Abstract
How do platforms integrate governance mechanisms that promote inherently distinct rules and incentives to manage peer relationships? Digital platforms combine market and community-based mechanisms to govern peer-to-peer interactions for value creation. However, these governance mechanisms play unique roles and interact in distinctive ways, thus shaping how platforms can leverage them for the governance of peer relationships. Through an analysis of sharing platforms, we identify under what conditions particular couplings of market and community mechanisms facilitate a stable governance configuration. We uncover how platforms leverage complementarities and avoid tensions among a set of core and elaborating governance mechanisms. The findings show that market and community mechanisms and their interactions constrain platform governance in different ways. When platforms have strong commercial identities and offerings, implement strict assurance instruments, or develop strong social institutions, they confine core mechanisms to a single governance structure and prevent innovative configurations. However, under specific conditions, platforms explore complementarities between market and community mechanisms which lead to either mixed or highly mixed governance configurations. The study uncovers platforms’ possibilities and constraints in developing stable governance configurations which hybridize market and community mechanisms.
Digital platforms have emerged as a new form of organizing, providing a digital interface and utilizing digital technologies to enable interactions between distributed participants at scale (Chen et al., 2022: 151; Kornberger et al., 2017: 79). Many peer-to-peer exchange platforms, frequently referred to as “sharing economy” (Gerwe and Silva, 2020), “peer production” (Benkler, 2017), “gig economy” (Frenken et al., 2020), or “distributed innovation” (Boudreau and Lakhani, 2009), rely on the participation of private individuals who contribute assets. These individuals assume complementary roles, that is, as peers (Einav et al., 2016), such as hosts and guests (Airbnb), drivers and riders (BlaBlaCar), sellers and buyers (Vinted), service providers and demanders (Fiverr), or app developers and users (App Store). Through their interactions, participants exchange resources and attract new people, thereby enhancing the platform’s network value. The governance of peer relationships is crucial for the development and competitiveness of platforms, as their capacity to create value and evolve depends on the continuous contributions of individual participants.
Platform governance involves creating “a set of overarching rules, constraints, and inducements” to coordinate and incentivize the relationships between a platform owner and participants as well as among platform participants (Boudreau and Hagiu, 2009; Chen et al., 2022: 153). Unlike conventional organizational governance, platforms’ intermediary role and ownership of a digital infrastructure have introduced novel, algorithmic governance mechanisms such as searching and matching, rating, verification, and participatory tools, tailored to coordinate participants within hierarchical, market, or community structures (Chen et al., 2022; Kornberger, 2016). In practice, many platforms adopt a hybrid form of governance, incorporating governance mechanisms associated with more than one structure, albeit to varying degrees. While most platform governance research has focused on the relationship between a platform owner and participants, examining how hierarchies and markets can be integrated to achieve a balance between platform control and participant autonomy (Chen et al., 2022; Kretschmer et al., 2020), less attention has been paid to the governance of peer relationships which can be organized through markets and communities.
Recent research acknowledges the potential for peers to engage both in competitive and collaborative interactions within markets and communities, giving rise to hybrid configurations for governance (Boudreau and Lakhani, 2009; Gulati et al., 2012). Effectively designed and managed hybrid configurations can facilitate the aggregation of varied resources, the exploration of participants’ innovation capabilities, and the generation of diverse forms of value (Benkler, 2017; Boudreau and Lakhani, 2009). Most research on platform governance concerning peer relationships, however, concentrates on the design of governance configurations which focus exclusively on either market or community mechanisms (Einav et al., 2016; Kirchner and Schüßler, 2019; Reischauer and Mair, 2018). There is also case study research that has looked into the functioning of single and mainly novel algorithmic governance mechanisms and their impacts (Kornberger et al., 2017; Luca, 2017). While these studies offer valuable insights, it is less clear how multiple market and community mechanisms (both conventional and algorithmic), which rely on distinct rules and incentives, can be combined and whether they interact in complementary or conflicting ways in platform governance (Altman et al., 2022; Benkler, 2017). Although market and community mechanisms are relatively well understood when they function independently or within a single structure, we know little about how they work synergistically. Despite a need for platforms to create effective ways to manage peer relationships, how they develop stable governance configurations which leverage complementarities and avoid tensions between market and community mechanisms remains underexplored.
To address this gap, we adopt a configurational approach which views platform governance as a system of core and elaborating elements serving different roles (Fiss, 2007; Siggelkow, 2002). We examine how peer-to-peer exchange platforms leverage market and community mechanisms and their interactions for the governance of peer relationships. We investigate under what conditions particular couplings of market and community mechanisms emerge as necessary or sufficient for a stable governance configuration. Our findings reveal that market and community mechanisms, and their interactions, constrain platform governance in different ways. When platforms have strong commercial identities and offerings, implement strict assurance instruments, or develop strong social institutions, they confine core mechanisms to a single governance structure and prevent innovative configurations. However, under specific conditions, platforms explore complementarities between market and community mechanisms, which lead to either mixed or highly mixed governance configurations.
Our research makes contributions to the fields of platform governance, new forms of organizing, and the sharing economy. First, we improve understanding of the governance of peer relationships and uncover which governance configurations achieve relative stability (Altman et al., 2022; Benkler, 2017). We show that market and community mechanisms mainly interact synergistically rather than at random. Second, our study reveals the design considerations of platforms when developing stable and innovative ways to govern peer relationships as a new form of organizing. This is achieved by clarifying the “causal architecture” where internal interactions between mechanisms can influence the emergence of stable governance configurations (Gulati et al., 2012; Puranam et al., 2014). Our findings capture elements of market and community structures that are either core or elaborating in the governance of peer relationships and how the capabilities of leveraging their interactions to produce stable governance configurations are dependent on specific platform contexts. Finally, our findings shed light on the controversy surrounding the sharing economy about whether platforms are practicing social innovation or are recreating market capitalism instead (Eckhardt and Bardhi, 2015). We show that market mechanisms tend to prevail in most configurations; however, the presence of community mechanisms is important, too.
Literature review
Platform governance of peer relationships
Governance refers to the formal and informal means (e.g. incentives, contracts, and norms) which an organization adopts to provide direction, control activities, and manage relationships between various stakeholders (Foss and Klein, 2013; Mair et al., 2015; Sitkin et al., 2010). Platform governance involves creating “a set of overarching rules, constraints, and inducements” to incentivize and control, thereby coordinating relationships between the platform owner and participants as well as between platform participants (Boudreau and Hagiu, 2009; Chen et al., 2022: 153). Recent platform research has primarily focused on the relationships between the platform owner and participants through the integration and interaction of hierarchical and market mechanisms (Chen et al., 2020; Gulati et al., 2012; O’Mahony and Karp, 2020). Such hybrid governance forms seek to strike a balance between centralized hierarchical governance, characterized by the tight control of platform owners, and decentralized market governance, which operates on independent decisions and economic actions of distributed participants (Chen et al., 2022; Kretschmer et al., 2020). However, there is little systematic analysis of the interactions and design considerations in hybrid governance involving market and community mechanisms that aims to optimize the governance of relationships between participants (Altman et al., 2022; Boudreau and Lakhani, 2009).
While both markets and communities facilitate decentralized governance of autonomous individuals, they differ in the focal relationship they encourage between peers, that is, what “linkages” are created, what interactions are promoted among participants, and how this is done (Gulati et al., 2012: 581). Markets facilitate economic transactions between participants driven by economic return, utility, and efficiency, with their interactions being typically immediate and objective-oriented (Kirchner and Schüßler, 2019). Market governance relies on pricing and contractual arrangements, formal and legal frameworks, enforceable sanctions, and assurance instruments to control participant behavior and address market frictions (Williamson, 1979, 1985). In contrast, communities foster social bonds between participants who seek to fulfill an intrinsic desire for reciprocity, trust, self-fulfillment, and a shared identity (Benkler, 2017). Interactions are ongoing and relationship-oriented (Benkler, 2017). Community governance comprises mechanisms that facilitate interpersonal interactions, informal norms, community spaces, and peer monitoring (Reischauer and Mair, 2018; Thornton et al., 2012). Market and community mechanisms differ in the values, interaction modes, and normative and regulative structures that they employ to incentivize and control the actions of participants (Zhang et al., 2020).
Building on the widely recognized potential of organizing autonomous individuals in markets and communities (Altman et al., 2022; Benkler, 2017; Boudreau and Lakhani, 2009; Gulati et al., 2012), we propose that platforms’ hybrid governance of peer relationships can simultaneously draw on the values, interaction modes, and normative and regulative structures of markets and communities, but to varying degrees, in order to incentivize and coordinate individual participants’ actions (Chen et al., 2022; Schüßler et al., 2021). Moreover, with the aid of digital technologies, platforms can deliver the values, interaction modes, and normative and regulative structures of markets and communities through conventional practices like monetary and social rewards, Terms and Conditions, Community Guidelines (Chen et al., 2022; Kolbjørnsrud, 2016), as well as algorithmic mechanisms such as searching and matching functions, evaluative infrastructures, and participatory tools (Kirchner and Schüßler, 2019; Kornberger et al., 2017).
Governing peer relationships with markets and communities
Early research on platform governance which incorporates market and community mechanisms primarily focused on the participants of open-source software development (OSSD), characterized by its technical specificity like open access to source code (O’Mahony and Ferraro, 2007; O’Mahony and Lakhani, 2011; Shah, 2006). More recent research has explored a wider range of settings in which market or community mechanisms are used to organize peer interactions (Benkler, 2017; Kirchner and Schüßler, 2019). This shift is aligned with the development of more universally applicable platform-based models that utilize tangible and intangible resources, exemplified by the sharing economy which facilitates the temporary access to underutilized physical and human assets between individuals (Gerwe and Silva, 2020). Some studies focus on community incentives and control mechanisms. Reischauer and Mair (2018), for example, captured governance practices in online sharing communities that created a sense of community, structured participant engagement, and facilitated ongoing interactions between participants. Chen et al. (2017) identified three community motivations—reciprocity, peer recognition, and self-image—for participants’ contributions to platforms. Other studies place emphasis on market mechanisms, examining pricing (Einav et al., 2016; Weyl, 2010), rating and sanctioning (Kirchner and Schüßler, 2019; Kornberger et al., 2017), certifying quality and insuring (Einav et al., 2016), and legally binding market rules and conditions (Kirchner and Schüßler, 2019). However, their focus has mainly been on a single governance structure and its associated mechanisms, particularly algorithmic ones, based on in-depth, small-scale case studies, and theoretical propositions.
There are a few studies which point to the potential to organize peer relationships in markets and communities simultaneously, although using different terminology. Boudreau and Lakhani (2009) contrasted the practices and conditions of organizing distributed innovations through competitive markets and collaborative communities. They suggested the potential of adopting a “mixed” or “nested” approach. While cautioning for internal tensions that could be present in incentives and platform control, they did not detail how such tensions would systematically affect the design of hybrid configurations. Gulati et al. (2012: 571) theorized the varying degrees and ways of engaging market and community-related mechanisms to organize autonomous individuals as critical design considerations for “meta-organizations” like platforms. Altman et al. (2022: 70) reviewed the growing literature of organizations engaging external communities and highlighted the lack of focused investigation into “hybrid governance” on platforms, characterized by their integration of multiple governance structures, and simultaneous leveraging of collaboration and competition as well as exploration and exploitation.
Despite the increasing significance in research and practice, systematic empirical research about the use of hybrid configurations to govern peer relationships is limited. We know little about how market and community mechanisms are combined and interact in a synergistic way (Altman et al., 2022). Therefore, we propose a configurational approach to examine how market and community-based mechanisms interact and shape the creation of stable configurations to govern peer relationships.
An organizational design perspective toward platform governance
From an organizational design perspective, organizing involves a system of synergistically operating rather than loosely coupled elements whose combined fit is associated with organizational performance (Fiss, 2007). Platform governance forms a system which relies on interdependent governance elements tailored to achieve platform goals (Misangyi and Acharya, 2014; Siggelkow, 2011). To achieve coherence and stability, some mechanisms are fundamental to the system’s functioning and relatively irreplaceable, while others serve as supporting mechanisms instead (Siggelkow, 2011). The former constitutes a core element which interacts with many other elements and has a large impact on future elements, while the latter is an elaborating element which reinforces core elements and suggests complementarity, that is, its presence increases the value of core elements (Siggelkow, 2002). When a core element’s interdependency with associated elements is very strong, it becomes necessary and it is unlikely to be substituted by an alternative element. Such an interaction will generate structural inertia, constraining governance choices (Hannan and Freeman, 1984). However, each governance element, no matter how core it is, also works interdependently with other elaborating elements in a governance configuration to achieve the desired outcome (Fiss, 2007; Kirchner and Schüßler, 2019). Together, core and elaborating elements constitute the system of platform governance to manage peer relationships.
Combinations of governance elements are not random but closely associated with the interactions between elements that serve distinct roles, leading to specific considerations of designers aimed at achieving alignment and stability. Complementarities between elements are considered to relate to clusters of elements—configurations—which are observed more frequently than others (Puranam et al., 2014). One proposed source of complementarity, for example, is that community interactions can provide fine-grained information, unavailable from transactions, which reduces market uncertainty (Benkler, 2006; Ladegaard, 2021; Powell, 1990). However, research points to tensions, too, that can have an impact on the development of configurations because extrinsic, monetary rewards might crowd out intrinsic, social incentives in peer interaction (Benkler, 2017; Boudreau and Lakhani, 2009). Literature not focusing on platforms notes a “crowding-out” effect of contractual and sanctioning structures as well as assurance instruments on community incentives and trust (Cheshire, 2011; Macaulay, 1963; Nissenbaum, 2004). Then again, new forms of organizing like digital platforms can introduce innovative ways of configuring conventional and novel governance mechanisms to create unique sets of value (Puranam et al., 2014).
Despite research acknowledging the importance of interactions between governance mechanisms, empirical research systematically examining such interactions in a platform context remains scarce. It is unclear whether and how these interactions shape how platforms configure stable governance configurations which leverage complementarities and avoid tensions. By adopting Qualitative Comparative Analysis, we take an exploratory approach to uncover configurations for the governance of peer relationships and their design considerations. We examine under what conditions particular couplings of market and community mechanisms emerge as necessary or sufficient for the development of stable governance configurations, thereby influencing the governance choices of platforms.
Methodology
Data and sample
Sharing economy platforms to access or exchange underutilized physical and human assets enable individual participants to interact directly or indirectly (Gerwe and Silva, 2020). Since the advent of Uber and Airbnb, they have expanded across a broad spectrum of sectors and entities, ranging from traditional businesses to non-profits and social enterprises. Given their widespread application and the principles they embody, sharing platforms can be viewed as a highly representative subset of peer-to-peer exchange platforms and form an ideal case to systematically examine how such platforms configure market and community-based governance mechanisms to manage peer relationships.
To ensure inclusiveness and representativeness of our dataset, we sampled cases with the following procedure (Thomann and Maggetti, 2017). We first conducted a comprehensive screening of platforms in five primary sharing economy sectors identified by industry experts (Owyang, 2014, 5 May) and generated an initial case set. We considered platform traits, including the time of establishment, country of origin, geographical coverage, and business model features to ensure the diversity and representativeness of selected cases. Next, we examined each platform against three key attributes of sharing platforms generally agreed in the academic literature (Gerwe and Silva, 2020; Mair and Reischauer, 2017): temporary access, underutilized capacity, and peer-to-peer interaction. We found several cases that did not fully meet all attributes and identified inconsistencies between platform self-identifications and the three attributes. We assessed these boundary cases individually to decide the extent to which they diverged from the attributes (particularly peer-to-peer interaction) to make a judgment about their inclusion. For example, although Car2Go provided a fleet of company-owned vehicles instead of private cars, it strongly recognized the free-floating car-sharing concept underlying resource access and redistribution in a community of individual drivers. 1 We thus included it in our sample.
The final sample comprises 56 platforms: ride-sharing (10), space-sharing (12), food-sharing (11), goods-sharing (13), and labor/skills/knowledge-sharing (10). We collected and content-analyzed all web pages and legal and community files of these platforms, including Terms and Conditions and relevant Blogs. Our data sources are complementary. Web pages present governance mechanisms but contain marketing content and avoid negative information. Legal files are more rigorous definitions and statements of how exchanges are established, operated, and controlled but are limited in describing interactive platform features and nonlegal arrangements. Our dataset includes 1840 screenshots of web pages and 143 legal files from August 2017 to January 2018. Half the cases were established between 2010 and 2015, and 45% started before 2010. The youngest platforms were founded in 2015 and had been in operation for around 3 years. While platforms’ governance mechanisms can change at an early stage, they generally stabilize after 3 or more years in existence. Therefore, our sample consists of representative cases with relatively stable and enduring governance configurations.
Methods, research design, and outcome of interest
The unit of analysis in this study is a platform’s governance configuration. We aimed to identify and analyze configurations of market and community-based mechanisms, examining the roles of and interactions between them. We adopted the case-oriented, set-theoretic method Qualitative Comparative Analysis (QCA) which analyzes the causal relevance between elements or combinations of elements and an outcome of interest. Fuzzy-set Qualitative Comparative Analysis (fs/QCA) understands cases as configurations of theoretically relevant elements of a fuzzy nature where elements are present in fine-grained degrees rather than a clear-cut presence or absence as in crisp-set Qualitative Comparative Analysis (cs/QCA; Ragin, 2008b). In our study, we applied both fs/QCA and cs/QCA to address our research aims at the different stages of the analysis, as specified below. QCA clarifies causal relevance by identifying the subset and superset relations between the elements and the outcome of interest (Fiss, 2007; Ragin, 2008b). Necessity analysis assesses to what extent an element is a superset to (i.e. is necessary for) another element. Sufficiency analysis captures to what extent a combination of multiple elements is a subset to (i.e. is sufficient for) an outcome. QCA evaluates the extent to which the element or the combination of elements is consistently necessary/sufficient for the outcome, that is, “consistency” for the necessary/sufficient relationship; while “coverage” gauges the relationship’s empirical importance (Ragin, 2008b).
Most QCA applications in management research focus on finding combinations of causally relevant elements which lead to a specific outcome (Misangyi and Acharya, 2014). Our research design is more exploratory. In view of our research aims, we focus on the relative interdependency of governance elements and their causal relevance with the outcome of a stable governance configuration (Vith et al., 2019). We clarify the “causal architecture” and its relevance for the emergence of stable governance configurations to reveal the underlying mechanisms. We use “governance element” or “element” for a better alignment with QCA terminology and to denote a higher level construct that can encompass multiple specific governance mechanisms performing a consistent function.
In the first phase of our analysis, to assess the interdependency between governance elements and uncover core elements which would prevent highly mixed configurations, we relied on fs/QCA’s necessity analysis as it detects strong interdependencies and non-substitutable elements. In the second phase, to uncover elaborating elements, especially those drawn from the alternative structure, which reinforce the core element(s) and bring flexibility to the governance configuration, we used cs/QCA’s sufficiency analysis as it can discern how and under what conditions governance elements from both structures are sufficient for constituting a stable governance configuration. Our outcome of interest for the sufficiency analysis is the presence of stable governance configurations. In our context, the stability of governance configurations is relative. All 56 platforms in our sample had been operating smoothly for at least 3 years and showed considerable maturity by the time of data collection (i.e. early 2018), as detailed in our sampling strategy. However, we acknowledge that platform governance is subject to ongoing evolution and may undergo substantial shifts due to complex external factors. Therefore, “stability” is a platform’s relative status within a relevant temporal framework, with our specific reference point being early 2018.
Governance elements and measures
As summarized in the literature review, markets, and communities differ in the values, interaction modes, and normative and regulative structures employed to govern relationships (Benkler, 2017; Thornton et al., 2012; Zhang et al., 2020). By iterating between our data and the literature, while building on an established framework (Zhang et al., 2020), we specified three dimensions of platforms’ governance of peer relationships—that is, alignment of individual incentives, coordination of interpersonal interaction, and facilitation of normative and regulative structures. We further captured six governance elements associated with market and community structures under the three dimensions. We derived a framework presenting the six elements in which market and community structures manifested in platform governance as a continuum (see Table 1). Furthermore, we obtained the qualitative measures of each element—that is, multiple lower level governance mechanisms performing the consistent function of the element—by analyzing iteratively between our data and the framework (see Tables 2 and 3, Column 2).
Governance elements of peer relationships.
Adapted from Zhang et al. (2020).
Measure and calibration of four elements (Direct method).
Measure and calibration of two elements (indirect method).
Our analysis of the 56 platforms corroborated that markets and communities were two predominant governance structures for peer relationships, consistent with the literature (Benkler, 2017). All platforms tend to adopt market and community mechanisms in each of the six governance elements, but to varying degrees. This close-to-dichotomy presence of market and community structures on sharing platforms allowed us to use, in each element, the “opposite” (or “negative” in QCA terminology) of the community structure to represent the market structure. Accordingly, a higher degree present in an element means that governance mechanisms from the community structure are more prominent than those from the market structure. In comparison, a lower degree in an element aligns more with the market structure, indicating a dominance of market mechanisms. Notably, the degree of an element is relative and non-linear. If a higher degree of an element is observed, it means that, while both are present, community mechanisms outweigh market mechanisms in the element. The closer the case membership to the cross-over (0.5) in that element, the more even the balance between market and community mechanisms. 2
Element 1: signaling organizational identity
A stronger signaling of a socially oriented identity of a platform (IDE) shows its focus on the community structure and developing social bonds between individual participants, reflected by explicit social goals related to collective values, relationships, and wellbeing, and the avoidance of only making revenues from the sharing activities. A weaker demonstration of a socially oriented identity, that is, a stronger signaling of a commercially oriented identity (~IDE) (~ reads: negative), focuses on promoting efficiency, scalability, and profitability and indicated a tendency to facilitate economic transactions and the market structure instead.
Element 2: promoting specific offerings to individual participants
When platforms motivate people with more substantial social offerings (OFF), it suggests a focus on the community structure. Such offerings actively promote social benefits like social and affective demands of human connection, feelings of pleasure and belonging, emotional support, and communication, or downplay or constrain the opportunity for people to make money from those needing their shared assets. By contrast, the market structure is highlighted when platforms provide more economic and utility offerings (~OFF), such as quality, convenience, price, and monetary payback.
Element 3: creating opportunities for repeated interpersonal interaction
Platforms that adopt approaches to create a higher chance of repeated interaction (REP) develop social bonds and favor the community structure. Such approaches include platforms playing a limited role in the search-and-match process and facilitating offline activities and online features for networking. Platforms with a lower chance or attaching little importance to enabling repeated interaction (~REP) depend more on economic transactions and the market structure.
Element 4: creating opportunities for unmediated co-presence
At the time of the data collection, communication tools for cognitive and social tasks, for example, audio and video, were still used sparsely by sharing platforms. So, increasing the chance for platform users to have face-to-face interaction was mainly through offline exchanges, meetups, and events arranged by the platform or participants. Platforms that use these means to facilitate a higher chance of unmediated co-presence between people (COP) strongly emphasize the community structure. By contrast, those that provide fewer of these means display a lower chance (~COP) and favor the market structure.
Element 5: developing central rules and norms
Active promotion of communal norms and shared experiences, for example, verbalizing, specifying, and promoting community guidelines and social norms, as well as facilitating shared experiences and spaces to cultivate a sense of community (NOR), indicates sustaining social bonds between platform users and the community structure. When platforms rely more on formal constraints such as contractual clauses and stringent sanction systems (~NOR), they develop extensive legal provisions and enforceable penalties to control and sustain the functioning of economic transactions, reflecting a stronger market structure.
Element 6: refraining from intensive assurance instruments
Sharing platforms refraining from intensive assurance instruments tend to adopt fewer, more qualified, and less impersonal assurance instruments in the sharing process (RAS) and rely more on the collective monitoring and control principles of the community structure. Platforms that utilize extensive assurance instruments to elicit information and lower transaction costs (~RAS) target transaction efficiency and security and draw more on the market structure (Cheshire, 2011; Einav et al., 2016; Parigi et al., 2017).
Membership and calibration
The next step was to decide a platform’s membership for each of the six governance elements. For fs/QCA, we used the sum of the qualitative measures (governance mechanisms) of a governance element as the raw membership for each platform in this element (see Tables 2 and 3, Column 2 and 3). On each qualitative measure, we applied an interval scale of “0–1” or “0–0.5–1.” The values of 0 and 1 represent sharply contrasting qualitative features, while 0.5 is assigned if an in-between feature is identified.
In fs/QCA, calibration is crucial for turning raw data into final memberships. Deciding the final memberships requires the raw memberships of cases to be adjusted and conform to dependably known standards. In this study, four of the six elements were calibrated with the direct method, using external criteria—theoretical and substantive knowledge that is external to the data at hand (Ragin, 2008b)—to decide the threshold of maximum ambiguity (0.5), full membership (0.95), and full non-membership (0.05; see Table 2, Columns 4 and 5). The final membership was obtained by converting the crossover-centered raw memberships into log odds and final scores ranging from 0 to 1 using a standard formula. We operated the direct calibration in the software fsQCA 3.0 (Ragin and Davey, 2017). The remaining two elements were calibrated with the indirect method since no reliable external criteria were identified (Ragin, 2008b). The indirect method relies on the researcher’s qualitative assessment of all cases to conduct an initial grouping of cases into six preliminary scores: Fully in (1), Mostly in (0.8), More in than out (0.6), More out than in (0.4), Mostly out (0.2), Fully out (0). The preliminary scores were regressed on the raw memberships using a fractional logit model and turned into final scores (Ragin, 2008a; see Table 3, Columns 4 and 5). The indirect calibration was conducted in Stata 14.0. By this step, we obtained all platforms’ final membership scores in the six governance elements. A score above 0.5 in an element meant a higher degree of the element, that is, the platform mainly adopted community mechanisms in that element, while a score below 0.5 represented a lower degree, that is, the platform predominantly incorporated market mechanisms in that element.
In the cs/QCA phase, we generated a new “Governance configuration” dataset consisting of all 64 logically possible governance configurations and defined their presence in the six elements and one outcome (see Table 1,
Data analysis
Our analysis comprised two phases. First, utilizing the “Platform” dataset, we looked for significant interdependencies—necessary relations—among the platforms’ governance elements to capture core elements. Since a platform’s governance configuration might not fully take in or reject a governance element from a structure but draw on it to a higher or lower degree, we opted for fuzzy-set QCA and did a pairwise necessity analysis between every two elements from market and community structures. The degree to which a platform draws on an element from market and community structures was represented by the platform’s membership score in that element, ranging from 1 (full membership) to 0 (full non-membership). 0.5 is the point of maximum ambiguity (Ragin, 2008b), representing the highest mix of both structures. A consistency threshold of 0.90 was applied (Ragin, 2008b). With a consistency value no lower than 0.90 and considerable coverage and relevance of necessity 3 (Schneider and Wagemann, 2013), the presence of an element can be considered necessary for the presence of the outcome. We referred to the element as a core element because substitution by its counterpart from the alternative structure is very unlikely. The presence of the outcome tended to necessitate the core element as baseline requirement or as initial step to form a relatively stable governance configuration.
The second phase identified elaborating elements by evaluating logically possible governance configurations and their association with the emergence of stable governance configurations based on the “Governance configuration” dataset. We conducted a crisp-set QCA of the 64 possible governance configurations, combined with a case-based analysis of the platforms using contextual information (Thomann and Maggetti, 2017). Of the 64 possible configurations, 14 configurations were present among our platform cases and deemed to be associated with stable governance configurations. These configurations received a score of 1 in the outcome “Presence of stable governance” (PRE). For the remaining 50 configurations that were logically possible but not observed in our cases (“logical remainders” in QCA terms), QCA allows the researcher to make assumptions about logically possible but empty configurations with existing empirical evidence and theoretical knowledge (Ragin, 2008b). We made assumptions about their outcomes based on the necessary relationships uncovered in Phase 1. That is, 26 configurations which did not meet these necessary relationships were considered “unlikely to survive” or “infeasible,” obtaining a score of 0 in the outcome. Such assumptions were empirically supported by our data and had considerable foundations in existing theory. Existing literature, both within and beyond the platform context, provides evidence that such necessary, non-substitutable relationships most likely exist between elements related to “identities,” “incentives,” and “institutions” from the same structure due to their high-internal coherence (Battilana et al., 2017; Benkler, 2017; Besharov and Smith, 2014; Siggelkow, 2011). Our necessity analysis results are consistent with these studies. For the remaining 24 configurations, we did not make any assumptions about their outcomes and left their outcomes “empty” in the Truth Table Algorithm.
The 64 configurations of the six elements and the outcome were added into QCA’s Truth Table Algorithm for the sufficiency analysis. Based on QCA standards and our empirical data, we applied a consistency threshold of 0.75 for the sufficiency analysis (Ragin, 2008b). The Truth Table Algorithm generated three solutions. The complex solution only presents the configurations observed in our sample and sufficient for the outcome. The intermediate solution makes those assumptions that align with empirical evidence and theoretical knowledge (Ragin and Sonnett, 2005). The parsimonious solution accepts all assumptions on the empty configurations that can make the solution as simple as possible (Ragin and Sonnett, 2005). We chose to present the intermediate solution that only made assumptions about those empty configurations for which we had empirical evidence and theoretical support, that is, the 26 configurations that did not meet the necessary relationships received an outcome of 0. With this approach, we detected how and under what conditions elaborating elements from the alternative structure reinforce elements from the primary structure and contribute to a stable governance configuration.
Findings
Core elements: tensions in platforms’ hybrid governance
Our necessity analysis reveals two elements with a particularly strong interdependency with other elements from one single structure. The interdependency is so prominent that the two elements become necessary for the latter elements to be present, thereby rejecting the alternative structure (see Table 4). The latter elements, while working coherently with the core, introduce tensions to constructing a hybrid governance configuration and generate rigidity. The two distinctive elements serve as the core elements in the governance configuration, with one from the market structure and one from the community structure. Aside from these two core elements, no other element passes the consistency threshold and necessitates another element.
Necessity analysis results.
Necessary relation: Consistency of Necessity (Cons.Nec) > 0.90. Cov. Nec: Coverage of Necessity. RoN: Relevance of Necessity.
Tensions and rigidity introduced by market elements
The core element in the market structure, a stronger development of contractual rules and enforceable sanctions (~NOR) is necessary for the respective presence of four elements, all of which come from the market structure. The four elements are a stronger signaling of commercially-oriented identity (~IDE), a stronger promotion of economic offerings (~OFF), the facilitation of a lower chance of repeated interaction (~REP), and a reliance on intensive assurance instruments (~RAS). Whenever one of the four elements is present in the governance configuration, platforms resort predominantly to contractual rules and enforceable sanctions to govern peer relationships.
For instance, a systematic development of contractual rules and enforceable sanctions is common among platforms that signal a commercially oriented identity (Consistency = 0.95, Coverage = 0.86). This finding suggests that using formal and contractual arrangements—for example, legal clauses in Terms and Conditions and penalties like fines and loss of deposit—to discipline individual behavior is considered more effective to realize a platform’s pursuit of exchange efficiency, scale, and revenue. These contractual constraints are also perceived to be most appropriate to support an image of promoting efficiency. By contrast, relying on communal norms and shared experiences to control individual action—through articulating community guidelines, promoting reciprocal norms, sharing stories and spaces, and developing collective value and belonging—is perceived as less effective to achieve commercial goals and optimize exchange efficiency and scalability. With less clear enforcement, these informal norms often require time and collective interactions before they are internalized by a large number of participants (North, 1992 [1990]). Informal norms are only used as a modest complement to the formal contractual constraints on platforms with a clear commercial identity. Contractual rules and enforceable sanctions are also perceived as more effective in meeting platforms’ promise of economic and utility benefits to users (~OFF), and when there is limited opportunity for repeated interpersonal interaction to cultivate communal norms and shared experiences (~REP).
The fourth element that necessitates strong contractual constraints is a reliance on intensive assurance instruments (~RAS). Contractual rules and enforceable sanctions (~NOR) are consistently implemented by platforms which rely on intensive assurance instruments (~RAS) to govern peer relationships (Consistency = 0.91, Coverage = 0.83). This necessary relation has two potential explanations. The first one is functional. Contractual constraints could be seen as more functionally compatible with assurance instruments and facilitate efficient and scalable transactions, especially for initial exchanges between strangers. Second, a platform’s dependence on assurance instruments from its early phases onwards could have led to exchange habits relying on such instruments and their continuous improvements, resulting in lock-in effects in user perception and behavior. Their presence hinders introducing communal norms to people who are used already to platform verifications and rating and monitoring systems. For instance, Airbnb constantly enhanced its rating and badge systems and insurance package since its founding, making them more dependable and widely used references for guests and hosts when making exchange decisions than the Community Guidelines. Both explanations indicate a mutual exclusion between communal norms and shared experiences as the core institutional constraint and assurance instruments as monitoring and controlling tools. Social institutions tend to be less compatible with adopting extensive assurance instruments. This finding questions the view that platforms apply various assurance instruments to create social trust between people (Newman and Antin, 2016, 14 March). By extensively implementing assurance instruments, but investing little in aligning them with communal norms, they are virtually reducing exchange uncertainty and improving security instead of generating trust that sustains (Nissenbaum, 2004).
Tensions and rigidity introduced by community elements
The second type of tension relates to a core element associated with the community structure—signaling of a pronounced socially oriented identity (IDE)—which is a necessary element for stronger development of communal norms and shared experiences (NOR; Consistency = 0.93, Coverage = 0.80). Platforms with specific communal norms and shared experiences to govern peer relationships are always among those conveying more distinct social missions and identities rather than commercial goals and identities. Mainly platforms which genuinely identify themselves with enabling and sustaining social bonds between participants are motivated to invest in social institutions. This finding implies that fostering communal norms and shared experiences might come at a high cost so only staunchly socially oriented platforms are willing to make such an investment. Food Swap Network, for example, connects local community members through recurring volunteer-hosted events where people share homemade food based on a set of norms. BeWelcome explicitly resists top-down platform decision-making and has a democratic and volunteer-based governance structure. Both platforms presented limited intention to rapidly scale their user base but focused on their social missions.
The status of a deep-rooted social identity as a core element with respect to community norm cultivation may also arise through tensions between the latter and claims of a strong commercial identity. Demonstrating commercially oriented identities and goals can signal a platform’s investment in scale-and-revenue-focused practices. It attracts participants with economic motives to exploit shared resources but can filter out those who are enthusiastic about a social connection and community feeling. It can be challenging for a platform that attempts to build a community founded on informal social norms to justify prominent commercial goals. When Couchsurfing changed from a non-profit to a “B Corporation” in 2011, and when it required a compulsory membership fee during COVID-19, it had to make great efforts to address a drop in user confidence, reassuring users that the change had nothing to do with the identity and norms of the Couchsurfer community but were imperative to relieve the severe financial pressures it was facing (Baker, 2011, 26 August; Lapowsky, 2012, 29 May).
Asymmetric tensions
The findings suggest an asymmetry in how market and community elements bring tensions into hybrid governance and therefore rigidity in designing governance configurations. The core element in the market structure (~NOR) and the elaborating elements it interacted closely with (~IDE, ~OFF, ~REP, ~RAS), differ greatly from the core element in the community structure (IDE) and its elaborating element (NOR). The presence of the four elaborating elements from the market structure and the one from the community structure constrain a platform’s core elements in distinct ways. For instance, the notion that stronger signaling of a particular identity (~IDE) limits a platform’s use of normative and regulative structures within the same structure (~NOR) is supported by the evidence from the market structure. Platforms with a strong commercially oriented identity always prioritize contractual rules and enforceable sanctions. However, the opposite does not apply to platforms with a strong socially oriented identity. A consistency of 0.80 for the latter is far lower than the threshold of 0.90, suggesting that for platforms with a prominent socially oriented identity, governing peer relationships with communal norms and shared experience is not imperative. Some of these platforms also make considerable use of contractual rules and enforceable sanctions, even though these are embedded in the market structure. Hence, there is a difference between having a commercially oriented and socially oriented identity in constraining the governance configuration. This finding suggests that formal and contractual rules are relatively value-free: using them with moderation could fit socially oriented platforms without causing too much tension. By contrast, commercially oriented platforms perceive informal and community norms as unnecessary, costly, or risky contradicting the optimization of efficiency, scalability, and revenue-making.
Contrary to the tension and rigidity stemming from the community structure, use of contractual rules or enforceable sanctions does not necessitate any other elements from the market structure. Implementing contracts and sanctions does not have any stringent requirements for the type of organizational identity and offering. It does not constrain what type of interaction mode is facilitated or how intensive the assurance instruments need to be. Platforms stressing interpersonal interaction and their unique offerings of improving human connection and social wellbeing—for example, BlaBlaCar—still formulated comprehensive legal clauses to regulate the contracts between users as well as enforced assurance instruments. However, if a platform has primarily based its governance on community norms and experiences, integrating commercial elements into its social identity would become challenging and could provoke opposition from the community. Using contractual rules and enforceable sanctions appears to be more versatile than cultivating community norms and shared experiences for the governance of peer relationships.
Elaborating elements: complementarities and flexibility in reconciling with core elements
Incorporating those rigidity-generating elaborating elements necessitates the choice of the core element from the same structure, but not all elements exhibited this property. In the second phase, we investigated how the remaining elements—particularly those from the alternative structure—were configured differently to complement these core and elaborating elements to constitute a stable governance configuration. Despite the much stronger coherence between governance elements within a single structure, we expected some level of flexibility in incorporating elements from the alternative structure based on the results from the necessity analysis. In the following, we identify the elements that enabled such enhanced flexibility in cross-structure complementarity and in governance configurations.
Table 5 presents the intermediate solution that minimizes the 64 potential configurations to seven models (Models 1–7). The symbol “-” denotes elaborating elements where some platforms lean toward the community structure while others favor the market structure. Within each model, different elaborating elements support and reinforce the core element, jointly establishing a stable governance configuration. Each of the four dominant models (Models 1–4), which encompass the majority of the 56 platforms, incorporate at most two elements from the alternative structure, indicating a lower level of hybridity with a favored distribution of elements from the primary structure. However, among these models, complexity exists since some platforms may prioritize the primary structure—market or community—in the remaining four/five elements, but in a more balanced way (fuzzy memberships close to 0.5 in these elements). These platforms show a relatively higher level of hybridity compared to other platforms which adopt the same model but significantly prioritize one structure in the remaining elements (fuzzy memberships close to 0 or 1). In contrast, for Models 5–7, a few platforms not only bring in more elaborating elements (four) from the alternative structure, but also exhibit a more balanced degree of hybridity within their rigidity-generating elements (Models 5–7), or even diverge from strong interdependency to favor the alternative community structure in their core element (Models 6, 7). Integrating a greater number of governance elements predominantly from the alternative structure as well as balancing both structures or prioritizing the alternative structure in the core element within the remaining elements in these models suggest a very high level of hybridity (Shepherd et al., 2019).
Sufficiency analysis results (Intermediate solution).
PRI: Proportional Reduction in Inconsistency.
• represents an element where the community mechanisms are prioritized; ⊗ represents an element where the market mechanisms are prioritized; - represents an element where some cases prioritize community mechanisms while others prioritize market mechanisms. • and ⊗ (in larger sizes) represent the core elements identified from the necessity analysis.
The findings suggest that for all seven models, integrating governance mechanisms predominantly from the alternative structure in certain elements do not significantly affect their alignment with the primary structure and the overall stability of the governance configuration. They further indicate that there is still considerable flexibility for platforms in designing a hybrid governance configuration. Necessity analyses between the elements and the outcome “Presence of stable governance” (PRE) were conducted before the sufficiency analysis, and no market and community element turned out necessary for the outcome.
Dominant models with a lower level of hybridity
Model 1 covers 28.6% of memberships for the presence of a stable governance configuration (Raw Coverage = 0.286) 4 and includes four configurations (2; 4; 7; 3). The model possesses three of the four rigidity-generating elaborating elements from the market structure—signaling of pronounced commercially oriented identity (~IDE), active promotion of economic offerings (~OFF), and reliance on intensive assurance instruments (~RAS)—which limit the choice of the core element to developing stronger contractual rules and enforceable sanctions (~NOR). Meanwhile, it presents flexibility in creating opportunities for repeated and unmediated interactions, as platforms can either prioritize the market or community structure in these two elements. This model covers platforms which generally apply market mechanisms but only strategically facilitate repeated or face-to-face interactions when necessary or deemed beneficial for market growth. Short-distance mobility platforms exemplify this model, where peers meet only for brief, economical, one-off access. They are often capital-driven with a clear commercial focus, highlight the economic offerings of efficiency and low cost, and use intensive assurance instruments that depend on and enhance their contractual arrangements. While face-to-face interaction occurs naturally during transactions, ride-hailing companies such as Uber and Didi Chuxing do not actively pursue opportunities for users to engage in repeated exchanges or expand the likelihood of face-to-face interactions.
However, some platforms under this model do implement mechanisms to increase the probability of recurrent or unmediated interactions. For example, while Uber and Fiverr.com share similarities in four elements, they diverge notably in the frequency and nature of repeated and face-to-face interactions. This distinction largely stems from the type of assets being exchanged and the intricacies of the information required for the exchange. Uber users primarily seek immediate rides, requiring straightforward details like the plate number, expected arrival time, and driver feedback. In contrast, Fiverr tasks, such as design, writing, and programming, involve more nuanced and tacit exchange information, necessitating intricate communication. This complexity in communication provides stronger incentives for users on Fiverr to engage in repeated interactions, leading the platform to introduce more community-oriented features, like social networking tools and online forums. These differences in managing interpersonal interactions are tailored to cater to each platform market’s unique requirements for eliciting information and reducing transaction costs (Einav et al., 2016).
Model 2 accounts for 28.6% of relevant memberships (Raw Coverage = 0.286) and covers four configurations (14; 13; 12; 1). Its core element of signaling a pronounced socially oriented identity (IDE) is reinforced by three elements from the community structure: creation of higher chance of repeated interactions (REP), development of solid community norms and shared experiences (NOR), and limited adoption of assurance instruments (RAS). Platforms employing this model to manage peer relationships exhibit greater flexibility in leaning toward either market or community structure to design their specific offerings and facilitate unmediated co-presence. This finding suggests that even platforms with a strong social orientation can emphasize economic and utilitarian offerings to attract users, and primarily bank on online peer interactions to realize their promised offerings and focal social objectives. A prime example is Zhihu, a Chinese online knowledge-sharing platform. While maintaining a free Q&A community, it also promotes paid services for premium knowledge products from industry professionals and experts. A few platforms within Model 2 still prioritize offline activities, as exemplified by community-driven initiatives like Food Swap Network, MamaBake, and ClothingSwap. Yet, most platforms recognize the value of combining online and offline interactions to ensure diverse member engagement and interaction. For example, while platforms like BeWelcome, Couchsurfing, and the entrepreneur network Economy of Hours primarily facilitate online searching and matching, they actively encourage offline meetups and events among members.
Model 3 comprises 14.3% of relevant memberships (Raw Coverage = 0.143). It covers two configurations (6; 8) and three platforms. Platforms adopting this model demonstrate commercial-focused identities (~IDE) and lean heavily on contractual arrangements (~NOR). However, they show flexibility in terms of the economic and social offerings they promote for participants. Unlike the ride-hailing platforms in Model 1, these platforms lack the spontaneous contexts for repeated and face-to-face interaction (~REP, ~COP). Most of these platforms have a slightly lower reliance on assurance instruments, which could see an increase in the future (RAS = 0.57). For example, despite its underspecified social goals and community norms, Love Home Swap attracted users with the home-swapping sector’s long-established reciprocal exchange mode where exchange partners stayed for free and took care of each other’s vacant home as well as the unique selling point of saving travel cost and increasing exchange efficiency with calculable virtual currency. Park On My Drive users post offers and make instant bookings online with little need for repeated and face-to-face interactions to acquire information about parking spaces besides locations. Yet, the platform promotes the notions of sustainability and community efforts to attract environmentally aware users.
Model 4 accounts for 14.3% of relevant memberships, including one configuration (5) and three cases (Raw Coverage = 0.143). This market-centric model benefits from the spontaneous context of face-to-face interaction (COP) and is flexible in its type of offerings to attract users (OFF). Long-distance carpooling platforms have a strong presence in this model, in contrast to the short-distance ride-hailing platforms of Model 1. Platforms like Kangaride and BlaBlaCar, even if primarily utilized for singular, long-distance trips, emphasize the significance of forging social bonds and trust over the mere monetary benefits. This configuration suggests a governance approach that, while market-centric, is enhanced by the positive social interactions experienced during extended journeys.
Rare models with a higher level of hybridity
Model 5 covers 7.1% of relevant case memberships and encompasses only one platform: HomeExchange (Raw & Unique Coverage = 0.071). The model integrates four elaborating elements from the community structure while maintaining coherence in its market-oriented identity and institutions. Compared to Love Home Swap in Model 3 (IDE = 0.02), HomeExchange demonstrates a more balanced social and commercial identity, seeking economic efficiency while building social bonds between home exchangers (IDE = 0.43). Reciprocal exchange is a predominant mode of exchange in its community, but the platform also adopted a non-calculable virtual currency, highlighting egalitarianism that all homes are equally valued despite sizes, locations, and amenities. The governance configuration largely prevented users from perceiving the platform as commercial and transactional.
Model 6 also covers one case only (Raw & Unique Coverage = 0.071). Despite its clear commercial goals and identity (~IDE), this model features prominent community norms and shared experiences (NOR), integrating three elements from the community structure: more substantial promotion of social offerings (OFF), higher chance of repeated interaction (REP), and avoidance of intensive assurance instruments (RAS). There is one representative case in our dataset, Quirky, whose peer interactions are conducted online only. However, a relatively stronger commercial identity (~IDE) and a limit to mediated interaction (~COP) does not prevent the platform from infusing prominent community norms into user interaction (NOR). This case is rare because it contradicts the necessity test result that predominant contractual rules and enforceable sanctions (~NOR) are necessary institutions for platforms with a stronger commercially oriented identity (~IDE). At the time of the data collection, Quirky was a for-profit business with a clear goal of commercializing the inventions of users and a structure for revenue distribution. However, Quirky’s emphasis on facilitating crowd-based product innovations through community collaboration neutralized the signal of its for-profit business considerably, making it a case with a balanced organizational identity (IDE = 0.43). It also implemented innovative practices to execute its balanced goals, for example, setting up rational user expectations for profits, designing a crowd-based revenue-sharing system, developing free Discussion and Following features, and moderating a collective- and process-monitoring system. Quirky tried to meet its goal of capturing economic value based on a collaborative product development community.
Finally, Model 7 covers one case: WeWork (Raw & Unique Coverage = 0.071). Despite its apparent commercial goal (~IDE) and the crucial role of assurance instruments (~RAS), the co-working platform’s unique selling point was its facilitation of social connections, physical interactions, and network resources for entrepreneurs (OFF, REP, COP). It also contradicts the necessity test result that consistently implementing contractual rules and enforceable sanctions are necessary for platforms with significant commercially oriented identities. WeWork compartmentalized its social norms in the shared workplace (NOR) from its corporate pursuit of profitability and scalability. However, the unique conditions that made this blend of governance elements possible are not usually available to other platforms, such as the physical spaces for frequent face-to-face interactions, entrepreneurial members’ strong desire to forge social bonds, and the openness of the entrepreneurship culture. This model bears similarities to a “compartmentation” (Kraatz and Block, 2008) or “separation” strategy (Benkler, 2017) of positioning competing structures in separate organizational units and activities.
Conclusion
Our study shows that platform governance of peer-to-peer relationships between platform participants combine market and community mechanisms in different ways, reflecting different forms and degrees of hybridity. Governance elements from market and community structures are not equal in their influence on how peer relationships are incentivized and coordinated, as they play unique roles and interact differently in establishing a stable governance configuration.
We identified two core elements (one from each structure)—stronger development of contractual rules and enforceable sanctions and signaling of a pronounced socially oriented identity—which are fundamental to platform governance, as they are highly interdependent with multiple elaborating elements from the same structure and are hard to replace. We also identified elaborating elements, from either structure, which entail and/or reinforce these two core elements to allow platforms to form a coherent governance configuration. The first type of elaborating elements brings rigidity to the design of hybrid governance configurations. They necessitate the presence of a core element from the same structure, considerably constraining platforms in their choices of governance structure. For instance, platforms that promote strong commercial identities and offerings, lack opportunities or motivation for repeated interaction, and implement intensive assurance instruments tend to establish strict contractual rules and enforceable sanctions as primary institutions to control peer relationships. In comparison, platforms focusing on cultivating social institutions, like community norms and shared experiences, always project authentic social identities. This asymmetric interaction between governance elements indicates that market and community structures tend to introduce tensions of a distinctive nature to platform governance. Yet, the second type of elaborating elements, especially those related to platform offerings and interaction modes, provide more flexibility for incorporating distinct governance elements and associated mechanisms from the alternative structure, without causing severe tensions. This enhanced flexibility in cross-structure complementarity allows platforms to promote diverse offerings and encourage varying degrees of repetitive and face-to-face user interactions, no matter what the other elements are of their governance configuration.
The distinctive roles of specific governance elements, and the asymmetric tensions and complementarities arising from the interplay of elements embedded in market and community structures, shape how platforms design their governance configuration in a hybrid form. We capture two approaches that increase the hybridity in platform governance of peer relationships: balancing the distribution of elements from both market and community structure across the configuration (i.e. even distribution), or balancing the degrees of market and community mechanisms within a particular element (i.e. strong hybrid intensity; Shepherd et al., 2019). Among the seven identified governance models, four (Models 1, 3, 4, 5) are built around strict contractual rules and sanctions and are principally guided by the market structure, with elaborating elements reinforcing this core element in varied ways. Only one model (Model 2) is built around social value creation and firmly grounded in the community structure. While utilizing economic incentives and online interactions, this model arranges elaborating elements in ways that consistently reflect and reinforce the platforms’ social goals and identities. Most platforms with a governance configuration built around either contractual rules and sanctions or a socially oriented identity only sparingly use elaborating elements from the structure different from the core element’s structure, leading to a lower level of hybridity with a pronounced distribution of elements from a single structure. Yet, some platforms in Models 1, 2, 3, 4 may have relatively higher hybridity than others due to stronger hybrid intensity within certain elements. In contrast, a few platforms have ventured further into using configurations that seem more contradictory (Models 5, 6, 7). These rare governance configurations have a very high level of hybridity both across and within elements. They either incorporate multiple elements from the alternative structure or strike a balance between both structures in the remaining elements. While seemingly contradictory, these platforms provide valuable insights into the potential for crafting more innovative governance configurations.
Contributions to theory and practice
Our findings make critical contributions to research and practice in the fields of platform governance, new forms of organizing, and the sharing economy.
First, with a focus on the less-developed area of how platforms govern peer relationships, we systematically uncover governance configurations created in a hybrid and potentially contradictory form while achieving relative stability (Altman et al., 2022; Benkler, 2017). We show that market and community mechanisms interact largely synergistically rather than at random. Each set of mechanisms creates unique complementarities and tensions that shape how platforms develop their hybrid governance configurations. Current literature highlights the potential for market and community structures to complement each other to foster innovation and create value through the efforts of autonomous individuals (Altman et al., 2022; Benkler, 2017). However, this potential and associated risks are primarily discussed in the context of theoretical propositions and case studies, rather than being investigated and validated through systematic research (Altman et al., 2022; Benkler, 2017; Boudreau and Lakhani, 2009; Reischauer and Mair, 2018). Drawing on organizational design and configuration perspectives (Fiss, 2007; Siggelkow, 2002), we bring to the fore the synergistic nature of platform governance configurations. We identify the distinct roles (“core” “rigid elaborative” “flexible elaborative”) and the consequent asymmetric interactions between market and community mechanisms as a fundamental dynamic shaping the design of specific governance configurations with hybridity both across and within elements. Our configurational analysis of 56 sharing platforms that create and capture value from peer interactions provides systematic evidence for current patterns, possibilities, and constraints of incentivizing and coordinating peer relationships on digital platforms. Although the hybrid governance of top-down platform-user relationship has been studied (Chen et al., 2022; Einav et al., 2016; Gulati et al., 2012), we show insight into how digital platforms integrate market and community mechanisms in stable and innovative ways to govern peer relationships instead (Reischauer and Mair, 2018; Schüßler et al., 2021).
Second, our study reveals the design considerations of platforms when developing stable and innovative ways to govern peer relationships as a new form of organizing (Gulati et al., 2012; Puranam et al., 2014). We explain the variation in platform governance by clarifying the “causal architecture” where internal interactions between mechanisms can influence the design of stable governance configurations. The primary set of design considerations points to the need to prevent intractable tensions and exploit the efficiency of a single, coherent governance structure (Altman et al., 2022). It involves selecting the core element within the primary governance structure when certain elements and conditions are favored or present on a platform. Key among these elements and conditions are commercial identities and offerings, absence of repeated interaction, assurance instruments, and social institutions. Such considerations tend to prioritize stability over innovativeness in creating hybrid configurations and are largely consistent choices across different platforms. The second set of design considerations entails the evaluation and exploration of potential complementarities offered by the alternative structure (Altman et al., 2022), by incorporating its elaborating elements and associated governance mechanisms. Such experimentation processes present more unique choices based on specific platform conditions and capabilities. Our findings show that choices of such elaborating elements that may enhance complementarity with the primary structures vary across different configurations and platforms. Under specific platform conditions or with innovative capabilities (Altman et al., 2022), platforms have more leeway to create governance configurations that more liberally combine elements from different structures without causing too much tension or even generate highly mixed configurations. These conditions and capabilities include a spontaneous context for face-to-face interaction, entrepreneurial members’ desire to forge social connections, an open entrepreneurial culture, a space of segregating inconsistent rules and activities, and specific characteristics of resource and exchange information. A crucial notion is that these considerations neither completely incorporate nor entirely exclude governance elements from the alternative structures but employ them to varying degrees to seek a balance leading to unique governance configurations (Battilana et al., 2017).
While theoretical literature underscores the strong impact of complementarity in creating “frequently occurring clusters of solutions” across new forms of organizing (Puranam et al., 2014: 174), our specifications of the nuanced design considerations enrich this literature by showing under which conditions of flexibility and rigidity can occur by detailing the specific elements and interactions which generate complementarities and tensions. Our findings are also consistent with organizational research beyond the platform context that argues that more conflicts between structures tend to exist in organizational goals than in means and that the latter is more manageable (Pache and Santos, 2013). Our research thus provides a systematic explanation for the varied forms and degrees of hybridity in platform governance by highlighting the significance of clearly distinguishing those elements that create tension. How a platform’s commercial identity constrains platform choices can create very different dynamics and evolutionary pathways compared to the influence of established community norms, with consequences for platforms’ agency to mitigate tensions and pursue innovation. Clarifying the “causal architecture” of platform governance design moves beyond uncovering the direct impacts of individual mechanisms as it identifies the interactive, sequencing, and context-based nature of its causal mechanisms (Gulati et al., 2012; Puranam et al., 2014).
In addition, our findings corroborate previous literature which argues that digital technologies and algorithmic mechanisms have broadened the governance choices for platforms (Chen et al., 2022). Yet, our findings also reveal that platform governance considerably draws on conventional organizational mechanisms, such as social identity demonstration and construction of formal and contractual institutions, on which digital technology has less impact. Using such conventional mechanisms as foundation, platform governance incorporates traditional, non-digital mechanisms like offerings and offline activities, as well as novel, digital mechanisms such as mediated interactive tools, assurance instruments, and virtual community spaces. While some digital mechanisms like mediated interactive tools and networking features can be used flexibly with conventional governance mechanisms, digitally enabled assurance instruments tend to limit platforms’ regulative structure considerably to a high reliance on contractual arrangements. It challenges the widespread view that platforms build social trust through various assurance instruments (Newman and Antin, 2016, 14 March). Instead, we found that these efforts, when lacking in community norm alignment, mainly reduce uncertainty and boost security, which diverges from their claimed community-building intention (Nissenbaum, 2004).
Finally, our study clarifies the complexity of how the sharing economy addresses social and economic value creation (Mair and Reischauer, 2017). While some platforms like Uber and Getaround blended the community structure into their offerings to increase their public legitimacy, they did not intend to or lack the conditions to implement community-oriented elements more systematically. However, there were platforms, too, experimenting with a highly mixed governance configurations, while trying to scale their activities. Yet, cases like Couchsurfing and HomeExchange experienced legitimacy challenges when trying to scale their “community.” By showing how platforms combine market and community structures in their governance configuration, we shed light on the long-standing controversy whether sharing platforms try to deliver on the promise of “new forms of collaboration, solidarity and social bonding” or just “recreate the inequalities of the capitalist markets, but in different ways” (Acquier et al., 2017: 2; Benkler, 2017; Eckhardt and Bardhi, 2015). We show that, indeed, conventional market-based incentives and regulative structures are dominant in most platforms’ governance configurations. However, platforms with a strong community orientation can also use economic incentives and contractual arrangements, although with moderation, without causing too much tension, and there are still platforms that are almost fully built around a socially oriented identity. Such differences in evolutionary trajectories are tied to platforms’ design choices and contextual factors, as we discussed earlier.
Future research
Our study has several limitations which present opportunities for future research. First, the variation in governance configurations is representative of platforms in a specific stage of development. We examined relatively mature platforms and our (sufficiency) analysis assumed that the governance configurations of platforms founded more than 3 years ago were relatively stable. However, platforms constantly experiment with their digital interface and governance configuration, and they continuously evolve as organizational form. Future research could generate interesting insights by tracking platforms’ ongoing developments to verify if the same pattern of core and elaborating elements sustain over time or significantly change instead.
Second, our analysis provided only some insights into how digital technologies are changing the nature of platform governance. We were mainly interested in getting a better understanding of how platforms combine market and community structures to manage peer relationships, pushing the question of how distinct incentive and coordination mechanisms—digital and non-digital—interact more into the background. However, with the fast-paced development of digital technologies, especially around (generative) artificial intelligence (AI), one might expect these technologies to become more influential soon. Future research should examine how transformative digital elements change the design and design considerations of platform governance, both in creating novel configurations or novel algorithmic mechanisms that may simultaneously fulfill the goals and functionalities of multiple structures.
Finally, while we identified some of the underlying conditions that allow platforms to create more innovative governance configurations, more research is needed to gain insight into the boundary conditions of platform governance. Interestingly, the two cases in our analysis—WeWork and Quirky—which presented configurations contradicting the necessity tests have been struggling since we collected our data. However, it is too easy to say that the contradictory configurations formed the root cause of their failure. It is worth conducting more research on such pattern-breaking cases to investigate the specific conditions that enable their unique configurations and assess how stable these are in the long run.
Supplemental Material
sj-docx-1-soq-10.1177_14761270241246603 – Supplemental material for Hybrid governance of digital platforms: Exploring complementarities and tensions in the governance of peer relationships
Supplemental material, sj-docx-1-soq-10.1177_14761270241246603 for Hybrid governance of digital platforms: Exploring complementarities and tensions in the governance of peer relationships by Yaomin Zhang, Jonatan Pinkse and Andrew McMeekin in Strategic Organization
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Supplemental material
Supplemental material for this article is available online.
Notes
Author biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
