From Policy-Practice to Means-Ends Decoupling in Organizations: A Systematic Review and Paths for Future Research

Jabbouri, Rachid; Schneckenberg, Dirk; Truong, Yann

doi:https://doi.org/10.7202/1088440ar

We address in this review paper the phenomenon of means-ends decoupling. Means-ends decoupling refers to the puzzle of why many organizations fail to achieve their intended goals despite investing substantial resources and compliance inducements. The relevance of the means-ends decoupling construct derives from its capacity to identify lower-level mechanisms that the traditional policy-practice lens fails to capture and provide new insights into the field-level consequences of decoupling practices (Boxenbaum and Jonsson, 2017; Wijen, 2014). Early research on decoupling suggests that organizational responses to external pressures, such as legislation, public policy, and social activism, often resulted in internal buffering of daily practices outside control and inspection. Such actions were implemented to decouple formal policies from firms’ internal technical core and daily work routines (Meyer and Rowan, 1977; Pfeffer and Salancik, 1978). Therefore, organizations adopted policies to respond to external pressures regarding formally stated legislation and regulation while keeping their internal practices unchanged (Déjean and Oxibar, 2010). In that sense, policy-practice decoupling has been recognized as an optimal way for organizations to shield their routines and strategic goals from external pressures (Westphal and Zajac, 2001).

The quantity and quality of decoupling research increased noticeably throughout the last decade. As a result, decoupling studies gain impact in the management discipline, with prominent scholars devoted to studying this phenomenon (e.g., Bromley and Powell, 2012; Tilcsik, 2010) and well-recognized conferences dedicated to its exploration (e.g., the Organization and Management Theory division of the Academy of Management had a track on Institutional Decoupling in 2008). However, early research on decoupling primarily focused on a single form of the phenomenon, namely policy-practice decoupling. While policy-practice decoupling helped organizations manage external pressures with minimal changes in practice, its symbolic purpose is increasingly debated. The level of pressures organization’s face has increased over the last decades, motivating studies that describe this significant change in the business context as “audit culture” (Strathern, 2000) or “audit society” (Power, 1997). The growing focus on accountability, assessment, and monitoring in contemporary societies has driven organizations to devote more effort to conforming to adopted policies and aligning the latter with internal practices and daily activities. These organizations act as “corporate citizens” (Meyer and Bromley, 2012) to display proper behaviours, especially for corporate social responsibility. The intended ends might be significantly different from or even contradictory to the core goals of these organizations. Scholars characterize these fundamental changes in the environment in which contemporary organizations perform as “the institutional environment rationalization” (Frank and Meyer, 2002; Zucker, 1987).

Following growing levels of control, monitoring, and evaluation, organizations feel compelled to devote more efforts to implement policies that they initially adopted in a symbolic fashion (Persais, 2010). However, even when organizations try to implement policies in their internal practices more genuinely, they still risk failing to attain the intended goals (Vo et al., 2016; Crilly et al., 2012). For instance, Jabbouri et al. (2019) studied how a research and development consortium failed to achieve the intended goal of a collaborative project even though its partners had allocated substantial resources to the project. Likewise, Dick (2015) showed how adopting part-time working by the UK police service only led to a marginal surge of this type of position within police units. Recent institutional work has raised attention to this new form of decoupling, namely means (practices) and ends (goals), to explain how the gap between practice and goals occurs even when organizations attempt to implement policies (Wijen, 2014; Bromley et al., 2012).

Given these significant changes in the institutional environment in which contemporary organizations operate, there is a need for re-examining traditional conceptions of decoupling and achieving a deeper understanding of the circumstances under which this phenomenon occurs, its corresponding consequences, as well as its inherent, institutionally embedded factors and causes (Bromley and Powell, 2012). Although some recent studies started to unravel some mechanisms leading to means-ends decoupling (Dick, 2015; Jabbouri et al., 2019), much remains to be done to explain the causal complexity lying in this form of decoupling (Wijen, 2014). To fulfil these ends, we review the extant literature on decoupling in this paper, clarifying the distinction between the two forms of the phenomenon: policy-practice decoupling and means-ends decoupling. Our study addresses this knowledge gap by pursuing the research question: What is the current state of means-ends decoupling research, and where can it move in the future?

Our research question addresses a critical paradox in the management literature: Even when organizations seriously attempt to satisfy the pressures of external stakeholders by implementing new policies, they risk failing to attain the goals set by these new policies. Thus, scholars need to elicit the instances when such a gap between means and ends may emerge and what can organizations do to reduce this gap. Providing insights into this puzzle contributes to extant institutional literature related to how stakeholders can induce organizations to meet their broader social expectations. Thus, we focus specifically on the emerging form of means-ends decoupling, which results from the increased rationalization and fragmentation of the institutional environment in which contemporary organizations operate (Boli, 2006; Strathern, 2000; Power, 1997). The lack of alignment between adopted policies and internal practices less likely occurs in modern organizations, while the gap between deployed means and envisaged ends more likely prevails in rationalised environments (Wijen, 2014). Interestingly, Boxenbaum and Jonsson and (2017) specify that means-ends decoupling even occurs when institutional actors do their best to fulfil to implement agreed-upon action at the organizational level. Figure 1 displays policy-practice decoupling and means-ends decoupling as the two complementary forms of the phenomenon.

Figure 1

**Distinctions and transitions between policy-practice and means-ends decoupling**

Within the broad range of literature review design (e.g., meta-analysis, bibliometric, or critical reviews) and their respective capacities for producing reliable and viable review studies, given the purpose of our study and the nature of our research question, we selected the systematic review approach as suitable for our research. Specifically, we aim at providing a comprehensive review of studies published within the field of decoupling. Furthermore, the diversity of methodological approaches employed in examining policy-practice and means-ends decoupling, and the conceptual nature of many decoupling studies, qualify the systematic literature review to provide a transparent, reliable, and viable review of extant decoupling research.

The remainder of this article is structured as follows: First, we delineate the employed systematic literature review approach stages. Second, we portray the fundamental tenets of the two forms of the phenomenon, with a particular focus on the emergent phenomenon of means-ends decoupling. Third, we present potential avenues for future research and advance a set of conceptual insights relevant to the manifestation of means-ends decoupling. Finally, we discuss these conceptual insights’ contributions to the current research and their role in framing potential practical solutions for contemporary governance issues.

Review Methodology

The literature review process is an essential instrument for organising the diversity of knowledge in a particular academic area. Such exercise aims to enable researchers to assess the existing body of knowledge and map the agenda for future research for a specific discipline or field. There are different approaches to conducting a literature review, including the meta-analysis review (Westhorp et al., 2013), bibliometric review (Randhawa et al. 2016), critical review (Massa et al., 2017), and the systematic literature review (Foss and Saebi, 2017). If appropriately applied, each of these approaches can be helpful given the aim and purpose of the performed review. To elaborate on our choice of the systematic literature review approach, we delineate key differences between these various approaches and the rationale behind our choice.

The critical review approach aims to assess and summarise the literature on a given research topic and hold the premise for enabling the emergence of new theoretical frameworks and perspectives (Torraco, 2005). Nevertheless, its use within management research remains limited. Rather than reviewing conceptualisations based on covering all articles on a particular research topic, the critical review develops theorising and combining perspectives and insights from different research fields and paradigms (Snyder, 2019).

The meta-analysis and the bibliometric review follow statistical methods for combining results from articles to identify and compare patterns and relationships within a research area (Westhorp et al., 2013). However, the challenge lies in the idea that the meta-analysis and bibliometric reviews apply only to studies that share their statistical measures and allow for results comparison, making these two review strategies unsuitable for reviewing studies following different methodological approaches (Tranfield et al., 2003).

For our research, we refer to the systematic review as “a research method and process for identifying and critically appraising relevant research as well as for collecting and analysing data from said research” (Snyder, 2019, p. 334). This approach has emerged as an alternative process for reviewing research in management and organization studies, which can compensate for weaknesses of traditional review processes (e.g., narrative reviews) and attenuate criticisms the latter often suffer from (Tian et al., 2018; Rousseau et al., 2008; Davies and Crombie, 1998; Denyer and Tranfield, 2008 Tranfield et al., 2003). The systematic literature review synthesises literature in the same substantive domain (Cortez et al., 2021, Palmatier et al., 2018). Following predetermined guidelines, the systematic literature review provides collective insights through theoretical synthesis, increases methodological rigour, and helps develop reliable knowledge through synthesising and accumulating contributions from a broad range of studies.

Against this background, the systematic review approach is appropriate for our research, given the purpose of our study and the nature of institutional decoupling research. The systematic literature review is suitable for obtaining a comprehensive summary of the current research state, as it helps grasp the diversity of methodological approaches employed in policy-practice and means-ends decoupling research and the conceptual nature of many decoupling studies. Consequently, following early research in management and organization studies (Cortez et al., 2021; Tian et al., 2018; Denyer and Tranfield, 2008; Davies and Crombie, 1998; Rousseau et al., 2008), we employ in this paper the systematic literature review process in conducting our literature review of existing institutional decoupling research.

Table 1

**Stages of the Systematic Literature Review**

Stage 1: Search Strategy

Phase 1: Research Panel

A primary phase to conducting a systematic review constitutes forming a review panel by soliciting experts in the fields of theory and methodology (Tranfield et al., 2003). For the sampling logic of panel members, we included expert researchers who have knowledge and expertise related to the phenomenon under exploration and the employed methodological approach. Our selection criteria consisted of selecting scholars by their capacity to provide rich insights about the subject under investigation and their ability to guide us through the systematic review process. Consequently, we included two colleagues with expert knowledge about decoupling and institutional theory and a scholar with methodological expertise in performing systematic literature reviews. The panel has helped us specify key terms, define inclusion/exclusion criteria identification, and consolidate the final sample of retained studies.

Phase 2: Keywords Design

To identify a set of relevant key terms for our search, we first relied on a preliminary analysis (Judge et al., 2007) to assess the literature size and relevance and outline the subject area of institutional decoupling. This primary analysis also included a brief overview of the cross-disciplinary perspectives and theoretical and methodological approaches employed to investigate the phenomenon of institutional decoupling. Second, following Snyder’s (2019) methodological procedures, we focused on terms directly related to the explored research question. Third, following Tranfield’s et al. (2000) systematic review guidelines, we solicited the review panel to validate the main key terms that efficiently capture relevant institutional decoupling studies. Accordingly, we decided to employ the following key terms for our search strategy: 1) “policy-practice decoupling”; 2) “means-ends decoupling”; 3) “symbolic management”; 4) “institutional theory”; and 5) “decoupling.”

Phase 3: Database Selection

Our choice of using the EBSCO Business Source Premier database for the search follows Foss and Saebi’s methodological choices (2017) in their systematic review paper. Employing the same database in performing their systematic literature review, Foss and Saebi (2017) used the EBSCO Business Source Premier database to provide a comprehensive and widely cited systematic literature review for business model innovation. Therefore, we combined the five key terms following our discussion with the review panel and concluded that using the terms “decoupling,” “symbolic management,” and “institutional theory” in independent searches results in a large number of less relevant articles in the search results. As such, we agreed to build two search entries that incorporate both “decoupling” and “symbolic management” (search 3) and “decoupling” and “institutional theory” (search 4). Table 2 summarises the four search entries.

In performing our search, we aligned our approach to the systematic literature review of Foss and Saebi (2017). Consequently, we searched the EBSCO database for academic articles published between 2000 and 2021, containing the key terms “policy-practice decoupling,” “means-ends decoupling,” “decoupling” AND “symbolic management,” and “decoupling” AND “institutional theory” in the title, subject, abstract, and/or keywords. In addition, we combined quotation marks with the Boolean operator “AND” to exclude irrelevant mentions based on grammatical coincidence. Table 2 illustrates the results from the four EBSCO database searches.

Stage 2: Exclusion and Inclusion Process

This first phase started by conducting our main search in the EBSCO database using the four search entries from stage 1.

Phase 1 of the Examination Process

The initial search led to identifying a total of 219 articles. This initial search of articles was admittedly broad. Therefore, we examined this first pool of articles to determine their relevance to decoupling research. The first phase of the examination process relied on an in-depth examination of the articles’ titles, abstracts, and keywords. This first phase led us to exclude 105 articles not sufficiently relevant for institutional decoupling. Thus, we retained 114 articles from this first phase.

Table 2

**An illustration of the employed search entries and their corresponding results from the EBSCO database**

Phase 2 of the Examination Process

Next, we reviewed all articles retained from phase 1 (n=114), with an in-depth reading of the full text. In this second phase of the examination process, we reviewed the articles’ full texts for employing theoretical frameworks that were primarily relevant to decoupling research. This phase 2 suggested that many authors discussed decoupling superficially but did not rely on any decoupling framework in their works. Accordingly, out of the 114 articles, we decided to exclude 54 articles for not being primarily based on decoupling theory. Table 3 summarises the three phases of the exclusion and inclusion proves.

Phase 3 of the Examination Process

The final sample yielded a total of 60 articles. To ensure that we have identified all the articles that have primarily employed a decoupling theoretical foundation, we solicited our review panel to identify any essential articles that might not have been omitted in the sample. The panel members identified no additional articles to be added to the sample. This independent panel assessment provides trustworthiness that our review comprehensively covers the current state of research on institutional decoupling.

Table 3

**Description of the exclusion and inclusion process**

Stage 3: Review Process

Phase 1: Thematic Analysis and Synthesis of Findings

The first phase of the reporting stage comprised a “thematic analysis” (Tranfield et al., 2003, p. 218). Then, following the procedures recommended by Cortez et al. (2021) and Snyder (2019) for performing a systematic literature review, we built a protocol for coding and structuring the key insights provided by the articles in our final sample. First, we created a set of codes to label the key elements from the articles which are most relevant to the aim of our paper. This process included the coding of the following properties: 1) the employed theoretical framework; 2) the key theoretical contributions; 3) the key conceptual insights; and 4) the main focus of the paper.

Second, we conducted a qualitative analysis (Strauss and Corbin, 1990) of our codes to synthesise the findings. Specifically, we conducted an axial coding process that involved grouping the large set of codes under broader theoretical categories we labelled as key conceptual insights. These key conceptual insights constitute the principal building blocks t structure our findings. Table 4 provides a synthesis of our findings. Guided by Cortez et al.’s (2021) methodological protocol to ensure the reliability and trustworthiness of our coding protocol, we solicited two review panel members again to review our coding process and results for validation.

To systematically report our findings, we started by synthesising research on institutional decoupling. Next, we categorised the primary forms of the phenomenon, shedding light on the main distinctions between the two forms of policy-practice and means-ends decoupling, and surfaced the fundamental tenets and the underlying conditions under which both forms prevail. In this respect, we categorised the different theoretical approaches and paradigms used by scholars to explore the two complementary decoupling forms. Further, we surfaced central themes and research questions from the literature. Specifically, we focused on the emergence of means-ends decoupling as an alternative explanation for the frequent failure of goal achievement among contemporary organizations. We identified two crucial gaps in the extant literature of means-ends decoupling and advanced two main insights contributing to filling these gaps.

Phase 2: Conclusion and Interpretation

In phase 2, we delineated the contributions of the conceptual insights for the current decoupling research and their role in framing potential solutions for contemporary organizations. We finally discussed the implications of our findings for research and practice and how they contribute to filling the identified gaps in the extant means-ends decoupling research. Moreover, we proposed potential avenues for future research to unravel the phenomenon of means-ends decoupling further.

Results

Early research on decoupling has distinguished between two different forms of the phenomenon; policy-practice decoupling and means-ends decoupling. This distinction has helped institutional researchers better understand the underlying consequences and decoupling conditions in the contemporary world (Bromley and Powell, 2012; Wijen, 2014). Policy-practice decoupling takes place when formal policies are not implemented and/or regularly violated. Instead, means-ends decoupling occurs when formal policies are implemented but tenuously link to intended outcomes. There is growing attention for studying the latter form of decoupling, which increasingly manifests in the contemporary world and has a significant impact on organizations (Bromley and Powell, 2012), given the continuing fragmentation and rationalization of the institutional environment and increasing pressures for a substantive implementation of adopted policies (Boli, 2006; Zucker, 1987; Frank and Meyer, 2002; Strathern, 2000; Power, 1997). In the following sections, we summarise our literature review on the two forms of the phenomenon, clarify the main distinctions between the two, and propose potential avenues for future research. Table 5 illustrates the main differences and distinctions between the two forms of decoupling.

Table 5

**Policy-practice versus means-ends decoupling**

Current State of Policy-Practice Decoupling

The “decoupling” term traditionally represents a mismatch between adopted policies and internal organizational practices. This form of the phenomenon refers to instances where the firm symbolically adopts policies. While tenuously monitored, evaluated, and implemented, the policies have little impact on the firm’s daily activities (Crilly et al., 2012; Fiss and Zajac, 2006; King and Lenox, 2000;). However, they help firms acquire legitimacy with minimal resources (Singh and Point, 2009). Early research on policy-practice decoupling suggests that adopted policies are often a weak predictor of daily work routines (Meyer and Rowan, 1977). Furthermore, policy–practice decoupling allows an organization to adopt multiple or even conflicting policies in response to external pressures without disrupting daily operations.

For example, Bussigel et al. (1986) show in their examination of medical schools the existence of inconsistent and incompatible organizational ends among which linkages are ambiguously articulated. Organizations that decouple policy from practice adopt various and often conflicting policies with little impact on daily operations. Similarly, several studies show that organizations adopt initiatives such as stock repurchase programs, total quality management programs, and long-term incentive plans for CEOs in response to institutional pressures, but that the implementation of these policies varies noticeably (Fiss and Zajac, 2006; Westphal and Zajac, 1994, 2001; Zajac and Westphal, 2004).

While early research on policy-practice decoupling substantially improved our understanding of why decoupling policy from practice occurs, it has done little to unravel the factors leading to decoupling means from ends prevailing in contemporary organizations. Table 6 provides sample studies from extant literature on policy-practice decoupling. The following section reviews the literature on means-ends decoupling and offers avenues for future research.

Table 6

**Summary of studies from the extant literature on policy-practice decoupling**

Current State of Means-ends Decoupling

Early institutional decoupling research focused on the gap between adopted policies and internal practices (Short and Toffel; 2010; Fiss and Zajac, 2006). Recently, scholars recognized that decoupling also occurs at the means-ends level if adopted policies or practices have a tenuous link to organizational outcomes (Bromley and Powell, 2012; Wijen, 2014).

Means-ends decoupling is more likely to surface in organizational settings with specific conditions: these settings are characterised by formal structures that have concrete organizational consequences, policies that are implemented and evaluated, and evolving work activities, but with little or no evidence supporting that the latter activities link to organizational goals and effectiveness. For example, internal constituencies often face significant structural pressures. Accordingly, they may pursue certain practices, although they recognise that these practices have limited utility and a tenuous link to internal goals. For example, certain organizational activities are implemented not because they link with organizational effectiveness but because formal operating systems such as evaluation and accounting push their establishment. As a result, they may have an unclear connection to organizational goals. The means-ends decoupling lens helps investigate “why organizations implement policies that tenuously link to organizational goals?”

Extant literature suggests means-ends decoupling to significantly impact organizational performance by causing “internal complexity” (Boxenbaum and Jonsson, 2017; Bromley and Powell, 2012), “endemic reforms” (Power, 1997), and “diversion of resources” (Sauder and Espeland, 2009). Wijen (2014) summarises the conditions under which the adoption of policies may not lead to achieving initially intended goals. The lack of field transparency caused by the compliance barriers of “causal complexity,” “behavioural invisibility,” and “practice multiplicity” leads policy’ regulators to elaborate a set of compliance inducements to ensure substantive compliance among organizational adopters. Complementing the compliance barriers, “setting rules”, “devising incentives”, and “transferring practices” constitute three compliance inducements for adopters.

Bromley and Powell’s (2012) and Wijen’s (2014) seminal works on means-ends decoupling advance understanding of this developing phenomenon. The highly increasing fragmentation and rationalization of the institutional environment in which contemporary organizations operate and the increasing pressures for a substantive implementation of adopted policies create favourable conditions for means-ends decoupling to prevail (Boli, 2006; Frank and Meyer, 2002; Strathern, 2000). While modern organizations operating in rationalised and fragmented environments are less likely to encounter policy-practice decoupling, they cope with decoupling at the level of means and ends. Under conditions of accountability and monitoring, means-ends decoupling provides an alternative explanation to organizations’ frequent failure to attain the envisaged goals for adopted policies, despite deploying significant resources and following explicit inducements. Table 7 provides a summary of studies from the extant literature on means-ends decoupling.

In the following section, we focus on the manifestation of means-ends decoupling in cooperative R&D projects, as they exemplary fulfil the criteria of an opaque field.

Means-ends Decoupling in Cooperative R&D Projects

Opacity exists in fields where observers have difficulty determining the specificities of prevailing practices, identifying causal relationships between adopted policies and organizational practices, and assessing the precise outcomes of policy implementation (Briscoe and Murphy, 2012; Bromley and Powell, 2012; Jiang and Bansal, 2003). It is challenging to observe whether policy-adopting organizations achieve a substantive compliance level. Wijen (2014) specifies that complex causal patterns, heterogeneous practices, and hardly visible behaviours result in three distinct compliance barriers: causal complexity, behavioural invisibility, and practice multiplicity. These compliance barriers in opaque fields influence policy-makers to elaborate policies in ways that incentivise adopters’ compliant behaviour. On the other side of the spectrum, three compliance inducements attenuate the compliance challenges organizations face in opaque fields.

Our comprehensive review of research on means-ends decoupling surfaces that we need to explore this emergent phenomenon in alternative settings. The research setting of many studies on decoupling deals with sustainability and socio-environmental standards (Crill et al., 2012; Haack et al., 2012; Lyon and Maxwell, 2011). However, this setting is not the only type of institutional field with opaque conditions where means-ends decoupling occurs. Given the scarcity of research that explores discrepancies between means and ends within settings beyond realms of corporate social responsibility highlights the importance of investigating this phenomenon within other opaque fields like public service (Frey et al., 2013), health care (Nigam and Ocasio, 2010), and climate change mitigation (Ansari et al., 2013). We focus on cooperative R&D projects, a field in which opacity renders goal achievement difficult for observers to measure and connect to deployed means, and practice multiplicity makes it hard for implementers to identify prevailing practices. These conditions favour the evolution of means-ends decoupling in cooperative R&D project settings. While this particular setting meets the criteria of an opaque field defined by early decoupling research, it has not received scholarly attention from institutional scholars. We address this gap by illustrating how cooperative R&D projects meet the criteria of an opaque field. Specifically, we detail the particularities of the field resulting in opacity. We also describe the compliance barriers and the set of inducements established by the field’s institutional actors to overcome barriers and ensure compliant behaviour among adopting organizations.

Table 7

**Summary of studies from the extant literature on means-ends decoupling**

Cooperative R&D projects fit the specificities of an opaque field, as compliance barriers render substantive compliance challenging for project partners and provoke, in turn, compliance inducements to attenuate compliance challenges and maximise goal achievement (Jiang and Bansal, 2003; Wijen, 2014). Figure 2 describes how compliance barriers and compliance inducements influence means-ends decoupling within cooperative R&D projects.

First, the compliance barrier of “causal complexity” evolves in environments in which multiple heterogeneous actors, factors, and effects interact (Espinosa and Walker, 2011; Levy and Lichtenstein, 2012). Causal complexity leads, in consequence, to uncertainty and ignorance about cause-effect relations and the precise nature of an institutional field (Davis et al., 2009). The uncertainty and ignorance, which causal complexity provokes, result in a lack of attention and knowledge about critical drivers of substantive compliance (Ocasio, 1997). Cooperative R&D projects exemplify this causal complexity which reins in opaque fields. A multitude of different actors constitutes R&D project consortiums. This diversity of stems from the need to engage highly experienced actors in hyper-narrowed specialisations.

Cooperative R&D projects often integrate actors from various disciplines, such as specialised startups in particular state-of-the-art technologies and leading research institutions with specific expertise (Giebe et al., 2006; Lundberg and Andresen, 2012). These actors’ diverse and multifaceted nature renders the comprehensive understanding of the cooperative R&D practices challenging. The “technical annex” that is developed for each specific work package guides implementers in cooperative R&D projects. The codifications and explicitly formulated rules that constitute the “technical annex” operate as a compliance inducement diminishing ambiguity, and uncertainty and fostering compliant behaviour among the project partners. Cooperative R&D projects’ designers act as institutional entrepreneurs, developing specific “technical annexes” that fit heterogeneous expectations in the consortium. These annexes’ particular guidelines establish the link between corporate action and adopted policies for involved partners.

Second, the compliance barrier of “behavioural invisibility” allows actors to conceal their non-compliant behaviour and avoid possible sanctions (Aravind and Christmann, 2011). Behavioural invisibility characterises institutional fields, in which actors maintain a low profile (Spar and La Mure, 2003), refrain from accepting external control (Howard et al., 2000), or locate in remote areas (O’Rourke, 2007). Therefore, behavioural invisibility risks decreasing the compliance motivation of adopters (Wijen, 2014). To overcome this motivation barrier, cooperative R&D projects developers signal that compliant partners receive material benefits, such as being selected for future projects or capturing value from the achieved innovations (Aarikka-Stenroos et al., 2017; Perkmann et al., 2013). These signals serve as compliance inducements to minimise the behavioural invisibility barrier and overcome low motivation among implementers. This way, R&D projects’ developers operate as institutional entrepreneurs creating ideal conditions for partnering organizations to comply, maximise goal achievement, and maintain a visible profile.

Third, the compliance barrier of “practice multiplicity” characterises institutional fields where actors face various divergent practices and heterogeneous routines. The multiplicity of practices makes it difficult for adopters to engage in compliant behaviour (Santos and Eisenhardt, 2009; Young, 2012). Moreover, it leads to ambiguity and a lack of knowledge about which one of the various coexisting practices results in substantive compliance and goal achievement. To overcome this compliance barrier, developers of cooperative R&D projects offer implementation options by enabling the transfer of best practices among project partners (Möller and Rajala, 2007; Raapersad, Quester, and Troshani, 2010; Perkmann et al., 2013). Thus, cooperative R&D projects establish ways to promote collaborative work and knowledge sharing among R&D partners, enabling capacity building and acquisition of necessary competencies for policy implementation (Santamaría et al., 2010). Overall, inducements reduce ambiguity and uncertainty arising from heterogeneous practices and surface the best practices that maximise the chances for compliant behaviour and goal achievement (Bromley et al., 2012).

Figure 2

**Means-ends decoupling within cooperative R&D projects**

As an interesting avenue for future research, our results encourage studying means-ends discrepancies in the alternative setting of cooperative R&D projects. Moreover, shedding light on such an overlooked opaque field can help better understand the factors and causes relevant to the imperfect achievement of adopted policies’ envisaged goals. The specificities of institutional fields and inherent conditions involve distinct lower-level mechanisms, shaping, in turn, the manifestation of macro-level means-ends decoupling (Vo et al.,, 2016).

Discussion

Many contemporary organizations which initially adopted policies symbolically have substantially increased their effort to implement policy goals (Short and Toffel; 2010; Tilcsik, 2010). This evolution results from the organizational need to respond adequately to the continuing rationalization and fragmentation of institutional environments and growing monitoring, control, and evaluation of various societal, governmental, and political bodies (Boli, 2006; Bromley and Powell, 2012). In this study, we identified gaps in the decoupling literature relevant to these fundamental changes. Based on a comprehensive review of extant literature, we develop a framework that delineates prevailing research questions for the decoupling field. Our framework builds on and further elaborates the seminal contributions of Bromley and Powell (2012) and Wijen (2014), which conceptualise means-ends decoupling as an alternative explanation for the frequent failure of organizations to achieve the intended goals of their adopted policies.

Our study makes three contributions to the decoupling field. First, our systematic literature review advances the traditional perspective on decoupling by pursuing the question of why means-ends decoupling occurs. We highlight and provide insights for this critical question relevant to organizations’ frequent failure to attain the envisaged goals for adopted policies in opaque fields (Bromley et al., 2012). Specifically, our review results show that a set of compliance barriers and compliance inducements, which are field dependent, lead to the manifestation of means-ends decoupling. This way, we provide further detail to the growing recognition of the importance of lower-level mechanisms in institutional research on decoupling phenomena, as indicated in Boxenbaum and Jonsson’s (2017) conceptual work (see also Vo et al., 2016).,

The second contribution of this study relates to the ongoing debate on opaque fields. Our review of prominent decoupling research reveals the need for exploring the nascent phenomenon of means-ends decoupling in contexts beyond sustainability and socio-environmental standards, which have primarily dominated early decoupling research (Crilly et al., 2012; Haack et al., 2012). We have focused on cooperative R&D projects as an alternative technological setting for studying means-ends decoupling. Our study shows that cooperative R&D projects represent an opaque field as defined by early institutional work (Briscoe and Murphy, 2012; Bromley and Powell, 2012; Jiang and Bansal, 2003). We show specifically how this institutional field is characterised by a set of compliance barriers, which, in turn, are attenuated by a set of inducements that ensure substantive compliance and maximise goal achievement for adopting organizations. We argue that field opacity in the context of cooperative R&D projects renders goal attainment challenging to assess and to link to the deployed means causally. Our study also shows that cooperative R&D projects’ designers act as institutional entrepreneurs offering compliance inducements to ensure compliant behaviour and goal achievement among partnering organizations.

Finally, the third contribution of this study is to delineate the conditions under which deployed means and envisaged ends maintain distinct trajectories. This finding enriches the cooperative R&D projects literature, as identifying the means-ends discrepancies in cooperative R&D projects help explain one of the persisting questions faced by R&D funding schemes in industrialised economies. For example, in policymaking, the research directorate of the European Commission struggles to understand why many funded R&D projects fall short of attaining their intended outcomes, despite the significant project funding means deployed by the Commission (Giebe et al., 2006; Lundberg and Andresen, 2012). Another example of implications in the entrepreneurship field is to help early-stage startups comply with ethical standards and norms in their respective industry. Press articles have recently raised the issue of startups not complying with country laws on work conditions for new graduates (excessive working hours is one). Even when there is a genuine intention to comply with the laws, the engaging nature of working for a fast-growing startup and the usual lack of resources may prevent truly committed startup founders from complying. Thus, industry unions and policy-makers may reflect on inducement measures that specifically address this type of organizations. Our findings offer some directions on how to mitigate compliance barriers through engaging in inducements. In sum, our study contributes to further enlightening the means-ends decoupling black box, develops promising avenues for future research, and provides a theoretically grounded baseline for theorising the emergent means-ends decoupling phenomenon. Figure 3 provides a summary of our key findings and contributions.

Future empirical research can refer to our framework to investigate the means-ends decoupling mechanisms in opaque fields. There are many opaque fields in which means-ends decoupling is likely to prevail in the future. However, while the decoupling practice has substantial implications for organizational performance, it has received little attention from institutional scholars. In this regard, our review results and theoretical framework provides a basis for examining decoupling practices in diverse institutional settings and show how specific settings meet the criteria of opaque fields where means-ends decoupling is likely to occur with all the possible consequences for the organizational and common interest.

Figure 3

**Opening the means-ends decoupling black box**

Conclusion

Our study was motivated by the lack of substantial academic research on the means-ends decoupling phenomenon. Our systematic literature review responds to the need for practically relevant theory around important contemporary topics (Corley and Gioia, 2011). We specifically find that field-dependent compliance barriers and compliance inducements involve in means-ends decoupling, examine their influence in decoupling practices in a cooperative R&D project, and delineate the conditions under which deployed means and envisaged ends maintain distinct trajectories. Our study results provide comprehensive, theoretically grounded insights that help investigate important governance issues for contemporary organizations regarding means-ends decoupling. We specifically advance the institutional perspective on the phenomenon of means-ends, which builds upon and complements the seminal contributions of policy-practice decoupling research. By pursuing the question of why means-ends decoupling occurs, our study findings summarise central viewpoints in the literature for the puzzle of organizational failure to achieve intended policy goals despite investing substantial resources and effort. In sum, our review contributes to building a comprehensive theoretical framework within which the significant phenomenon of institutional decoupling can be better understood and addressed by management research and practice.

Abstract

Résumé

Resumen