms should first invest in becoming a strongly sustainable firm before investing in external collaboration” since they “will benefit from co-aligned sustainability collaboration with external partners only if they are leading in sustainability practices within their own practices” [157]. Others, however, highlight that “collaborative forms of governance are best viewed as dynamic, problem-solving processes in which learning about social-ecological change is an essential component” [158].

In the context of sustainability, some studies have referred to polycentric governance, whereby there are many authorities involved that act as centers for decision-making [153]. Polycentric governance “creates possibilities for moderating vertical interplay among institutions” [153]. Such governance is often tripartite; meaning that it includes “representatives from businesses, civil society, and the state; therefore, they can be distinguished from more traditional types of alliances, such as strategic alliances between business organizations, social alliances between business and nongovernmental organizations (NGOs), and public-private partnerships” [132].

Other than tripartite partnerships, studies show various types of collaborative governance that can be applied to environmental and social sustainability challenges, including joint ventures [159], licensing [27], and social franchising [160,161], to other forms of network governance [151]. Generally, scholars, in a similar fashion to the commercial business context, differentiate between formal and informal governance [162]; in other words, contractual and relational governance [125,163]. Indeed, while parties may collaborate by abiding by the rules and norms and contract sets, parties may also rely on trust to create synergistic relational rents [157]. To discuss relational governance, studies often discuss ‘trust’. However, since trust is also a relational dimension of social capital, it will be discussed as a catalyst.

Governance affects how the relationship between different parties are organized and therefore, crucially, it has an impact on how the partners can learn from each other. For instance, some scholars note that the distinct equity and non-equity forms of governance impact inter-organizational learning [164]. They highlight that while equity forms require a tight coupling between the parties, non-equity forms are often associated with loose coupling. These scholars propose that the non-equity forms are likely to be associated with exploratory learning with a diverse set of partners; whereas equity forms are likely to be associated with an exploitative learning homogenous set of partners [164]. Therefore, the way in which the partnership is governed would affect the degree and type of inter-organizational learning.

4.3. Context

Environmental conditions are often used to explain why firms need to engage in inter-organizational learning in the first place [120]. In the words of Liu, Esangbedo and Bai [161], “the purpose of organizational learning is to achieve a new understanding of the external environment of the members and organizations through an effective mechanism of formation, dissemination, and sharing”; hence environmental conditions set a rationale for learning. Indeed, one study finds that “companies adapt their strategies and orient them towards CSR to stakeholder pressures also independently from what they are actually learning from them” [138]. This finding shows the crucial impact of the external environment in shaping firms’ organizational behavior when it comes to sustainability issues. For instance, the “environmental catastrophe following the explosion at the Fukushima nuclear plant in March 2011” has led to “growth in anti-nuclear sentiment, a rise in the stock prices of renewable energy companies, and an interest in clean technology and renewable energy firms” [165]. Such external events often motivate companies to engage in various partnerships with stakeholders.

Environmental conditions are often explained using theories such as stakeholder theory [138,166] or institutional theory [133,164,167]. These scholars focus on ‘institutional pressures’ or ‘stakeholder pressures’ that create a form of legitimacy crisis on firms, which then motivate them to enter collaborative relationships and engage in inter-organizational learning. Most scholars refer to Suchman’s [168] seminal definition of legitimacy which is “a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions” [137,156,169]. Drawing on the seminal work of DiMaggio and Powell [170], the review articles discuss three types of isomorphic pressures that lead firms to engage in inter-organizational learning [148,155,167,171,172]: coercive, mimetic and normative.

Coercive pressures are often associated with government policy and regulations about sustainability challenges [139,164,173,174]. For instance, a study highlights how “legal trends, such as the European Union Directives, significantly influence operation systems, product-markets, and business strategies of firms” as well as their collaboration patterns [121]. Another study, on the other hand, highlights how in multinational corporations, depending on the regulatory environment, the absorptive capacity of a subsidiary changes from that shared absorptive capacity of headquarters [143]. Another study highlights that often cross-sector partnerships and multi-stakeholder platforms are positioned to address government failures [120] and the creation of positive externalities [127,174].

Normative pressures may arise from industrial or societal norms [164]. For instance, Lin and Darnall give the example of the participation of electric utilities in the US Department of Energy’s Climate Challenge Program “to collectively improve their public image and reduce the climate emissions for the utility sector as a whole” which helped these firms “to ameliorate normative pressures from their professional networks, and conform to values and social norms exerted by the industrial associations” [164].

Mimetic pressures are to do with the competitive environment of the firms. For instance, one study finds that firms’ engagement in CSR is difficult to imitate by other firms even when there exist conditions for mimetic pressures [167]. The authors highlight that this is because the knowledge that is needed for substantive CSR engagement is sticky. However, the study highlights that such substantive engagement may be facilitated by the selected governance structure, culture, or capability development [167].

It is essential to highlight, however, that coercive, normative and mimetic pressures, of course, affect the actors through their cognitive filters or perceptions. Zou, Xie, Meng and Yang [167] highlight that the perceptions of decision-makers about the institutional pressures faced are shaped by their accumulated experience and knowledge. Therefore, the engagement of firms with a proactive or reactive strategy due to isomorphic pressures also come down to firms’ perceptions of these pressures. This perception, indeed, would both affect the motivation to engage in a partnership and the intention or willingness to learn [175].

Finally, the institutional environment may also impact the kind of alliances and partnerships that focal firms form. For instance, “in industries with diffuse stakeholder pressures, firms that seek to strategically differentiate themselves engage in proactive alliances, e.g., the alliance between Starbucks, a leading specialty coffee company, and the environmental non-profit Conservation International” [145]. On the contrary, “intense public criticism and impending regulatory pressures in the extractive and energy production industries drive reactive alliances as seen between DuPont and the non-profit World Resources Institute” [149]. In sum, the institutional environment impacts the perception of actors, and shapes their intention to partner, and therefore, affect who they partner with and how much they can learn [145].

4.4. Catalyst

In the context of sustainability-oriented alliances, scholars define social capital as “social cohesion and strength of relationships among partners” [176], or as “a set of resources such as trust, norms, and values that are accessed through a network of social relations and can be mobilized to facilitate action” [126]. These studies discuss how social capital and the idea of investing in social relations can bring about new opportunities for actors [177]. For instance, the social capital of social enterprises offers some benefits to corporations such as access to community and local knowledge when they engage in collaborations [134].

Arya and Salk [145] propose that “firms with greater social capital across hierarchical levels within and between cross-sector alliance partners will positively influence learning that can enhance diffusion and integration of codes of conduct into corporate culture compared with firms with lower social capital.” In a similar vein, others highlight that social capital “instil[s] a shared vision and strategic alignments toward common goals and collective outcomes [151]. At an individual level, boundary spanners can facilitate the accumulation of social capital [145]; at an organizational level, on the other hand, a similar facilitation role is taken by broker organizations who connect otherwise unconnected contacts [126,172].

Scholars identified three dimensions of social capital, in line with the seminal work of Nahapiet and Ghoshal [178]: structural, cognitive and relational. Structural dimension “refers to the overall pattern of connections between actors” [126]. These patterns of connections include the “roles, rules for decision-making and communication, procedures, precedents and networks that facilitate mutually beneficial collective action” [177]. Relational dimension “describes the personal relationships people have developed through a history of interactions” [126]. Finally, the cognitive dimension is described as “shared representation, interpretation, and systems of meaning among partners” [126]. Scholars highlight that social capital “may be produced and/or increased” as a result of inter-organizational learning [158].

4.4.1. Structural Dimension of Social Capital: Effective Coordination

The structural dimension of social capital is to do with coordination patterns, roles and responsibilities of parties involved.

Coordination is to do with “communicating potential solutions, setting priorities for particularly promising ones, and assimilating various solutions” [179]. In a partnership setting, different partners would provide different types of resources, knowledge and values, and effective coordination helps actors integrate these different types of resources, knowledge and values to come up with a solution to a sustainability challenge [179]. In the context of cross-sector partnerships, a study finds that cross-team coordination mechanisms have proved helpful by allowing organizations to “understand each other’s unique circumstances’’ and thanks to these coordination mechanisms, organizations were able to fuse different value frames [12].

Sustainability-related initiatives may require changes in the traditional relationships between some partners. For instance, a firm needs to extend or transform an existing relationship with supply chain partners to align itself with the principles of the circular economy [180,181]. Effective coordination mechanisms and establishment of clear roles and responsibilities improve inter-organizational relationships and, more specifically, inter-organizational learning outcomes [131,157]. Notably, an agreement on how collaboration is coordinated, along with how roles and responsibilities are distributed, require attention in the initial partnership design phase [182].

4.4.2. Relational Dimension of Social Capital: Trust

The relational dimension of social capital that impacts inter-organizational learning is trust. Scholars highlight that trust “facilitates the openness for exchange of tacit knowledge, which is relatively difficult to communicate or trade in markets, and durability of relationships, which otherwise may collapse when problems arise between exchanging partners in pure market relationships” [183]. The literature highlights that the stronger the trust between partners, the stronger their ties and the more they can learn and innovate in a partnership [122,183,184].

Scholars have defined trust in different ways and focused on different types of trust. One definition is “a belief, sentiment or expectation about an exchange partner that results from the partner’s expertise, reliability and intentionality, or from the partner’s honesty and benevolence” [122]. In other words, trust is identified as “the expectation that the partner will pursue cooperation, fulfil obligations, and try to maintain the relationship between the other parties” [161].

Scholars differentiated between personal trust which is “extended primarily to another human being” and system trust, which “concerns trust in the steering mechanisms of social interaction and the functionality of so-called expert systems (e.g., money, power, companies, and networks)” [156]. Others defined institutional trust, similar to the concept of system trust [185]. They highlighted that trust in institutions specifically play a significant role in the management of relationships between actors from different sectors [185]. Finally, some scholars defined a specific type of trust that makes sustainability collaborations distinct: aspirational trust. Aspirational trust “reflects a vision of the potential that may transcend one’s organization, expressing one’s personal, “pro-social” ideology and motivation for action” [152].

Having prior relationships with a particular partner is identified as a factor that would help trust formation [152,186]. According to Vinke-de Kruijf, et al. [187], “when actors interact with each other over a longer period of time, they gain additional information about other actors that are rooted in relational experiences.” If their experiences are positive, then they will be more inclined to trust others [187]; however, partners may also be engaging for the first time in some sustainability-oriented alliances and partnerships. In these cases, their perceived reputation may yield an initial bonding trust [188]. Trust can even enhance relationships and improve learning when there are cognitive differences between partners since it acts as a glue and helps actors empathize with their differences [184].

4.4.3. Cognitive Dimension of Social Capital: Optimal Distance

The review demonstrates that the cognitive dimension of social capital is to do with shared cognition, values, logics, norms, and culture [120]. However, the review shows that in the context of sustainability-oriented alliances and partnerships, it may not be straightforward to share cognitive elements due to the complexity and subjectivity of sustainability. Although, based on the review, it is possible to identify two characteristics: the proximity (or distance) of cognition and institutional logics [189]. The degree of similarities in these dimensions is also commonly referred to as compatibility between partners or inter-organizational fit [166].

Scholars define cognitive distance as differences of partner organizations with regards to their organizational frames, which are “interpretations used to make sense of the world.” In other words, cognitive distance is to do with the “similarity in actors’ frames of reference, and mental modes facilitate effective and efficient communication and transfer of knowledge, although some extent of differentiation is needed for new ideas, creativity, and innovation to emerge” [183]. On the other hand, institutional distance is referred to as field-level differences between organizations with regards to their institutional logics which are “taken-for-granted assumptions and practices that shape the behavior of organizations in specific societal sectors” [189].

Some scholars highlight that these cognitive and institutional distances between partners pose both opportunities and challenges in terms of inter-organizational learning [189]. Some highlight that an optimum level of frame plurality can be achieved in collaborations [11]. Nevertheless, others argue that “different logics, values, interests, and knowledge systems need to converge” [131]. In other words, they argue that “shared mental models of interpretation may improve the firm’s capability to perceive focal issues in strategic nets and may empower the firm and the network to better respond to environmental challenges” [165].

Some scholars focus on measuring the impact of distance on the partnerships’ and firms’ sustainability performance [189]. Others take a longitudinal understanding of distance and argue that “the initial cognitive distance between the parties reduces through interactions and becomes a bidirectional exchange of knowledge” [140]. Indeed, some argue that it is this process of social learning that lead to changes or shifts in value frames [184]; which will be further discussed in Section 4.7.

For example, the following differences between for-profits and social enterprises due to logics are evidenced [134]: the value creation objectives (private value vs. public value), ownership structure (for-profit vs. non-profit), organizational governance (hierarchical vs. participative), accountability (to shareholders vs. to stakeholders). In the context of environmental research partnerships between scientists and for-profits, a study found that the differences in dominant logics can lead to different expectations regarding the outcome of the research [174]. As such, for-profit firms are associated with a market or commercial logic that drives them to expect “exploitable results through short-term applied research” from such partnerships, while scientists may target generation of publications out of the research partnership as an outcome [174].

Similar tensions are also commonly observed in firm-NGO interactions [12,128,190]. A recent study highlights that the impact of such cognitive and institutional differences on learning may depend on partners’ “value empathy” [190]. The “value empathy mechanism involves interventions which not only create an exchange of resources in the context of an individual project but also an ongoing capability to absorb knowledge across sectors” [190].

The review shows the learning from a partnership depends on the cognition and institutional backgrounds of collaborating parties. While the differences are more substantial in the context of cross-sector partnerships, even in the context of inter-firm alliances, partners have varying degrees of corporate environmentalism or corporate sustainability, which means varying value frames about sustainability [186,188]. It is plausible to expect that there would be an optimal distance between the partners whereby they are different enough to learn from each other and, at the same time, close enough to be able to understand each other’s language and work together, which would act as a catalyst to inter-organizational learning [11,189].

4.5. Inhibitors

The review shows that partner opportunism and power imbalance may inhibit inter-organizational sustainability learning.

Opportunism is defined as “self-interest seeking with guile, leading to deceit-oriented violation of implicit or explicit promises” [122]. Therefore, an opportunistic partner would manipulate the goals or the outcomes of the partnership towards its interests; which would diminish trust between partners and inhibit inter-organizational learning [122,166,171,180]. It is the various forms of governance that often protects an organization from a potentially opportunistic partner [134]. For instance, through equity-based governance, as in the case of joint ventures, firms can incentivize their partners financially against possible opportunism [127]. Partner opportunism can also be tackled with non-economic mechanisms, including trust and development of social capital [156]. Indeed, the expectation of partner opportunism increases in the existence of another inhibitor: power imbalance; meaning that partners who perceive themselves weak or inferior in the relationship dynamic would expect their partners to act opportunistically, and even engage in opportunism themselves, to shield themselves from potential opportunism and damage the relationship [186].

Scholars note different types of power, including but not limited to operational power, informational power, economic power, or social power [191]. Scholars also highlight that “if the power base of stakeholders is weak or if critical actors use their power to resist, learning is hampered” [191]. Some argue that “where knowledge exchange takes place it is likely to be because power is also being shared—to a greater or lesser extent—helping to provide a more conducive decision environment where the proponent/authority is also willing to receive” [169]. In line with this, others find that power imbalance between partners negatively moderates the relationship between incompatible logics and partnership survival [171]. These scholars highlight that the power imbalance would further amplify the cognitive differences between partners and may lead to situations whereby the more powerful partner imposes their dominant logics and frames to the weaker partner [171]. Relying on resource dependence theory, these scholars argue that one way to combat power imbalance is through the mutual dependence of partners, which are “bilateral dependencies regardless of whether the partners’ dependencies are balanced or imbalanced” [171]. These mutual dependencies are also reflected as complementarities between partners and “the extent of which each partner contributes unique strengths and resources to synergize new value” [166].

Finally, like other discussed characteristics, power imbalance and opportunism are not stable during a partnership relationship [174]. One study found that “balance in power and dependence develops over time” and “in the process of balancing the relationship between power and dependence, the relative absorptive capacity also increased, especially in the knowledge base and dominant logic dimension” [174].

4.6. Inter-Organizational Learning Process

Studies have discussed different types of learning processes, namely single-, double-, and triple-loop or higher-order learning, and exploitative and exploratory learning are discussed. Here, what is meant by learning is ‘relationship learning’ which is “a joint activity in which two parties strive to create more value together than they would create individually or with other partners” [192].

Single-loop or first-order learning is identified as “fixing errors from routines” [158] without questioning “boundary conditions, frames, assumptions”, in other words, “the usually ‘tacit’ assumptions implicit in the paradigm” [193]. It “contributes to insights and approaches for improving performance and efficiency (e.g., in skills and practices) to meet existing goals” [194]. This type of learning is characterized as the “optimization of existing routines, practices and systems” [184]. One study highlights that actors do not reflect on whether a particular system is sustainable or not during single-loop learning [184].

Double-loop or second-order learning is identified as “correcting errors by examining values and policies” [158], or “reflecting on the assumptions which underlie our actions” [177]. In other words, it is about “reflecting on existing frames and [that] actors have become capable of viewing and adapting these frames, paradigms, and values by a process that has been coined frame reflection” [193].

Triple-loop learning is identified as “designing governance norms and protocols… to improve the capacity of an organization to engage in single- or double-loop learning” [158]. This kind of learning “encourages a more open-ended and deep-seated discussion about what the primary challenges are and ways to reshape the values, norms, and social structures to address it” [194].

Exploitative learning focuses on “applying successful practices into large-scale manufacturing, dissemination of existing technology, and standardization of the current routine to enhance efficiency” [148]. In other words, exploitative learning is to do with “the acquisition of new behavioral capacities framed within existing insights” [191]. On the other hand, exploratory learning is associated with “new searches and experimentation” and “risk-taking and variance-increasing activities in learning, experimentation, flexibility, discovering, and distant search” [195].

It is important to note that, to contribute towards SDGs, all types of learning are needed [158]. While single-loop or first-order learning helps firms with building new or enhancing existing sustainability-oriented capabilities, double- and triple-loop learning helps firms reflect on existing values and shift their frames.

4.7. Outcomes

The review demonstrates different types of outcomes: firm-level, partnership-level and system-level [120].

4.7.1. Firm-Level Outcomes

Firm-level outcomes can be described in two categories: firm-level learning and general outcomes. As introduced earlier in the introduction, the value generated from a partnership is not limited to the interaction value of learning. Therefore, herein, the outcomes of learning are recognized as capability development and shifts in value frames, while other outcomes of partnerships are identified as ‘general outcomes’ in the model presented in Figure 2.

According to Vinke-de Kruijf, Bressers and Augustijn [187] “there are mutual relations between an interaction process and actor characteristics: the characteristics shape the process and are also shaped by the process”; meaning, “learning involves changes in actor characteristics.” As highlighted in the Partner Characteristics section, existing resources, capabilities, and absorptive and disseminative capacity are included in these characteristics. In this paper, the focus as an outcome is on capabilities because a large number of studies focus on how firms can develop new capabilities through learning that takes place in sustainability-oriented partnerships [49,139,141,165,196,197,198]. Development of new capabilities and shifts towards different frames are often associated with an improved absorptive capacity [192,197,199]. Therefore, partner characteristics improve as a result of the learning process [187].

In the context of sustainable supply chain management, collaborative relationships with suppliers positively impact the development of organizational capabilities, which then positively impact a firm’s sustainability performance [200]. Similarly, in alliances between governments and firms that aim to foster radical eco-innovations, close interactions in alliances have yielded a generation of new capabilities that can help firms address sustainability problems [127]. These can be technical or operational capabilities that can help firms reduce their environmental or societal impact [141]; equally, these capabilities may also be dynamic capabilities such as alliance/partnership management or ‘external integration’ capabilities [4]. This capability is an organization’s ability to organize and manage relationships with external partners [201]. Indeed, through partnerships, firms would not only learn ‘what to do’ and ‘how to do it’ to become sustainable; but also, would learn how to engage with various partners better.

While these capability-related changes are likely to result from single-loop learning [197], more profound changes can also be observed. Thanks to close interactions with partners that lead to cognitive learning, shifts in value frames can also be observed [127]. Collaborative partnerships may require “reframing, which involves perspective-taking and the possibility of enlarging or revising one’s frame to take account of how their counterparts view the situation” [202]. Le Ber and Branzei [12] find evidence of “partners not only updat[ing] their frame concerning each other, but they also do so in reference to, and in conversation with, each other” whereby they go beyond the reframing process to frame fusion in which their frames are continually changing.

At the organizational-level, studies have differentiated between narrow or simple and broad or complex frames [189]. While some studies propose that the engagement of various types of partnerships depend on these simple or complex frames [189,203], others focus on how, within a partnership, these frames shift over time. An optimal frame plurality is achieved whereby various frames evolve within the lifecycle of a partnership; some frames lose traction, yet others are sustained [11]. These kinds of frameshifting, frame fusing, or frame-breaking outcomes are likely to be associated with double- or triple-loop learning [191].

There are also general outcomes of partnerships associated with the creation of associational value, transferred resource value and synergistic value [9], as well as interaction value, which is considered as learning in this article. Thanks to the development of new capabilities and shifts in value frames, as well as the creation of associational, transferred resource and synergistic value, partnerships would help firms improve their environmental performance [136,146,147,189,195,198], social performance [133,167], sustainability performance [200,204], innovation performance [205] and may even improve economic performance [121].

It is important to note that not all studies find evidence that inter-organizational learning, or sustainability collaborations, in general, impact performance positively [173,206]. This study, as shown in Figure 2, expects to see an improvement in the sustainability performance of a focal firm, both due to gained capabilities and shifts in frames, as well as other types of value created within a partnership. However, it may also be methodologically problematic to measure the exact contribution of a partnership on a firm’s sustainability performance. Often, firms engage in a portfolio of partnerships [203] and it is this portfolio that allows firms to create synergistic value at the portfolio-level [119,120]. Furthermore, realizing the benefits of partnerships either through the value created thanks to learning, or other value dimensions, may require time and, therefore, observation in longitudinal settings.

4.7.2. Partnership-Level Outcomes: Alliance/Partnership Success

Amongst other performance outcomes in the review, partnership-level outcomes have received the least attention. The definition of partnership success has been somewhat vague: while some took partnership success as the sustenance of the partnership [171], others provided an organization-level definition focusing on what different organizations take away from the partnership [207]. Partnership success can generally be considered as an accomplishment of goals that are set out by the parties involved within the timeframe agreed upon [176]. One study explicitly focuses on the role of institutional logics and resource dependencies on partnership success [171]. Others refer to ‘partnership’ or ‘alliance‘ success; however, they do not explicitly measure the impact of various factors on partnership success [186]. Sanzo, Álvarez and Rey [207] proposed that “the existence of a process of learning within the non-profit will probably enhance the firm’s perception (and also the non-profit’s own appreciation) about the non-profit’s capability to contribute to partnership success”.

Partnership-level outcomes are expected to be in a dynamic relationship with (shown as a feedback relationship in Figure 2) to the partnership characteristics, catalysts, and inhibitors. Generally, collaboration experience can help to build capabilities to manage such partnerships more effectively in the future and assess what modes of governance are more appropriate under which circumstances, which was also discussed in the previous section [51,127]. Also, within a continuing partnership, partners can reflect on the partnership outcomes collectively, assess whether the chosen mode of governance allowed creating a solution space for a particular problem, and evaluate whether the expected learning outcomes are achieved by all parties involved. Furthermore, within a continuing partnership, partners can also reflect on the inhibitors and catalysts that impact the learning process.

It should be noted that poor learning may also impact partnership performance negatively and may later act as an inhibitor, or a positive learning experience may enhance trust (feedback relationships). These feedback relationships explain why a focal firm would partner with an organization that they had an amicable relationship with before and why it may refrain from partnering an organization that they have lost trust in due to a previous engagement that resulted in poor performance.

4.7.3. System-Level Outcomes: Contribution to Sustainable Development Goals (SDGs)

System-level outcomes are macro-level societal or environmental benefits [208]. Several studies identified the system-level outcomes of inter-organizational sustainability learning by discussing how these alliances and partnerships help to address SDGs [5,19,131,176,193,203,209], or previously discussed millennium development goals [210]. For instance, Dzhengiz [203] provided evidence from electric utilities focusing on alliances and partnerships that address SDG 7 and 13. Kolk and Lenfant [211] focus on the role of partnerships for SDG 16, promotion of just, peaceful and inclusive societies, while Le Ber and Branzei [12] focus on the role of partnerships for SDG 3, access to healthcare. Even though the majority of the articles have not explicitly referred to how studied alliances and partnerships have contributed to SDGs, all articles in the review focused on alliances and partnerships that addressed various SDGs implicitly (see Table in the Appendix A).

Among others, only SDG 5 was not addressed by the studies in the review. All articles in the review have contributed to SDG 17, as all articles discussed the role of partnerships and alliances to tackle sustainability challenges. SDG 17 was followed by SDG 8, as expected, highlighted in 40% of the articles with the emphasis on sustainability-oriented innovation that can be generated through partnerships. Finally, around 5% of articles focused on partnerships that aimed to tackle SDG 13, climate action and SDG 12—sustainable cities (3%).

5. Future Research

As a review article, this study identifies some limitations of the extant research and offers paths for future studies, following the categories in Figure 2.

5.1. Partner Characteristics

Articles in the review have frequently focused on partners’ characteristics such as partners’ capacity to learn or absorb knowledge. However, they have not taken ‘learning’ as a bi-directional concept, whereby both partners need to absorb and disseminate knowledge. This shows that studies that focus on inter-organizational sustainability learning can benefit from a more comprehensive bi-directional understanding of learning, hence, an understanding of both parties’ absorptive and disseminative capacities. To do so, future studies should integrate the recently growing literature on disseminative capacity [92,93,149,150].

In line with this comment, it should be also noted that this study also provided a limited understanding of the learning phenomenon since it mostly focused on actors from the private sector, their learning needs and outcomes, and their potential contributions to systemic and wicked sustainability challenges. Future studies can empirically study the partner characteristics of other societal actors and how these different characteristics would impact the bi-directional process of learning.

Moreover, studies in this realm can benefit from a paradoxical understanding of inter-organizational sustainability learning by focusing on how embeddedness into existing capabilities may create vicious or virtual cycles, drawing on the literature of organizational paradoxes [212,213,214,215,216,217]. The review shows that, to some degree, a firm’s organizational capabilities and absorptive capacity will determine how much it can learn from a sustainability-oriented alliance or partnership. Focusing on longitudinal studies, future studies can further interrogate how firms overcome this embeddedness paradox in the context of sustainability and provide evidence for firms that were able to turn their vicious cycles into virtuous ones through mindful interventions [213,218].

5.2. Partnership Characteristics

Articles in the review have frequently referred to “governance” as a partnership/alliance characteristic that would impact learning. However, comparative explorations seeking to understand which modes of governance allow more room for learning, and under what conditions, have been somewhat limited [148,158,164]. Future studies can focus on the relationship between different modes of governance (such as equity vs. non-equity alliances, network governance, polycentric governance) and learning types or levels (exploratory, exploitative, single-/double-/triple-loop).

Besides, it is plausible to expect that different SDGs would require different forms of learning, and hence, different forms of governance. However, a holistic approach to how firms address different SDGs through different types of governance that allow a different type of learning is yet to develop. Such an approach can be developed, drawing on the literature on alliance or partnership portfolios [219].

Furthermore, studies also highlight that there would be differences in the contractual governance of partnerships, depending on the partner type (inter-firm vs. cross-sector) [220]. However, there has generally been limited attention to this area in literature, which requires further attention [221,222]. Notably, future research needs to investigate the differences in inter-organizational learning in inter-firm and cross-sector partnerships for various governance forms and examine the how governance affects the learning mode and type and the impact of learning on sustainability performance.

5.3. Context

The review showed that both institutional theory and stakeholder theory highlighted the role of the external environment and legitimacy pressures on firms as a motivator of engaging in inter-organizational sustainability learning. However, articles in the review have not focused on the role of the ‘internal context’ of firms also setting pressures for such engagement. Meaning, employees’, and managers’ roles in initiating inter-organizational sustainability learning processes need to be understood further.

Furthermore, it would be interesting to measure the impact of ‘legitimacy crisis’ on inter-organizational learning using a quantitative method. While existing studies highlight the legitimacy crisis as a factor that explains partnership formation, it is also evidenced, in some studies, that partnerships formed to address a legitimacy crisis may not always yield learning opportunities [223].

5.4. Catalyst

Articles in the review frequently focused on at least one of the dimensions of social capital, social capital’s role in improving inter-organizational sustainability learning and partnership outcomes. However, it is essential to highlight that most studies have analyzed different dimensions of social capital in isolation, with a few exceptions [126]. Some solely focused on trust as a relational dimension [122,156,188]; others on cognitive and institutional differences [12,189]; and yet others on structural dimensions and coordination patterns [179]. Future research needs to address the impact of social capital on inter-organizational sustainability learning, especially using longitudinal studies since social capital changes in different phases of a partnership.

Furthermore, the extant literature often studied how social capital may generally catalyze the relationship between the partners and improve partnership outcomes; however, it is most likely that the impact of social capital on partnership performance is mediated by inter-organizational learning [122]. Future studies can take into consideration more sophisticated models to test these webs of relations.

Finally, future studies should also take into consideration to what extent social capital improves inter-organizational learning when there are strong influences of various inhibitors such as power imbalance and partner opportunism. In line with this, recent literature lacks a longitudinal analysis of both inhibitors and catalysts, looking at whether and how these factors change in different phases of an alliance relationship.

5.5. Inhibitors

The impact of partner opportunism on inter-organizational learning and partnership performance is widely discussed in the context of inter-firm alliances [224,225,226]. Surprisingly, in the context of inter-organizational learning within sustainability-oriented alliances and partnerships, partner opportunism has not been studied as much. This is likely to do with the researcher bias in the field and the expectation that sustainability-oriented alliances and partnerships are more altruistic [11]. However, this does not mean a firm’s relationship with an NGO would involve less opportunism than a firm’s relationship with another firm. The review shows that partner opportunism and learning paradoxes must be unpacked further in the context of sustainability-oriented alliances and partnerships. Future research can focus on building a comparative analysis of how firms’ opportunistic behaviors differ depending on the partner type and partner status.

Another inhibitor, power imbalance or asymmetries, is rarely studied in this literature. Furthermore, when it is studied, it is typically through the lens of resource dependence theory [171]; and rarely combined with theories of learning and knowledge development [174]. Power imbalance or asymmetries also impact what can be learned from whom within a partnership setting. Future studies in this field can benefit from studying inter-organizational learning, capability development and frameshifts through the lens of political processes of power, and draw on recent literature that integrates learning and power [227,228,229].

5.6. Inter-Organizational Learning Process

There has been growing attention given to triple-loop learning in the context of sustainability in recent years [106,107,158,194]. However, as a construct, triple-loop learning has also received some criticisms in the broader literature of organizational learning [230]. Studies within the context of sustainability have not elaborated these debates regarding ‘triple-loop learning’, and the concept appears to be misused or gets reified as did other concepts such as absorptive capacity [231], and recent clarifications about the concept can contribute to the work of future scholars [230].

Second, there have been studies that focused on partnership formation motivations that proposed how “resourced-based motivation is more likely to be associated with firms’ participation in exploration alliances” [148]. In contrast “institutional motivation is more likely to be associated with firms’ participation in exploitation alliances” [148]. However, these studies have not focused on how some prior exploitation alliances may, in the future, yield exploration alliances with the same partners or vice versa for exploration.

Longitudinal assessments of partnership portfolios can further enhance our understanding by showing how firms engage with various partners, as well as explore and exploit. Furthermore, it is likely that, as in the context of commercial inter-firm alliances [94,95,232], firms would use their alliance and partnership portfolios: (a) to balance the tensions between exploration and exploitation of sustainable products, technologies and processes, (b) to balance the tensions between the triple bottom line of sustainability (environmental, social and economic sustainability), (c) to balance the tensions between short-term and long-term concerns. However, the literature on partnership portfolios has, thus far, been limited in the sustainability context [203,219]. While some studies highlighted how, at the level of the dyad, a single alliance provides a space for learning, portfolios are likely to provide a broader space for resolving tensions mentioned above.

5.7. Outcomes

In the extant literature, there was some focus on frameshifts [11,12] and capability development [49,233], but a limited quantitative assessment of how such learning enhances firm performance (environmental, social, economic). A reason why this has been a barrier is also to do with the dyadic focus of this literature [120]. Future research can, instead, focus on portfolios to measure: (a) what kind of alliances/partnerships improve what kind of performance (environmental, social, economic), and (b) how the impact of partnership type on performance is mediated by various types of learning. Such approaches have been limited [135,234], and often used different measures or different types of collaborations. Therefore, the results have been somewhat conflicting [173], which is why future research should clarify the impact of inter-organizational sustainability learning on various performance outcomes.

Still, in comparison to firm-level outcomes, partnership-level outcomes have been studied even less. In the commercial inter-firm alliance context, partnership effectiveness or success has been frequently discussed [235,236,237] However, in the context of sustainability, only a few studies have focused on partnership terminations [186] and partnership success [171,207]. Future studies can further explore how inter-organizational learning and firm-level outcomes impact perceived partnership performance and study the impact of catalysts and inhibitors on this relationship.

Even though many studies used system-level arguments to explain partnership formation motivations, the system-level outcomes of partnerships themselves require further research. Most studies discuss why firms enter partnerships to tackle ‘systemic’ sustainability issues. However, they do not explain whether and how these partnerships effectively solve some societal and environmental sustainability challenges, or which SDGs these partnerships contribute towards and to what extent the impact of these partnerships can be measured. Future research should further focus on the effectiveness of these partnerships and their contribution to tackling sustainability challenges. A way this can be done is by identifying metrics that firms use to measure and communicate their sustainability progress according to different SDG areas, and to measure the improvements or the impact of a partnership using these metrics [238,239,240]. Another way is to link inter-organizational learning to sustainability transitions literature [241,242] and discuss how the single-, double- and triple-loop learning that takes place in alliances and partnerships may yield changes at the system-level [243,244].

Finally, it should be noted that the model presented in Figure 2 demonstrates potential relationships based on the extant literature. This model can be considered as a conceptual framework that helps us organize the existing literature through the lens of inter-organizational learning. However, this model is not tested empirically in this review article. Therefore, while some relationships and mechanisms are discussed herein, future research should further test this model, and especially investigate how the firm-level learning, firm-level general, partnership-level and system-level outcomes relate to and impact each other.

6. Conclusions: Contributions to Theory and Practice

This study contributed to theory in several ways. First, the study synthesized different theoretical perspectives and concepts and demonstrated how these different concepts build a complex picture of inter-organizational sustainability learning that is represented in Figure 2. Second, the study clarified various outcomes of sustainability-oriented alliances and partnerships, and one of these outcomes is at the system-level, which can be understood as a contribution towards SDGs. The paper outlined how such a contribution does not only result directly from the partnership itself but also firm-level outcomes through the development of capabilities and shifts or expansions of firms’ value frames. Besides, the paper outlined that these system-level outcomes make sustainability-oriented alliances and partnerships distinct, as commercial alliances and partnerships do not focus on societal or public value creation. Finally, drawing on the review findings, the paper identified gaps and proposed areas for future research. Doing so, this study contributed to the literature on sustainability-oriented alliances and partnerships of firms.

This study also contributed to the practice of sustainability practitioners. Previous research has already evidenced that sustainability practitioners are increasingly involved in managing alliances and partnerships, and therefore, they need skills to work in an interdisciplinary and collaborative manner. This study further adds that to manage sustainability-oriented alliances and partnerships effectively, practitioners need to focus on how best they can learn from their partners, develop capabilities that enhance their sustainability performance and broaden their mental models—their value frames to different interpretations of sustainability. Furthermore, the model presented in the study is intended to guide practitioners in identifying what catalyzes and inhibits their learning relationship with partners and under which environmental conditions. Furthermore, the model helps in interrogating what characteristics a firm and its partners need to enhance the learning outcomes (such as capabilities, resources, absorptive capacity, disseminative capacity). Using such a model can improve the work of practitioners and the impact they can generate from partnerships at the firm-, partnership-, and system-level.