Introduction

In our essay introducing this symposium (Foss and Klein 2023), we described our book Why Managers Matter: The Perils of the Bossless Company (Foss and Klein 2022) as a work in applied organization design. We explained that we authored the book in response to a growing strand of (mostly practitioner) literature arguing that flat, lean, agile, and fluid modes of organization are consistently better than more conventionally structured ones. We pointed out that this literature neglects the role of organizational design and the contingent role of the environment, both proximate and distant (product and factor markets, technological developments, institutions and policies). Our essay laid out some research implications of our perspective on organization design including the need for more empirical evidence, a more robust understanding of authority, and a better appreciation for the relevant contingencies.

The book itself adopts an occasionally polemical tone, partly to engage the interest of practicing managers and also because we enjoy rocking the boat. But we were also upset about some claims and arguments in the bossless company literature. Many of them contradict not only standard ideas in the management and organizations research literature, but also logic and data. Moreover, bossless-company ideas are positioned as applying to any organization, even though research suggests that the performance of any organizational form depends on a variety of contingencies and the “bossless” organization, in particular, can only be realized under special circumstances. We worry that implementing the ideas from the bossless company narrative in organizations where they just do not fit can do considerable harm.

We knew our critique of the bossless company narrative would be controversial, partly because it draws on feelings and beliefs that are strongly rooted in contemporary culture. Some of the literature on tearing down hierarchies, empowering employees, and replacing rigid and mechanistic structures with more “humanistic” ones adopts an underlying moralistic tone. It is often assumed that managerial hierarchies—and authority more generally—are inherently oppressive and unjust, making the bossless company narrative a story about human dignity, workplace liberation, personal autonomy, and so on. Indeed, some reactions to the book (especially among consultants and authors who promote bossless organizing) suggested not just logical or factual disagreement but ideological opposition.Footnote 1

However, while the book is written for the practicing manager it draws on established research in management, organization, and economics. As Jackson Nickerson notes in his contribution, our analysis of flat hierarchies “draw[s] upon the science of organization to distinguish reality from illusion by deeply examining these and other organizations based on scientific foundations instead of specious hyperbole” (Nickerson 2023, 5). Our critique of the bossless company narrative builds on classic contributions from scholars such as Chester Barnard, Alfred Chandler, Ronald Coase, Henry Mintzberg, Herbert Simon, Richard Cyert, James March, James Thompson, Max Weber, and Oliver Williamson, and applies their ideas on organizing to an examination of the validity of the bossless company narrative. We also ask how their core ideas may need to be adjusted given the changes to technology and preferences we have witnessed over the last few decades. We believe that many of our specific arguments are reasonably novel. Therefore, it makes sense to also engage in a scholarly discussion of the book.

The editors of the Journal of Organization Design invited several organization design scholars to provide brief reflections on the book. We are grateful to these commentators as well as the editors for organizing the symposium and allowing us to contribute opening and closing essays. The contributors have raised a host of interesting and important issues related to the theme of the book, including thoughtful objections to some of our arguments as well as suggestions on different roads forward for research in organization design. These include the need to distinguish more strongly between top- and middle management, to include broader psychological ramifications of bossless companies, to treat in greater detail the implications of self-selection into distinct kinds of organizations, to consider more carefully the implications of our argument that to some extent humans are biologically hardwired for hierarchy, and to explore the particular challenges for flat organizations that pursue “social” goals.

In this brief comment, we summarize some reactions to the essays, clarify a few misunderstandings, and suggest additional work to be done.

Gaps and critiques: the reviews

Frank Martela’s essay, “Managers matter less than we think: How can organizations function without any middle management?” agrees with our overall point that managers are necessary (and with most of our specific points as well). Or at least top managers: “managers matter in the sense that larger organizations need someone having an oversight over the wholeness” (p. 4). Martela is less sure about the role of middle managers, however. As he notes, “if working substitutes are found to all tasks traditionally taken care of by middle managers, an organization can be functional and successful without any managerial layers” (Martela 2023, 2). Martela suggests more broadly that we make a distinction between hierarchy and structure. Wikipedia, an example we use in the book to similar effect, has little hierarchy but much structure. Some self-managing organizations exhibit the same feature. We agree entirely with this point, although it could have been brought out more explicitly in the book.

Implied in Martela’s distinction between top and middle management is that we did not sufficiently distinguish between the two. While we did discuss the specific role of middle managers (e.g., pp. 163–65), our arguments are directed at managers and management in general. Martela is interested in organizations like BuurtzOrg and Reaktor that have strong top management but few middle managers and also points out that information and communications technology can substitute for some middle-management functions. We agree and in the book we also note that, for example, the Oticon spaghetti organization and Valve, while having powerful top-managers, largely did without middle management.

We thus agree with Martela’s observations that one can “Combine autonomous, motivated and empowered teams with dedicated IT solutions to help them coordinate inter-team and inter-unit dependencies”, reducing the need for middle managers. We also agree that this “seems to be the secret behind most examples of self-managing organizations I’ve researched or read about—and most examples mentioned by Foss and Klein (2022)” (Martela 2023, 8). We maintain, however, that the extent to which an organization can have “autonomous, motivated and empowered teams” depends not only on management’s deep understanding of the task structure and motivational drivers of efforts, but also on technology: complex technology with strong task interdependencies is only compatible with team organization if the interdependencies can be encapsulated within teams. When this is not possible, dedicated IT-solutions may not be sufficient to ensure a flat, “modular” organization. There will be a need for coordination mechanisms and for making sure that cooperation happens across functions, teams, etc. This is where the need for middle management arises. While we grant the general point that “it is possible to build large organizations without any middle managers, if the organization is able to find functioning substitutes for managers through team empowerment, through ICT, and through other structural solutions and practices” (Martela 2023, 9), the challenge is identifying the conditions under which this is, in fact, possible.

Martela’s focus on the distinction between top and middle management seems to derive from his view that the task of the top manager cannot be “automated”, at least to the same extent as many middle-management tasks. We concur. Thus, in the book, we note that it is unlikely that the judgment performed by top-managers perform can be performed by an AI or other software tools (Foss and Klein 2022, 238–240). Martela does not explain why he thinks top management tasks cannot be automated. We suggest that like an entrepreneur, a top manager performs “judgment”, which we think of as dealing with unique situations, usually oriented towards an uncertain future, where there are few current or past data points that are helpful to decision-making. While this may be called “intuition” or similar, we are not in the realm of mysticism and black boxes. Thus, there are powerful arguments in cognitive science that while AIs can make excellent predictions based on past and current data, only humans are (so far) capable of forming causal theories based on a few data points (e.g., Griffiths and Tenenbaum 2009). Such insights are likely key to understanding organization when data and software and AI tools exist in abundance and warrant much more investigation.

Ambra Mazzelli finds that we have developed an “unobjectionable argument”—a characterization we find unobjectionable! We especially appreciate her recognition that, while the book is written for the nonspecialist reader, we “incorporate a wide range of research drawn from business history (Chandler), economics (Coase, Williamson), organization theory (Simon, March, Cyert, Barnard), and sociology (Weber)—making it interesting for practitioners and academics alike” (Mazzelli 2023, 2). For the academic reader, “one of the many merits of the book rests on its ability to offer a comprehensive examination of the nature and purpose of managerial hierarchy that integrates both micro- and macro-structural perspectives.” That is certainly something we aimed to do.

Mazzelli places special emphasis on the teleological distinction between firms and markets, a theme that runs throughout the book: firms are goal oriented, though those goals are not always designed explicitly and consistently and may emerge endogenously from the interactions of those inside the organization. For this reason, bosslessness—which assumes that all coordination takes place spontaneously—does not always facilitate the adaptation needed when tasks and processes are interdependent.

The second part of Mazzelli’s commentary asks whether new forms of organizing might emerge, transforming the nature of managerial hierarchy without dispensing of it entirely. “What we may observe in the future … is an increased variety of hierarchical forms” (p. 5), especially among organizations pursuing nonfinancial objectives (something we did not cover in the book). One such form is the distributed platform with volunteer contributors (like Wikipedia) in which mission orientation and “trust, transparency, and openness” (p. 6) mitigate the opportunism we often associate with flatter structures. However, the potential cost of resulting coordination failures may be particularly high, giving rise to more formal hierarchy and governance than would otherwise be optimal. “This makes the choice of organizational design in social organizations dilemmatic, involving either (1) minimal internal transaction costs, a better probability of survival, but some risk of incurring internal cooperation and coordination failures that produce significant external costs; or (2) high internal transaction costs, a greater risk of failure, but a minimal probability of producing external transaction costs” (pp. 6–7). Mazzelli suggests that some social organizations will embrace this dilemma by adopting a paradox approach rather than a contingency approach, leading to an increased variety of organizational forms.

Matthew McCaffrey draws on his own work on the videogame company valve (Moller and McCaffrey 2021) to extend our argument on the benefits and costs of flatter structures. As he notes, “Valve is a useful case study for several reasons, especially because it shows how, when formal hierarchy is abandoned, informal hierarchy often replaces it. Yet it is also a fascinating example of how organizational structure can profoundly influence the strategy-structure problem and the way that a company creates value” (McCaffrey, 2023, p. 4).

McCaffrey interprets the Valve case slightly differently from us; while we argued that Valve’s flat model worked reasonably well because there are few interdependencies across projects (games), he argues that game development requires substantial interdependency within projects, suggesting that the flat model should have failed early in the life of the company. The reason Valve has survived, he says, is because it stopped making games and became a distribution platform. “[I]nstead of abandoning flatness to protect its future, Valve changed its future to protect its flatness. In practice, this has meant many failed and abandoned projects resulting in a general move away from software development and toward more modular and less interdependent projects” (p. 4). This is a good point and, while we noted in the book Valve’s strategic shift, we did not connect it explicitly to a mismatch between the firm’s original strategy and structure.

While McCaffrey agrees with our call for a better empirical understanding of bossless companies, he also urges scholars to develop more theory, not only on the benefits and costs of flatter structures, but also on the environmental factors pushing firms toward one model or another. “Foss and Klein mention technological change as a key driver of efforts to delayer and decentralize business, and there are undoubtedly many other factors that encourage or discourage the development of flatter firms, and that influence which specific approach that a company (or rather, a CEO or an entrepreneur) chooses to adopt. In particular, we need some essential conceptual work on the institutional foundations of the bossless company, as well as the relationship between these firms and public policy” (p. 9). We heartily agree!

Jackson Nickerson provides a concise and insightful summary of our bossless-company critique and deftly interprets the second part of our book in the language of Williamson’s transaction cost economics. As he puts it, “the accumulated science of organization … illuminate[s] that much of the benefit of hierarchy comes from its ability to facilitate coordination and cooperation” (p. 6). Specifically, “people, tasks, and processes are often interdependent or tightly coupled. This tight coupling implies that if an adaptation (reactions from a surprise) is needed at the person, task, or process level, then many other adaptations throughout the organization need to be made, generating high demand for coordination and cooperation to facilitate these changes” (p. 6). This is exactly our core argument: hierarchy is an efficient way to respond to unanticipated disturbances in the face of high interdependence.

Nickerson suggests that we pay more attention to “vacillation” between organizational forms, an idea explored in his own work (Nickerson and Zenger 2002; Boumgarden et al. 2012). To achieve the benefits of coordinated adaptation while also realizing gains from worker empowerment and leveraging distributed knowledge, managers could modulate between flatter and more hierarchical structures, just as firms have been observed to vacillate among strategies and structures to develop multiple capabilities or strategic ambidexterity (Kang et al. 2017). As we discuss in the book, several firms that experimented with near-bossless models eventually reverted to more conventional structures, and vacillation theory might provide some insight on the timing of such moves.

Libby Weber notes that although we “only allude to transaction cost economics (TCE) (Williamson 1985)”, our “logic is consistent with this approach” (Weber 2023, 1). In fact, she notes that our basic logic is in line with recent advances in TCE that focuses on internal organization: “While TCE has not been traditionally applied to choosing an efficient hierarchical form, recent works by Foss and Weber (2016; Weber et al., forthcoming) do just that, dimensionalize simple hierarchies, U-forms, M-forms, and project matrices and match them to transaction characteristics to govern internal transactions most efficiently. In this book, Foss and Klein extend this thinking to the choice of bossless companies” (p. 1). Like Nickerson, Weber highlights the importance of coordination to our reasoning and, per implication, an expanded TCE theory that is more detailed about internal organization than received TCE. She also draws attention to psychological costs of bossless organization, something we did not discuss in the book: “In bossless companies, there is no clear roles or rules, nor is there clear formal power, this likely draws employees’ attention to individual motivations. Individual-level hedonic goal frames are likely to lead to effort or visceral opportunism. Furthermore, activation of individual-level gain goal frames results in resource and status opportunism. Thus, in contrast to the prediction that bossless companies allow millennial and Gen Z employees to be more intrinsically motivated, they may in fact be more motivated to be opportunistic instead. As a result, a traditional hierarchy may lead to more firm-level motivations, promoting cooperation and more effectively mitigating transaction costs” (Weber 2023, 11).

While Weber is supportive and appreciative of the TCE aspects of our book, she does not buy our argument that humans may to some extent be hardwired for hierarchy. She argues that we have overinterpreted the biological and evolutionary evidence which in any case deals with social hierarchy rather than firm hierarchy. But we disagree that this distinction is essential. In the book, we argue that informal, social hierarchies and formal managerial hierarchies typically interact and overlap. For example, in the ostensibly hierarchy-free Valve, social hierarchy was established very quickly. In fact, an advantage of firm hierarchy is that it appeals to and may draw on our tendencies to establish social hierarchies. Weber goes on to argue that we “link [the] biological evidence to gender roles in family hierarchies and then in business. They cite Niche, as well as Hebrew, Christian, Muslim, and Hindu creation stories, making a case for men leading the family with women playing a subordinate role.” We are not sure where Weber gets this idea as we do not make this case at all—we simply point to various kinds of evidence of hierarchy existing in history. We are reporting the evidence, descriptively, not making normative claims about how firms (or societies!) should be organized. For example, we note that CEOs are most likely to be tall men, without endorsing this pattern.

Maciej Workiewicz (2023) thinks that one “of the book’s most interesting points is the role of fads and fashions in the rising popularity of bossless forms of organizing at the expanse of traditional hierarchies. Most young people today desire not a cushy managerial role but rather to be a leader or entrepreneur. Over the last two decades, and especially since the financial crash of 2008, we’ve witnessed a progressive disillusionment with hierarchies, institutions, and elites in general (Gurri 2018). … The authors of Why Managers Matter worry that such ideologies are overshadowing still-important fundamentals” (p. 4). He extends this into a critical point: “Where the book could have gone further is in discussing to what extent the desires of the public, in general, and potential employees, in particular, should be taken into account in designing organizations. To the frustration of many, workers march to the unyielding rhythm of demographic forces. Each generation, from the silent to boomers, Generation X, millennials, and Gen Z, brings its own ideas, expectations, and, very importantly, numbers. The rift between the silent generation’s acceptance of militaristic organization design and the boomers’ preference for more flexible approaches like matrix created a significant shift in American corporations in the 1970s.”

As we also indicated in our introductory essay (Foss and Klein 2023), organizational design takes place within a cultural context. In the book we discuss some cultural roots of the bossless company narrative, tracing some of its origins to the counterculture as it took shape on the U.S. West Coast. More generally, the extent to which organizations can delegate (and to whom), the way the structure their incentive plans, how much they communicate laterally and vertically, and so on are all affected by culture. This may not only hold across different contemporary cultures, as suggested by international business research (e.g., Sethi and Elango 1999; Steers et al. 2009), but also over time, as Workiewicz (2023) and our book suggest.

Final reflections

These commentaries are thoughtful, reflective, and suggestive of many new directions for research on the flattening hierarchy and on organization design more generally. Here we discuss some of these research opportunities in more detail, then turn to a discussion of engaged scholarship in organization design.

Organization design as a multi-disciplinary undertaking

An interesting theme across the commentaries is the variety of theoretical lenses, frameworks, and constructs that can be applied to bossless organizing. In this respect, Martela’s neat commentary is the one that stays closest to traditional organization design thinking. Nickerson and Weber interpret our arguments and their extensions in terms of transaction cost economics (and, in Weber’s case, motivational psychology), while McCaffrey draws also on the Austrian tradition in economics. Mazzelli relates to both core organization design ideas that derive from the thinking of Herbert Simon as well as more recent ideas on social goals. Workiewicz (2023) draws attention to the cultural matrix within which organization design takes places. By drawing on unusual concepts, theories, and disciplines the commentaries point to important, but unexplored opportunities for research.

This is particularly striking, because in the book we made a point of drawing not just on what many would see as organization design theory proper, that is, information processing (Galbraith 1977) and contingency/configurational approaches (Mintzberg 1983), but also on ideas from history, economics, psychology, organization theory, ethics, and other disciplines and fields. Beyond the suggestions for extensions of our basic reasoning offered in the commentaries, we could have gone into more detail on issues related to with procedural justice, workplace motivation, team interaction, and other issues covered in the vast organizational behavior literature. Organization design is not an engineering exercise decoupled from the “softer” aspects of the internal and external environments, but on the contrary need to take these into account. To paraphrase Peter Drucker’s famous quip, culture may eat organization design for breakfast.

The broader lesson is that progress in organization design thinking is likely to come from exploring the intersections between organization design and the disciplines, fields, and critics identified in these commentaries (as well as others not discussed) rather than from extensions and refinements of existing organization design theory. To be sure, opportunities remain for refining such theory, for example, in terms of understanding new emerging organizational forms (cf. Martela and Mazzelli’s commentaries). However, organization design theory has certain gaps that derives from its traditional macro-structural focus. It is less strong when it comes to microfoundations, as partly reflected in what is colloquially known as the “OT/OB divide.” A number of the commentaries (notably Weber and Mazzelli) point to the value of building more explicit and solid microfoundations for organization design theory. But organization design can also benefit from relating to other relevant macro streams such as culture research or research on organizational belief systems or (even more macro), ideas on institutions and institutional logics. These weaknesses threaten the relevance and applicability of the theory as organization designers may fail to take employee reactions to organizational design initiatives into account or misunderstand how organization design is influenced by the prevailing culture, whether in the organization itself or in the surrounding society.

Organization design as engaged scholarship

We think our book may be seen an example of “engaged scholarship”, that is, deliberate attempts by scholars to engage with managerial practice in a way that goes beyond consulting and has a public purpose (Van de Ven 2007). As mentioned above, we authored the book partly a critical reaction to claims in the practitioner literature we think are exaggerated, often wrong, and potentially harmful. Of course, engaged scholarship can be controversial.

In our case, the book is positioned against the provocative bossless-company narrative which invites controversy, so it’s not surprising that the book has been somewhat divisive. Some critics on social media were less than enamored with our thesis (Ravier 2022), claiming we failed to understand and appreciate the true message of the worker self-management literature. As noted above, others took our message to be a naïve defense of “traditional” hierarchy (e.g., Robledo 2022). In fact, we devoted considerable space to the future of hierarchy and how authority is changing, from what we called Mark I authority (command and control) to Mark II authority (defining and enforcing the organizational rules of the game). Mazzelli’s commentary explores this in more detail.

Some of the more negative reactions can be understood in the light of the view many seem to hold of hierarchy, particularly with respect to its supposed inherent asymmetries of power. For example, here is Christian Bourion (2005, 85) describing such asymmetries:

Begin with a confrontation with an aversive situation, add confrontation with the hierarchy, and you have a situation with double constraints, giving rise to all sorts of pathology. If the subordinate succeeds brilliantly, he finds himself in competition with his superior. Thus, the hierarchical situation is essentially conflictual, whether the subordinate succeeds or fails. The weight of the hierarchy is that much heavier when the actor is at the bottom of the pyramid, and the emotional conflict that results from the double confrontation is all the more profound if the gap between what is demanded by the situation, and what is demanded by the superior is great.

As we noted in the book, this view of hierarchy as inherently dehumanizing and oppressive inspires some strands of the bossless-company narrative. Flatter hierarchies are liberating and empowering. These social, historical, cultural, and often emotional attitudes and beliefs must be navigated carefully when engaging the broader community.

Engaged scholarship also highlights the fact that practitioners and academics may have different lodestars. Many of the scholars we cite in the book (e.g., Weber, Coase, Williamson, Chandler, even Simon) are not well known among practitioners, even though Mintzberg, who is also cited with considerably frequency, is more of a crossover figure. Likewise, popular writers like Tom Peters and Frederic Laloux are unknown to many academics, and some academic management theorists are more influential among practitioners than among scholars. For example, a number of people have written to us pointing out that we missed the contributions of Elliott Jacques (e.g., Jacques 1976, 1997), particularly his 1990 Harvard Business Review article, in praise of hierarchy (Jacques 1990). In this article Jacques, who had an unusual background including a study of psychoanalysis, was dealing with the interplay between hierarchy and employee psychology which we above, referring to Weber (2023), identified as a major research opportunity. This is a telling example of how knowledge production related to organizations in academia and among consultants, gurus, and executives may develop along different, parallel trajectories. Because organization design as a field is inherently close to the significant concerns of practitioners, it also has the potential to bridge these trajectories. In other words, there is an enormous potential for organization design scholars to be or become engaged scholars. We hope Why Managers Matter is helpful in this respect!