Abstract
Alignment of organizations with external imperatives is seen as a sine qua non of proper organizing and strategizing by many fit and complexity scholars. Any deviation from this management mantra engenders organizational decline and, ultimately, mortality. We put this axiomatic principle under empirical scrutiny and use the law of requisite variety as our organizing principle to do so. The law is an iconic cornerstone of this matching contingency logic and it has served to legitimize a wide range of fit decisions in, e.g., leadership, organizational learning and corporate governance. Inspired by organizational vignettes inhabiting antithetical complexity regimes, we introduce a novel concept, which we label ‘agentic misfit’. In this way, we deconstruct deterministic assumptions related to environmental fittingness, we challenge teleological orientations in the fit literature, and we flesh out the viability of non-matching human agency amid complexity.
Keywords
Introduction
Complexity and human agency are inextricably intertwined with the essence of management studies. On the one hand, complexity is the archetypical means to describe structural properties in and around organizations. Our markets, organizational arrangements, networks and, overall, ‘environments’ are characterized as complex as ever by management scholars (Boisot & McKelvey, 2010). On the other hand, management scholarship investigates the ability of humans to organize structures and processes through purposeful action. Human agency and its usefulness are what management research largely studies (Greenwood & Suddaby, 2006). Therefore, works couched in a process tradition (e.g. Garud, Gehman, & Kumaraswamy, 2011), simple rules in strategizing (e.g. Sull & Eisenhardt, 2015), institutional perspectives (e.g. Greenwood & Suddaby, 2006), microfoundations (e.g. Barney & Felin, 2013) or methodological approaches (e.g. Lorino, Tricard, & Clot, 2011) have rightfully made the conceptual connection between human agency and complexity.
However, ‘in many cases attention is directed towards systemic qualities in a manner that implies little scope for significant human intervention’ (Child & Rodrigues, 2011, p. 803). As the latter authors note, individuals’ power, intentions, orientation or cognition are often neglected or considered irrational to be applied in complexity settings, thus begging an answer on who initiates commercial, political and organizing action therein (Anderson, 1999). This prioritization of the ‘systemic’ rather than the ‘agentic’ leaves complexity theories with a lacuna of studies ‘on how purposive action taken by key organizational actors may buffer, and even to some extent shape, external systems’ (Child & Rodrigues, 2011, p. 804). Rather, complexity studies most often see the external environment either (1) as a constraint to human action or (2) as the impetus for adaptive action. Therefore, an ‘outside-in’ perspective has often led to an ‘action void’ or an overly adaptationist view of managing amid complex regimes (Child & Rodrigues, 2011). Especially in fit/matching studies, the centrality of environmental determinism and the causal efficacy of adaptation are paramount (Poulis & Poulis, 2016).
We problematize those perspectives by empirically connecting four themes: the law of requisite variety (LRV; Ashby, 1956) which is articulated as the need for organizations to internally match external variety/complexity (Boisot & McKelvey, 2010). The law acts as the ancestral cornerstone of fit-as-congruence studies, which associate environmental matching with enhanced performance. We claim that this fit logic which the LRV has spawned is the main doctrine in complexity realms, which promote the exclusive efficacy of adaptation as the prudent organizational choice. Nevertheless, we claim that fittingness in complex settings is theoretically mono-dimensional. Those settings are conducive to multiple manifestations of human agency, which enable non-adaptive possibilities for action, too.
We negate neither the conceptual appeal, parsimonious articulation and theorizing potential of the LRV nor fit’s zones of applicability. However, empirical substantiation is seriously lacking before the law can be used as a doctrine. Motivated by this observation, we identified a sharp contrast between LRV’s premises and our fieldwork experience: in the course of a research project, we identified surviving organizations of misaligned (LRV-disconfirming) internal/external complexity; an observation which we understand as the result of agentic misfit. The latter is a novel empirical concept that challenges entrenched assumptions in fit and complexity theorizing and constitutes the nucleus of our study.
Therefore, we contribute in three ways: first, we showcase that uncritically embracing borrowed concepts such as the LRV entails shortcomings that call for empirical scrutiny. Thus, we refine an iconic concept that is not empirically supported and largely misconceived. Second, we challenge the centrality of fit-as-congruence in organizational scholarship and its concomitant adaptive imperative as the exclusive means for superior performance. Instead, through agentic misfit, we promote human agents’ reflective judgement as the medium that shapes (non)matching action. In turn, by questioning the one right matching contingency that is the conceptual cornerstone of the LRV, we also challenge normative assumptions related to non-survival prospects of misfit organizations. Third, by fleshing out agentic misfit, we empirically document practically-laden and viable ways of dealing with complexity (‘enactment through disruption’ and ‘quiescence through prescience’) that the literature has not captured yet. Contrary to fit studies, these dealings elucidate that variably complex regimes induce possibilities for action that can be reasonably misaligned with external imperatives. Thus, we make an important claim that human agency matters in complexity studies, contrary to its undertheorized status or its treatment as an irrational, non-viable pursuit. 1
We structure the paper as follows: we critically approach the law’s tenets and argue that fit is LRV’s ‘translation’ into management research. We discuss main complexity themes, which emulate this matching logic and we introduce agentic misfit as an alternative understanding of organizational acts. Thus, we express our disagreement with the one right adaptationist approach that fit and complexity scholars promote and we set empirical boundaries to LRV. We explain our methodology by building upon Tsoukas and Hatch’s (2001) second-order complexity and we present findings from eight misfit cases inhabiting sectors of high (maritime logistics) and low (drug retailing) complexity. We conclude with implications for organizational theorizing.
Requisite Variety and Fit in Organizational Scholarship
The LRV (Ashby, 1956) is a quintessential foundation of complexity thinking and has an impact across domains as wide as biology, engineering, or information systems. In management studies, scholars use it (1) to stress that mortality is engendered in organizations which do not possess sufficient internal variety (2) to support empirical findings, build hypotheses or justify methodological choices (see Poulis & Poulis, 2016). According to the former perspective, an organization needs to possess properties which are as complex as the properties of the system against which it attempts to cope with. Otherwise, organizations will decline and perish.
Specifically, management research utilizes requisite variety in two normative ways: (1) to stress the need for matching environmental contingencies through adaptation (Boisot & McKelvey, 2010); (2) linking this with an organization’s survival prospects, e.g. through the mediating effect of gaining competitive advantage (Lepak & Snell, 1999) or avoiding managerial inertia (Ferrier, 2001). Therefore, according to LRV interpreters, non-matching implies: (1) either internal arrangements are more complex than external contingencies (thus, wasting strategic capability and accumulating useless slack) or (2) internal arrangements are less complex than external contingencies (thus, exposing the organization to risk from volatility and turbulence). The need to respond to complexity through requisite internal complexity is ‘textbook knowledge’ and a central element of management research (Schneider, Wickert, & Marti, 2017, p. 199). Thus, it is of particular interest to fit scholars due to the ‘poor’ survival prospects of non-matching organizational configurations. 2
Fit as the conceptual offspring of the LRV
The modern reincarnation of the LRV is fit-as-congruence, 3 which occupies a celebrated position in management studies and is understood as the alignment of internal arrangements with environmental imperatives. Lack of fit is seen as the antecedent of failure (Carmeli, Gelbard, & Gefen, 2010) and the ‘road to disaster’ (Heracleous & Werres, 2016). It leads organizations into unviable zones (Godsiff, 2010) and results to ‘inefficiencies, substandard performance, and the potential death of the organization’ (Pérez-Nordtvedt, Payne, Short, & Kedia, 2008, p. 785). Consequently, in order to avoid collapse, organizations’ goal should match internal/external complexity, with the latter determining the ‘fitting’ degree for the former (Lynn, 2005). This uni-directional view dominates the management discourse and legitimizes environmental determinism in organizational action. Moreover, echoing fit-related remarks, complexity scholars also note that ‘a system must possess complexity equal to that of its environment in order to function effectively’ and achieve fitness (Uhl-Bien, Marion, & McKelvey, 2007, p. 301). Thus, the relevance of fit is paramount across many themes that constitute management scholarship.
After a burst in the 1960s to 1980s and a decline thereafter, fit theories are now revitalized (Van de Ven, Ganco, & Hinings, 2013). This is logical since the concept permeates management studies’ legacy. Fit is the common denominator across themes as wide as the configuration tradition, contingency theory, industrial organization, population ecology, adaptive capacity, dynamic capabilities, McKinsey’s 7-S, etc. (Carmeli et al., 2010; Volberda, van der Weerdt, Verwaal, Stienstra, & Verdu, 2012). Essentially, fit-as-congruence implicitly reflects LRV’s appropriation by management scholars. Its orthodoxy is rarely challenged, which has led fit authors to claim that higher performance emanates ‘only to the extent that there is fit between the environmental imperatives and the strategy being deployed’ (Katsikeas, Samiee, & Theodosiou, 2006, p. 867).
We disagree with such dogmatism without negating zones of LRV applicability and, hence, fit. In turn, we chose the LRV to promote our disagreement given the law’s relevance to fit studies. Problematizing its core thesis, we pose a question: given a perceived external complexity, should managers configure their internal environment in ways that fit the said complexity? Viable cases where misfit is noted imply that the LRV is disconfirmed. Thus, boundary conditions can be sketched. To the best of our knowledge, this is the first empirical study that explores this conditioning, which is particularly interesting given the LRV’s association with core management themes.
Complexity Science(s)
Complexity studies is not a homogeneous field but a heterogeneous amalgamation of foci and themes (Burnes, 2005). For example, complexity may be understood as the exponential consequences that sensitivity to initial conditions may generate (see chaos theory; Thietart & Forgues, 1995) or as the co-existence of multiple logics in a field (see institutional complexity; Delbridge & Edwards, 2013). This heterogeneity is inevitable given the ontological and epistemological diversity associated with complexity research (Maguire, McKelvey, Mirabeau, & Öztas, 2006). As such, the resultant multi-vocality might be seen as healthy pluralism. However, a striking inconclusiveness has also made complexity to be seen as a non-scientific exercise (Poulis & Poulis, 2016).
Given this inconclusiveness, we understand complexity in parsimonious terms and adopt the definition in Maguire et al. (2006, p. 166): ‘A complex system is a whole comprised of a large number of parts, each of which behaves according to some rule or force that relates it interactively to other parts.’ Therefore, complexity comprises variety of and interactions between elements including human (e.g. managers) and non-human (e.g. departments) actors or socially constructed (e.g. institutions) and naturally occurring (e.g. climate) forces. The higher the variety and the interactions, the more complex the industry or the more complex the organization and vice versa. Consequently, (lack of) fit implies that there is (mis)alignment between those noted external and internal complexity levels.
This parsimonious understanding lends itself to a human agency framing that may revitalize the fragmented complexity discourse and implies a focus on organizational decision-makers’ agency. We certainly acknowledge the value of sub-fields such as chaos theory, complex adaptive systems or dissipative structures, which are interested in complexity as a general property of structures (Maguire et al., 2006). However, traditions of complexity therein largely focus on how wider structures self-organize and emerge following agents’ interactions at lower levels of analysis and/or through inputs from the external environment (Maguire et al., 2006; Chiles, Meyer, & Hench, 2004). Therefore, the scholarly effort is to formalise generalizable rules and patterns for whole systems. This focus is fair enough. However, it offers limited regard about the organizations’ agency 4 in the making of observed patterns (Child & Rodrigues, 2011; Heylighen, Cilliers, & Gershenson, 2006).
We acknowledge that several complexity studies explore agentic nuances (e.g. intentionality in entrepreneurial ecosystems; Roundy, Bradshaw, & Brockman, 2018). However, most complexity fields demonstrate a Newtonian quest for generalization and prediction of a system’s orderly or emergent arrangements without a commensurate effort to elucidate the role of human agency in these arrangements. Given complexity studies’ origins in cybernetics, this lack of attention to human agency is unsurprising. Cybernetics seeks to identify general laws that govern systems (Schneider et al., 2017) and not to explain the situated specificities of their becoming. That is a fair focus for cybernetics. Nevertheless, organizing-wise, focusing on aggregate conceptualizations that aim to understand system-level developments impedes clarifying agentic underpinnings. However, organizational research is those underpinnings to a large extent. After all, complexity is not a system’s objective property irrespective of the agent in it or observer of it. Rather, ‘it is determined by the position, perspective and purpose of those who seek to describe it’ (Maguire et al., 2006, p. 166). Therefore, a neglect of those agentic themes not only opposes the essence of management scholarship’s orientation and scope but also ignores a fundamental premise of complexity theories.
Given these limitations of ‘traditional’ complexity-cum-management studies to elucidate human agency, we adopt Tsoukas and Hatch’s (2001) second-order complexity (see Methodology). Second-order complexity allows us to surmise the complexity of an organization or of an external environment and thus, identify fit/misfit decisions. However, we do so through organizational decision-makers’ eyes (hence, the human agency framing) and not through an ex-post observation of any objective systemic property (e.g. the self-organization of an industry) or of a systemic outcome (e.g. the eventual order of an organizational arrangement). Moreover, we differentiate ourselves from a typical approach related to the ‘how’ of agency in complexity theories: according to our working definition, interactions across elements induce complexity. However, as noted earlier, in the complexity literature, what stands out as pertinent in the course of those interactions is the constraint of responding, i.e. a systemic actor is forced to ‘adjust its behaviour to that of other agents’ (Burnes, 2005, pp. 78–9).
Therefore, first, systemic properties such as order or emergent self-organization assume higher analytical and hence, theorizing importance in several complexity studies. Consequently, the role of agency assumes an, at best, secondary role. Second, even when agentic contributions are acknowledged, they are mostly limited to an adaptationist approach, which emulates the matching logic of the LRV and fit-as-congruence. In the same way that natural subjects do, organizational systems continuously adapt to one another and their environment through increasing internal complexity (Uhl-Bien et al., 2007). Otherwise, they become obsolete and die. This Darwinian adaptation/survival nexus that is unreflectively borrowed from the sciences is so prevalent in complexity studies that it constitutes ‘the underpinning nomological network of the entire field’ (Poulis & Poulis, 2016, p. 518). However, this ignores the possibility that organizations may strive to reproduce themselves (Poulis & Poulis, 2016), enact complexity through their choices (Goh & Pentland, 2018) or simply ignore others without being committed to an adaptive imperative (Heylighen et al., 2006). Consequently, the contemporaneity of the adaptive task masks not only non-responsive possibilities for action but also agents’ past experiences and future orientations that may make contemporary fit an unwelcome compromise.
Introducing Agentic Misfit
Given this popular understanding of complexity as a property of structures that constrains action, we sought instead to theorize the value of a ‘miniscule’ focus on ‘agentic misfit’, which we define as decision-makers’ deliberate choice to arrange their organizations’ internal environment in ways that are not congruent with the external one. The literature so far mono-dimensionally considers such instances of misfit as disastrous (Heracleous & Werres, 2016) and as indicative of managerial inability due to high costs (Strong & Volkoff, 2010), poor team outcomes (Maruping & Agarwal, 2004), efficiency loss (Pérez-Nordtvedt et al., 2008), or low trust and high conflict (Burton, Lauridsen, & Obel, 2002). In turn, such unfortunate instances of misfit ought to be rectified through more adaptation (Sia & Soh, 2007).
For us though, agentic misfit is not an act of irrationality or inability. It is not the result of poor managerial judgement, myopic environmental scanning techniques, or substandard configurational abilities. Rather, it is a conscious managerial choice and the result of a reflective modus vivendi, which engenders existential concerns and thoughtful self-awareness. Hence, it is ‘agentic’. Even more so, it is ‘agentic’ because misfit is deliberately chosen against a visualized pool of fitting possibilities. Instead of promoting emulation, conformity, legitimacy-seeking, compliance or compromise as archetypical instantiations of fit, we show that agentic misfit prioritizes entrepreneurial flexibility and stems from aspirational ethics. It solidifies the authenticity of one’s legacy against a volatile world. Thus, notwithstanding other fitting options, it epitomizes practical wisdom against complex and ordered regimes.
For reasons of balance, we emphasize that fit-as-congruence is a rational choice that is expected in many contexts. Thus, we do not seek to portray agentic misfit as a canonical or inherently prudent outcome of decision-making. In fact, it is a bold and risky choice, which can be logically expected in certain contexts and by certain organizations. For example, as we empirically showcase, entrepreneurial, aspirational or innovation-driven organizations in stagnant industry contexts may be prone to agentic misfit (Koch, Wenzel, Senf, & Maibier, 2018; Eghenter, 2018). On the contrary, organizations oriented towards legitimacy-seeking by external audiences (Patala, Korpivaara, Jalkala, Kuitunen, & Soppe, 2019) or ones infused with mimetic tendencies (Lieberman & Asaba, 2006) may not be ideal candidates for misfit actions. Moreover, organizations whose identity is based on conservative norms or who act as symbolic gatekeepers of traditions in volatile industries (Raynard, Kodeih, Greenwood, 2019) may exhibit agentic misfit, too. On the contrary, organizations which occupy a central, beneficial position in a stagnant industry (McKague, Zietsma, & Oliver, 2015), or organizations which lead structural changes in volatile sectors (Kalpokaite & Radivojevic, 2019), may be better off with a fit approach. Overall, such a balanced approach is in line with requisite advances in management scholarship. For example, one may consider institutional entrepreneurs (Greenwood & Suddaby, 2006) or social movements (de Bakker, den Hond, King, & Weber, 2013) as opposed to bureaucratic organizations (e.g. state-owned enterprises; Chiu, 2006). One may also juxtapose creative organizations (e.g. a theatre or an advertising agency; Jones, Svejenova, Pedersen, & Townley, 2016) against safety-driven and high-reliability entities (e.g. a nuclear plant or the army; Leveson, Dulac, Marais, & Carroll, 2009) to realize that both fit and misfit may be promoted depending on focal contexts and idiosyncratic circumstances.
Therefore, agentic misfit is not an ex-ante mechanistic choice that is arbitrarily taken by managers. It is not a monolithic decision of one or a team of managers without prior consultation with external agents or without reflection upon resources and constraints. Apparently, such a simplistic representation would ignore emergent phenomena, conflicts, tensions, contextual contingencies etc. (Delbridge, 2007; Kornberger, Leixnering, & Meyer, 2019). If we ignore these nuances, we also contradict ourselves in terms of what we critiqued previously, i.e. the lack of focus on agency in complexity studies. Thus, as we demonstrate, agentic misfit takes place iteratively and following reflection upon multiple temporalities of action. It is realized following assessment of fitting or less fitting alternatives and amalgamates into something concrete only after a conscious evaluation of the nuances associated with such a bold decision.
To the best of our knowledge, this is the first empirical concept across the complexity and fit literatures that explicitly captures purposeful acts of internal misalignment as viable modes of organizing and strategizing. Consequently, it represents a drastic departure from conventional wisdom which promotes calculative intentionality towards optimal matching. It equips our theorizing arsenal with an enhanced understanding that lack of congruence is not an ‘accident’. Rather, it can be a fortunate instantiation of prudence. In this way, we open up a wide realm of research opportunities to understand its performance consequences and the conditions that enable its deployment in organizations.
Thus, agentic misfit (1) challenges fit as a monolithic antecedent to higher performance, (2) questions the term ‘adaptive’ in complexity definitions and (3) conditions the LRV. It indicates that (mis)fit is not something that externally happens to organizations. Rather it is a choice from within, which engulfs past experiences and future aspirations into present concerns. As such, it implies that the external environment is not the habitually constraining impetus of ‘appropriate’ organizing and strategizing. Rather, internal features (e.g. agents’ ethos, experiences, teleological orientations) may also drive meaningful action. Even if the external environment is utilized to visualize alternative possibilities, it does not necessarily induce adaptive behaviours intra-organizationally. Rather, agentic misfit leads to variable and viable (in)action.
This non-matching manifestation of human agency is also a segue that moves us away from the core focus of complexity studies, i.e. the collective structures that complexity generates. Such a system-level focus has generated many insights but has also led to agentic foundations being neglected or ill-conceived. We assert though that the role of minutiae in collective outcomes should be a concern of complexity scholarship. Especially in archetypical social aggregation theories such as complexity, disaggregating explanations from concerns such as agentic misfit essentially ‘obfuscates explanation by hiding the actual mechanisms, processes, and actors that lead to the emergent outcome’ (Barney & Felin, 2013, p. 147).
Fragmented yet insightful agentic perspectives show that organizations can be proactive contextual shapers or consciously depart from structural arrangements and survive perfectly well. Organizations may morph their environments (Greenwood & Suddaby, 2006) or consciously adopt no response to external contingencies (Poulis & Poulis, 2016). Quiescence is a conscious choice that reflects moderation and restraint coupled with a deep knowledge of competitive dynamics amid conflicting institutional demands (Pache & Santos, 2010). Organizations may even choose to silently disappear, should this serve strategic concerns or address mutating externalities (Greenwood & Suddaby, 2006). These choices also challenge how scholars portray complexity itself. Many organizations (precisely, authors writing about them) do not perceive complexity as a desirable opportunity, nor are they willing to embrace it. Rather, they perceive it as something detrimental that must be matched to avoid decline.
We acknowledge that matching is a legitimate orientation under both stable and changing conditions. Barriers to this pursuit such as bounded rationality or incentive conflict need to be often tackled through requisite configurations (Gulati, Lawrence, & Puranam, 2005; Weigelt & Sarkar, 2012). However, we do not subscribe to the aetiology of increasing internal complexity as the exclusive need to do so. Rather, actors may foresee a possibility and mobilize resources that may enact it into reality or simply consolidate apraxia towards organizational reproduction. Therefore, a wide array of options is open: the external environment may be actively resisted, ignored through quiescence or reenacted through mobilizing resources and due to organizational identity and self-identification concerns (Kodeih & Greenwood, 2014). This discussion allows us to reframe a major management dualism: Is adaptation and, in extent, organizing and strategizing a result of environmental determinism? Alternatively put, should external complexity be ignored or circumvented as a result of agentic misfit?
Methodology
We employ eight purposefully selected cases of agentic misfit combining instruments such as interviews, documentation and observation (Eisenhardt & Graebner, 2007). Cases include four organizations from the shipping industry as archetypical of high external complexity and four cases from the drug retailing sector as representing low external complexity (all in Greece). Based on an iterative process through secondary data, pilot interviews and discussions with peripheral informants (Poulis, Poulis, & Plakoyiannaki, 2013), we identified primary cases (Table 1).
Sources of data.
Sources of data
Documentation (e.g. board minutes in shipping cases and promotional material in drug retailing cases) and direct observation helped us contextualize interview responses, identify relevant themes for exploration, and cross-check accuracy of primary responses. For example, observation in shipping cases confirmed the leader’s centrality or the prevalence of the organization’s legacy (e.g. formalization of seating arrangements during board meetings, or old photographs of founders/owners as material symbols on office walls).
Initially, we conducted interviews with peripheral informants and pilot respondents: First, all interviewees agreed on the complexity of the two sectors. This enabled the identification of misfit cases, i.e. shipping firms with low (SH1, SH2, SH3, SH4) (Table 2) and drug retailing firms with high (DR1, DR2, DR3, DR4) (Table 3) internal complexity. Figure 1 showcases this complexity typology which includes instances of both fit and misfit. All quadrants reflect legitimate goals to pursue and their performance consequence (survival) does not mono-dimensionally emulate the LRV doctrine of the one right matching contingency (see Discussion).
Indicative quotes – shipping.
Indicative quotes - drug retailing.

Sampled cases.
Second, respondents enabled identification of thematic areas for subsequent interviews. For example, peripheral informants helped us understand the complex framework governing shipping (high external) and the simple organizational setup and family-centric ethos (low internal) of traditional shipping firms. They pinpointed the protective framework in drug retailing and the new generation of pharmacists who foresaw additional possibilities by moving beyond it. Therefore, we organized primary interviews accordingly.
Third, preliminary interviews pinpointed paradigmatic cases for further inquiry:
- Shipping-wise, respondents signposted firms, which follow simple rules through low internal complexity (Sull & Eisenhardt, 2015). Access was enabled through those early respondents and by the fact that the Greek managed fleet is the world’s largest. Hence, several relevant firms could be readily identified. 5
- Drug retailing-wise, we were informed of pharmacists who pursued business studies and collaborated with cosmetics firms with a bold objective: to disrupt a stagnant industry. Hence, a new generation of drug professionals became well-versed in advanced sales methods and marketing techniques.
Selection of cases was not dictated by a quest for replication or representativeness. Rather, we selected paradigmatic misfit cases with preliminary informants helping us in identifying and accessing those cases. Therefore, a form of selection bias was inevitable since the non-matching organization was our pre-selected empirical unit. This focus was desirable since the purpose was neither to select LRV-confirming organizations nor to achieve sample heterogeneity. ‘Matching’ cases are arguably plenty and certainly well represented in the literature. Therefore, any such focus would have a reduced marginal utility.
A main selection criterion was the organizations’ survival record. In order to safeguard that misaligned practices are not detrimental, selected cases demonstrate at least ten years of market presence. All four shipping firms operate for more than 50 years while sampled pharmacies operate for 20 to 30 years. Across both settings, we interviewed upper echelons respondents since others might not be able to reflect upon complexity. These primary interviews resulted to transcripts of 314 pages and included (1) the CEO/managing director and departmental directors in shipping firms and (2) pharmacy owners and their store managers, i.e. their immediate subordinates. In total, we conducted 14 interviews in shipping cases and 8 interviews in drug retailers (Table 1). Our organizing principle was the generic distinction between external and internal complexity and we sought to explain how and why each case’s characteristics induce agentic misfit. Similar cross-case findings were identified, which was expected following our aforementioned sampling strategy.
Capturing complexity
Given the temporal stability of organizational fields, capturing complexity is highly challenging (Tsoukas & Dooley, 2011). To tackle this, we employ Tsoukas and Hatch’s (2001) second-order complexity, i.e. we make sense of complexity through decision-makers’ voices and ‘ask managers how they see their organization’s environment and the challenges it poses’ (Schneider et al., 2017, p. 201). Focusing on the perceptions of those who cope with complexity is chosen because managers shape the phenomena we study. We treat decision-makers as reflective participants whose judgement mediates the relationship between actual choice and the objective structures they connect with. After all, for an abstract aggregation as complexity then ‘every organization perceives its own distinct environment’ as situated in a relevant system (Schneider et al., 2017, p. 184).
We used the two features in our working definition of complexity as our interview guides: the diversity and interactions between agents (Maguire et al., 2006). Despite the numerous meanings associated with ‘complexity’, ‘most definitions attribute its emergence to combinations or interactions among heterogeneous elements’ (Garud et al., 2011, p. 738). Therefore, ‘complexity results from the interaction between the components of a system’ (Cilliers, 1998, p. 2) and a market or an organization is complex because ‘they are diverse and made up of multiple interconnected elements’ (Chiva, Grandio, & Alegre, 2010, p. 119). Thus, in line with Wright and Nyberg (2017), we coded data by having two sensitizing principles in mind: variety is the foundational platform of capturing complexity and indicates the number of elements in an organization’s internal or external environment. It answers the question of ‘how many’ (e.g. employees or departments for internal variety or suppliers and customers for external variety). Interactions within or across entities ‘upgrade’ variety to complexity. Externally, within-interactions mean alliances between competitors or intra-organizationally, or how resources are interlinked through cross-functional teams (Schneider et al., 2017). Externally, across-interactions mean how different institutions or legal frameworks govern the rate of change in a sector or intra-organizationally, or the extent of collaborative complexity with partners (Schneider et al., 2017). The higher the variety and the interactions within and across entities, the more complex the internal or external environment. 6
Having achieved this portrayal for both internal and external complexity, we then surmised whether there is a matching empirical tendency or not. Capturing complexity this way avoids the utopian pursuit and the pitfalls of any measurable match, i.e. we did not ask interviewees to reify or quantify complexity for matching purposes. We were only interested in whether there is a perceived mismatch between internal/external complexity. This was identified in both sectors through the meanings that organizational respondents and peripheral/pilot informants ascribed to their experiences. Their views helped us position sectors and cases across a bipolar empirical tendency: high or low complexity. In turn, we mapped cases in misaligned configurations of high (low) internal / low (high) external environments (Figure 1).
Presentation of Findings
We first demonstrate the high external complexity of shipping and low external complexity of drug retailing. This is the initial springboard to then showcase that antithetical intra-organizational arrangements take place due to idiosyncratic pursuits of selected cases. We illustrate this by using indicative quotes, which may not be exhaustive but, nevertheless, they (1) illustrate the imbalance between internal/external complexity and (2) shed light on enabling conditions for those non-matching configurations.
High external complexity in shipping
Shipping is a highly complex industry with a plethora of interacting players punctuating the field (Lützhöft, Grech, & Porathe, 2011). This is logical given that 90% of world trade takes place by sea. Shipping organizations act against a complex and uncertain environment due to numerous events ranging from regulatory change related to safety and the environment, infrastructural diversity in the world’s ports, and adverse weather, up to numerous ship sizes and changing patterns of trade and sea routes (Justice, Bhaskar, Pateman, Cain, & Cahoon, 2016). Thus, the sector is characterized as an unpredictable, volatile, cyclical and international environment (Goulielmos, 2002) where a large number of agents interact in obscure, non-linear and intractable patterns (Caschili & Medda, 2012). What makes our chosen context even more complex is that Greek-owned shipping is the largest in the world (circa 20% of the world’s tonnage).
Low internal complexity in shipping: Quiescence through prescience
Our shipping cases are archetypical examples, which disregard the complexity they encounter. They do so not because they do not acknowledge it. Rather, drawn by their past experiences and accumulated practical wisdom, they disregard it because of its overwhelming nature. This is a conscious managerial choice, which comes from prescience and a copious understanding that quiescence should be exercised in such a regime. As long as a comfortable market position is secured, then maintaining the status quo is a priority against a complex context.
Traditional ship owners exist as they were 100 years ago and they will still exist after 100 years in the future, exactly as they are now. They do not change their ethos, practices and ideas. We are one of them. (finance manager in SH1)
A mindset of preservation is associated with an owner who takes decisions. A certain commitment to the status quo is aggravated by his/her past successes, his/her centrality in securing cargo (i.e. the revenue stream for shipping companies) and a long-standing, traditional way of doing things that can be traced back to the industry’s family-centred background.
My subordinates cannot help me; they do not have the relevant information and ability to do that. This is a one-man-show in exactly the same way it used to be since the beginning of the industry. (CEO in SH3)
Below, we detail manifestations of low complexity. As can be surmised, all pinpoint to an inherent simplicity, which constitutes a conscious choice irrespective of the turbulence of the sector. For example, despite the huge shortage of competent seafarers, which has given rise to multiple nationalities onboard, these firms insist on a traditional way.
Our ships carry the Greek flag. This means that Offi|cers on the ship must be Greek – without exception. . . it is carried forward as a legacy from the founders of the company and it has to be respected. (technical manager in SH4)
The traditional nature of the sector, a legacy from the past that continuously intrudes into the present, is a staple element of decision-making and organizational arrangements that is never ignored.
Our organizational structure and decision-making processes are quite simplistic. Departments are and remain separate, each one specializing in what they do. . . we are a traditional ship owning company; that is our mentality and this mentality is a standing element in shipping. (chartering manager in SH1)
This iterative leverage from the past permeates all aspects of operations including human resource management practices, modes of decision-making, susceptibility to innovation, etc.
There is no external recruitment at any level of the hierarchy unless an emergent situation calls for it. We employ our seamen and our office employees when they are young and we promote them within the organization. All directors in the company have been working here for more than 15 or 20 years. . . (CEO in SH3)
Moreover, this legacy of keeping things as they are has served the company well enough and this record of success is an omnipresent factor dictating present action.
Shipping firms, and especially traditional shipping firms, do not wish to shake out the market. If one makes competitive moves that change the status quo, someone else will definitely retaliate, since many of us have the size to affect market structure. Why then initiate such a process? (CEO in SH2)
At this stage, we put forward our first propositions:
P1a: Agentic misfit in complex regimes is more likely in organizations that enjoy a sustainable market status in accordance with their expectations and cemented beliefs. Such organizations will proceed with acts of quiescence through prescience, i.e. they will disregard external complexity and commit to reproducing existing internal arrangements in an effort to maintain this status.
P1b: In such cases of agentic misfit, adaptation is neither desirable nor preferable and the law of requisite variety is unlikely to be predictive of organizational acts.
Low external complexity in drug retailing
Drug retailing in Greece is a low complexity sector. Its highly protected nature, the predetermined number of owners and their limited interactions, the extremely low entrepreneurial risk, and the minimal technology usage in a pharmacy make it a stagnated field (Dounas, 2008) and perhaps the most well-regulated industry in the country (Venizelos, 2018). The state not only unilaterally approves new drugs but also centrally fixes their pricing as well as drug wholesalers’ and retailers’ profit margin (Yfantopoulos, 2008). Legislation also dictates a certain number of stores per geographical area and population figures that must be owned by specific licensed individuals (Zacharakis, 2014). Therefore, there is no possibility for a supermarket to sell drugs on its premises. For these reasons, the sector can be safely characterized as one of extremely low complexity and lack of competitive intensity in terms of price setting, locational choices, product diversification, etc.
High internal complexity in drug retailing: Enactment through disruption
Our drug retailing cases are characteristic cases of agentic misfit against their ordered regime. External context was simply a platform for enactive agency. It was neither a setting for matching pursuits nor a constraint that calls for adaptation as an unwelcome compromise. External imperatives may be acknowledged but the impetus for action is intra-organizational.
Yes, I fully understand that other stores see themselves as mere sellers since there is no peer pressure to do otherwise. . . They act by the rules imposed by the Ministry. Well, I am glad that I am not one of them! (owner in DR3)
The organizations’ ethos, competence and idiosyncratic aspirations led them to enact a new code of practice through disruption of the status quo. Hence, instead of emulating existing arrangements, our cases chose intra-organizational arrangements that oppose the stagnated regime in the country.
Yes, of course we became more complex following all these initiatives. But we also became better and this is what we inherently wanted. If someone wants to change things they do not agree with, they have to try harder. And we are all proud that we made it. (owner in DR1)
An increased internal complexity is evident in several activities. Cross-functional meetings where advances in one’s own work are presented, practices such as staff peer assessment, systematic feedback, relationship-building and engagement with customers are standing elements of the ‘deviant’ culture in sampled drug retailers.
We employ salesmen, pharmacists, cosmetologists, a secretary, an accountant, myself, my wife. . . We organize meetings, we train each other on a wide array of things ranging from pharmacological properties of drugs to our salesmen and sales techniques to our pharmacists. We attend external seminars, too, and we are trying to establish closer relationships with the university that will help us enhance our skills, e.g. on marketing tactics or customer service. (owner in DR4)
We made an attempt to document the reasons for those non-conventional acts. Conventional wisdom would dictate that these are costly and, thus, unnecessarily complex. They move over and above formal Ministry directives and are not justified by the protected market status. Hence, why should organizations proceed to misaligned actions, which require extra effort, costs and commitment? Simply selling drugs would appear as a rational, fitting choice.
We envisioned this store with a purpose in mind and we are determined to support this vision. Yes, it definitely requires a non-conventional mindset but the ultimate objective is to change the way our pharmacies serve the public. We do not see ourselves as sellers of drugs; we educate people on drug usage, personal care, healthy lifestyle. . . (owner in DR4)
Visions, aspirations and expected rewards translate into a pursuit for enactment through disruption. Decision-makers therein do not wish to proceed to something in addition to what is being offered. They want to disrupt current arrangements so that actual change of regime is implemented.
We wish society to perceive us as an aid to their daily concerns. Not as an alternative offering but rather as the characteristic example of how a completely new establishment stemming from private initiative should be a standard of excellence that the country needs. (owner in DR2)
Following collective (but uncoordinated) activities, a new generation of pharmacists changed the business model of pharmacy management and, consequently, the way the public sees drug retailing. This became possible through their training in business management which took place in leading schools of the country.
Following initiatives such as ours, pharmacies are not mere selling points of prescribed drugs any more. They became something more, which I label as an enhanced retailing experience. Following my marketing classes, I realized what ‘experience’ means for a customer; something that I have never thought of before. . . well, at least in this way. (store manager in DR4)
A main takeaway is that external imperatives are a backdrop for reflective judgement. They are not constraints calling for adaptation. Rather, constraints are actively deconstructed so that a new possibility is enacted. This is a striking difference compared to the dominant treatment of complex structures as impediments to voluntary action or, at best, as platforms for compromised adaptation. Intraorganizational drivers related to founders’ aspirations, skills or ethos generate mechanisms of amplifying internal complexity. Thus, we propose:
P2a: Agentic misfit in ordered regimes is more likely in organizations that identify a discrepancy between internal features (their aspirations, ethos, skillset) and external structures (stagnant market, institutional and societal forces). Such organizations will be inclined to proceed with non-matching action and strive to enact a new setting through disruption.
P2b: In such cases of agentic misfit, adaptation is neither desirable nor preferable and the law of requisite variety is unlikely to be predictive of organizational acts.
Insights Across Cases
In shipping cases, the complex environment did not induce matching, adaptive practices. Actors, drawn from past experiences and enabled by their market status, preferred to disregard complexity’s overwhelming nature. In drug retailing cases, the ordered environment did not enable conformity and simplicity. Actors, inspired by a creative ethos and driven by their aspirations, preferred to disrupt established norms by complexifying themselves. Overall, the consequentiality of complexity was not limited to an adaptive imperative. Actors, due to their skills, risk orientation or practical wisdom, utilized complexity in variegated ways that do not conform to a fitting logic. Therefore, using complexity as the driver of a solely adaptive response ignores not only the multiple manifestations of human agency but also management studies’ own heritage: the role of equifinality, i.e. the diverse paths towards an end and the value of proactive organizing where external imperatives play a less important role. Thus, we propose:
P3a: Acts of agentic (mis)fit are simultaneously retrospective, contemporaneous and teleological, i.e. they are inherently linked with an organization’s and its members’ past, present as well as an eventual purpose and what is to be achieved in the future.
P3b: Temporalities of action related to (mis)fit decisions imply that adaptation following only contemporary fit imperatives may be neither desirable nor preferable and the law of requisite variety is unlikely to be predictive of organizational acts.
These findings problematize fit theorizing, which assumes that organizations relentlessly seek to become superior performers through adaptation. Certainly, such a focus is fair enough. However, it also under-estimates many organizations’ less maximalistic or non-conventional aspirations. Our cases show that deterministic adaptation of organizations to an external environment is neither a necessary nor a sufficient condition for survival. Survival can be achieved through e.g. non-conformity and deviant responses. Enactment or quiescence, disruption or prescience are also successful, non-adaptive modes of organizing. Most importantly, they appear to be prudent choices in an environment that seemingly calls for matching representations. An organization’s adaptive capacity may be one of its great virtues. Yet, it may also be a stumbling block to unimagined possibilities or may obscure a wider understanding of how organizational success is defined.
Therefore, complexity is not only a constraint. Rather, it acts as a conceptual springboard that enables actors to either circumvent or disregard their environment’s constraining properties. Overall, complexity induces the visualization of alternative possibilities. It drives actors to imagine agentic misfit’s transformative or reproduction potential. Hence, it is a catalyst for the deployment of non-adaptive agency. As shown, this tendency is aggravated in particular circumstances, which allows us to propose:
P4: Agentic – as opposed to deterministic – explanations of organizing and strategizing become more theoretically and practically relevant when there is drive for high structural (environmental) change or high structural (organizational) maintenance. In such cases, the law of requisite variety may not be predictive of organizational acts.
Discussion of Findings
Following aforementioned findings, Figure 1 is ‘updated’ to Figure 2. Therein, we demonstrate boundary conditions for the LRV and its zones of (non)applicability: a zone of agentic misfit (reflected in grey quadrants A and C) and a zone of deterministic fit (quadrants B and D; see examples further below). Thus, Figure 2 indicates that human agency in complexity has two orientations: matching and non-matching. 7 Therefore, without negating LRV-confirming instances, we illustrate our problematization in relation to (1) the exclusive efficacy of fit-as-congruence, (2) the adaptationist oeuvre in complexity-cum-management studies and (3) the LRV’s monolithic appropriation in management. Thus, Figure 2 summates our contribution.

Zones of LRV (non)applicability and (mis)fit.
Figures 3 and 4 represent our empirically grounded models and exemplify the gist embedded in our propositions. They reflect that variably complex environments do not necessarily translate into an inherent constraint or an adaptive compromise as LRV, fit or complexity theorists suggest. Instead, those environments induce instances of agentic misfit and illustrate why the LRV might not apply:
- Quadrant A is where our shipping cases reside. As illustrated in Figure 3, agents act against the backdrop of their past experiences, which consolidate action to a simplistic imperative. Present conditions that enable them to visualize specific possibilities for action include a reflection upon the overwhelming nature of their complex environment and intra-organizational concerns such as their commitment to the status quo or their non-maximalistic goals. So, instead of emulating external complexity internally, these conditions led organizational actors to disregard it through prescience. Therefore, this LRV-disconfirming engagement with complexity demonstrates agentic misfit’s potential to enable organizational reproduction in a complex regime.
- Quadrant C is where our drug retailing cases reside. As seen in Figure 4, agents are driven by their future aspirations for environmental enactment. Present conditions that enable them to visualize possibilities for action include a reflection upon the stagnant nature of their ordered regime, which is misaligned with their aspirations, personal ethos and skillset. Therefore, instead of compromising to an adaptive imperative and simplistic internal arrangements that accord with the low complexity of their sector, they opted to build up an internal structure that induces the possibility of environmental enactment. They sought to complexify themselves so that the transformative potential of their agentic misfit can be realized.
- Quadrants B and D in Figure 2 denote LRV-confirming cases and reside in the zone of fit-as-congruence. For example,
(1) Quadrant D: multinational corporations (MNCs) in corrupted, emerging economies (EE) may likely choose a matching orientation following coercive and normative pressures (Spencer & Gomez, 2011). In fact, complying to corruption through matching may be a ‘necessity for being competitive’ (Collins, Uhlenbruck, & Rodriguez, 2009, p. 89). Thus, relationship with a powerful local network may suffice to facilitate congruence with such markets and manage the arbitrariness and pervasiveness of corruption (Rodriguez, Uhlenbruck, & Eden, 2005). On the contrary, incongruent responses to corruption imply non-matching arrangements that are an unnecessary waste of vital resources (see Jensen, Li, & Rahman, 2010).
(2) Quadrant B: a luxury goods firm addressing affluent customers cannot ignore a complex market that is punctuated with aggressive competitors (Kapferer, 2014). Thus, requisite levels of internal complexity that match the pressing demands for a superior value that justifies excessive prices appears as the only prudent choice (Tynan, McKechnie, & Chhuon, 2010). If luxury firms do not complexify themselves through sophisticated, adaptive marketing strategies (Donzé & Fujioka, 2015) they may decline and perish.

The reproduction potential of agentic misfit.

The transformative potential of agentic misfit.
Our section on agentic misfit, our findings and the configuration of cases in the four quadrants open up some major empirical questions for management research: Why would certain industries be likely to exhibit agentic misfit and others not? What characteristics might be able to predict that? We try to offer some relevant insights below.
As shown, drug retailing cases are proactive contextual shapers and morph their environments through civic action and identity work (Creed, DeJordy, & Lok, 2010). They did not compromise to an established norm nor emulated external arrangements since conformity impeded their operational potential. Thus, they chose to ignore structural standards (Seo & Creed, 2002) and committed irreversible resources for something new (Välikangas & Carlsen, 2019; Walker, Schlosser & Deephouse, 2014). Therefore, contexts where instances of agentic misfit are more likely are the ones that are simultaneously characterized by two features: a stagnant state of being (e.g. due to a rigid regulatory framework) and, at the same time, ones where entrepreneurial agency and disruption is likely to enhance value (e.g. societal welfare through an innovative offering). Thus, major ‘candidate’ contexts include regulated, protected industries, which provide sub-optimal services to customers and lend themselves to disruption.
On the other hand, shipping cases, drawn from past experiences, pursued simpler forms of organizing. Their orientation was acceptable profitability and preservation of traditional norms. Thus, they consciously adopted no response to external contingencies yet survive perfectly well. Certain conditions privilege such dominant organizations, which are in favour of maintaining existing arrangements (Greenwood & Suddaby, 2006). Their choice for ontological continuity reflects a deliberate attempt to maintain or an indifference in altering their inhabited contexts. Therefore, instances of misfit are also likely in industries, which are highly volatile and whose inherent uncertainty is seen as an identity threat by organizational agents. Preservation of traditional norms, maintaining authenticity, safeguarding one’s ontological security, paying tribute to a glorified past are reasons that may enable misfit. They may consolidate a conscious decision to avoid fitting with external, volatile contingencies irrespective of any economic orientation.
Certainly, deploying agentic misfit may not be a prudent choice in other industries and circumstances. For example, (1) in contexts where external legitimacy is sought after (e.g. new ventures in mature fields; Kislov, Hyde, & McDonald, 2017), (2) when compromises are necessary (e.g. in contested or corrupted settings; Jenkins & Delbridge, 2017), (3) whenever emulation promises better outcomes (e.g. adopting practices of prestigious others; Jones & Massa, 2013), (4) when public conformity is anticipated (e.g. in political or institutional arenas; Song, 2019) or (5) when normative compliance is expected (e.g. collaboration in traditional realms; Hibbert & Huxham, 2010) then, in such cases, fit-as-congruence is a more likely scenario. It is important to re-emphasize though that abovementioned fit/misfit scenarios are inextricably related to organizations themselves, too. This is because the same industry feature may lead to both fit and misfit choices across organizations inhabiting the same industry and depending on their leaders’ orientation or beliefs, organizational structures, resource endowment, employees’ skillsets, etc.
Only a future empirical programme can map organizations along zones of fit/misfit and elucidate the environmental contexts or agentic conditions, which enable either option. At this stage, our data only allow us to assert that any answer cannot ignore (1) the purposefulness and identity of focal organizations and (2) the receptiveness of the wider context to disruption and desirable change or its susceptibility to maintenance and conservatism. Given these nuances, our findings extend scholarly discussions in three ways: first, by conceptualizing agentic misfit, we showed that external complexity is not necessarily constraining and consequential in its adaptive sense. Rather, it may simply be a platform for reflective judgement that signposts actors towards visualized possibilities. Thus, through our misfit framing, we challenge one of the most celebrated and enduring ‘dualities’ in management: adaptation as the prerequisite for survival (Burnes, 2005). Second, by fleshing out ‘quiescence through prescience’ and ‘enactment through disruption’ as qualitatively distinct modes of engaging with complexity, we move away from a myopic ontological reification of environmental complexity in the literature, i.e. as a given entity that can be accurately measured and one that should be internally matched (Katsikeas et al., 2006). Thus, we empirically refine an iconic law such as the LRV and extend the explanatory breadth of (mis)fit scholarship. Third, by promoting a chronological ordering of future aspirations, past experiences and present concerns for an understanding of (mis)fit decisions, we illustrate the value of integrated temporalities of action against a contemporary fit orientation, i.e. the norm in relevant research. Thus, we alert scholars that a merely synchronic fit task may generate erroneous insights (Garud et al., 2011; Kodeih & Greenwood, 2014; Poulis, 2020; Poulis & Poulis, 2018).
Conclusion
We connected theoretical strands that revolve around a central management mantra: the efficacy of fit in complex regimes. While this quest for fit is recognized in certain contexts, we showcased zones of non-applicability, which remain largely under-represented or considered irrational in the literature. To the best of our understanding, this is the only empirical study that frames this choice vs. determinism dilemma and, hence, (mis)fit within a complexity perspective. We situated human agency among environments that actors (1) inherited and sustained or (2) problematized and strived to enact and we showed that their agency was channelled through means other than adaptation. Importantly, we chose to discuss agentic misfit through the LRV since the law has been used exactly for that reason across both stable and complex settings: to serve as a rule that solidifies the efficacy of matching, i.e. a main, recurrent theme in management scholarship, which nullifies the role of human agency towards environmental transformation or organizational reproduction.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
