Abhorred by some as a means for social domination, adored by others for its apparent success, hierarchy (from ancient Greek hiereus (ἱερεύς) “priest” and arkhe (ἀρχή) “rule”) remains a dominant feature of contemporary organizations. Despite recent trends towards “flatter” models of management, hierarchical order remains persistent. But why?
This essay sees “hierarchy” as a particular form of group coordination; hierarchy is thus defined by its distribution of roles and decision-making responsibility among the members of a group.
It has three fundamental features: a plurality of social strata, or ranks, to which individuals are assigned; subordination, meaning that people of lower rank are responsible for carrying out the decisions of people of higher rank; and concentration of decision-making power, which implies that there are fewer people at the top of the organization than there are at the bottom, hence its characteristic representation as a pyramid.
I start by analyzing some of the basic structural characteristics of hierarchy. There are two reasons which make hierarchy an attractive form of coordination: its time-related efficiency and, by extension, its capacity to solve collective action problems in large organizations.
The efficiency dilemma
The economist Kenneth Arrow figured that for every type of organization, there is generally a set of desirable characteristics for collective decision-making. However, no form of decision-making can possibly satisfy all desirable characteristics at the same time. Thus any procedure for taking decisions on behalf of a group of individuals will require a trade-off. This goes for any kind of group.
A fundamental trade-off comes with the number of decision-makers. Decisions can be made by few or by many. Each extreme comes with its benefits and disadvantages.
Let us start with many. Consulting many people before agreeing on a decision has the notable advantage of integrating a high amount of individual preferences and knowledge in the decision-making process. And yet, the higher the number of people involved, the more likely the process will be time-consuming. This is because dialogue requires a division of time among participants: while someone speaks, others must wait their turn.
Giving everyone an opportunity to speak in large groups can thus create an endless discussion (there are a few exceptions to this: participatory processes during which all have the possibility to express themselves simultaneously. Voting, surveys, and suggestion boxes are but a few examples of procedures that allow for efficient, although limited, expression of individual preferences.)
One way to avoid never-ending deliberation is to restrict it to limited periods of time. However, the higher the number of participants, the less time there is for each to participate. Large deliberating assemblies end up requiring substantial time to achieve constructive dialogue, and even more to achieve consensus. As such there are limits to the extent to which time can be constrained.
Philosopher Jeremy Bentham also noted that the larger the assembly, the more the “active aptitude” of its members becomes diluted: given their reduced influence, members have less incentive to actively participate. This leaves us with a “vacant seat” problem: some members may decide to opt out in order to spend their time elsewhere, resulting in a decision that is not as representative as one might have wished.
Now let us consider what happens when only a few people are involved in taking a decision. In such situations, there is a risk that decisions will reflect the narrow perspective of those who are in a position to dictate their will. Nevertheless, by avoiding the usual deliberation and confrontation of participatory processes, this may in fact achieve considerable economies of time. The resulting trade-off is illustrated in the following table:
This is one of the structural features of hierarchy: it is simply better at getting things done quickly and efficiently, although this does not guarantee that decisions are well-informed.
Hierarchical organizations usually compensate for this weakness by implementing additional bottom-up information transmission processes, whereby people at the bottom of the pyramid codify and synthesize available knowledge for the benefit of decision-makers. Still, one other reason explains the prevalence of hierarchy: its capacity to deal with organizational growth and complexity.
Dealing with growth and complexity
As an organization grows, its internal dynamics become increasingly complex. Increasing the number of agents within an organization equally increases the probability that some agents will end up having overlapping interests for perceived exclusive resources (whether for financial gain, promotions, reputation, control over projects, or the attention of superiors.)
Theories of ecological competition argue that in such environments, “the intensity of competition depends on the number of actors in the competing population”. In addition, psychologists have noted that an individual’s cognitive capacity to form significant committed relationships is generally limited to about 150 people.
Both ecological dynamics and cognitive limitations thus suggest that as the number of organizational agents increases, so does the probability of political behaviour, dissension and conflict. The capacity for fast decision-making makes hierarchy almost ideally suited to deal with such political problems. Troublemakers can quickly be sanctioned. Whenever an internal or external crisis arises, central authorities can watch, observe, and intervene at the moment that suits them best. Additional layers and subdivisions can easily be created in order to rearrange subordinates into manageable groups. This provides hierarchical organizations with a coercive structure that allows them to react rather effectively to unexpected events.
The capacity for coercion also makes hierarchy an efficient solution for dealing with collective action problems. In fact, many social innovations would probably never have emerged without hierarchy. This is notably the case of irrigation, which saw its first extensive agricultural use in the Fertile Crescent. Manual irrigation is particularly prone to the free-rider problem: because it takes time and effort to bring out water from a river, it is much easier to let your neighbour do the work first and then simply divert his water onto your own land.
Relying on a coercive hierarchy helps avoid such problems by making sure that everyone follows the rules. Many parts of the developing world, such as the rice paddies of rural China, still employ designated “water guardians” to monitor irrigation practices among farmers and warn authorities of any abuse.
The following sections turn to the human and societal elements that contribute to making hierarchy such a predominant feature of contemporary organizations.
An evolutionary heritage
There is an important psychological factor that particularly facilitates the smooth functioning of hierarchy: our natural responsiveness to authority and social status.
German sociologist Robert Michels argued that hierarchy satisfies a basic psychological need to be led. Whether this inclination is innate or socially constructed has been the subject of considerable debate.
Many psychological experiments, such as those of Stanley Milgram, showcase a widespread tendency to bend to authority, a tendency which undoubtedly facilitates the subordination that is required for hierarchy to exist. Milgram believed that humans evolved the capacity to organize into social hierarchies because it provided an important survival advantage.
More recently, clinical psychologist Jordan Peterson has argued that social hierarchy is, in large part, a natural consequence of our hard-wired evolutionary baggage. To illustrate his point, Peterson uses the example of lobsters, which we share a common evolutionary ancestor with. Peterson argues that, like humans, lobsters exist in hierarchies and have a nervous system attuned to social status, through the production of serotonin.
The higher up a hierarchy a lobster climbs, the more its nervous system will be subject to an increase in serotonin. The more defeat it suffers, the more restricted the serotonin supply. Lower serotonin is in turn associated with negative emotions that make it harder for the lobster to climb back up the ladder. According to Peterson, hierarchies in humans work in a similar way. We are wired to live in them.
While this analogy with lobsters is highly interesting, Peterson offers no definitive empirical proof. The animal kingdom provides many examples of hierarchies, with the highest level of organisation observed in insects. These are as closely related to us as lobsters are. They also have serotonin and nervous systems. Plants have serotonin. In animals, including humans, most of the serotonin is produced and used in the intestine to help digestion. Hence serotonin can have different effects in different organisms. The latest research shows that serotonin moderates our response to positive and negative outcomes, which somewhat supports Peterson’s conjecture, although much research remains to be done.
The role of shared values
Another intriguing observation is that people working in efficient organizations rarely think about hierarchy and subordination. Their work is most often given sense by a prevailing organizational ideology: a soldier fighting a war thinks that he is doing a service for his country; a consultant working on his presentation slides thinks that he is adding value to the client’s business; a teacher thinks that he is educating tomorrow’s citizens. Few think in terms of their contractual obligation to obey orders.
This is not surprising. It has long been recognized by both scholars and gurus that workers motivated by strong personal values are less likely to shirk than those who are solely motivated by pure economic gain. The values with which workers identify are often part of what makes hierarchy work so well. Some accordingly argue that “the most effective hierarchies are ones in which subordinates have enough loyalty to the organization and to its leadership that they obey without needing to be forced.”
The adherence to shared sense-making values, which often legitimize hierarchy, has the notable benefit of reducing shirking and associated surveillance costs; it increases the efficiency with which organizations engage in concerted, large-scale collective action.
Many social institutions also reinforce hierarchy
The economist Per Molander points out that in a hunter-gatherer society, where everyone survives at subsistence level, there is usually little material inequality. It is only when there is a surplus that inequality between individuals can develop. This observation is confirmed by the fact that early states show some of the most extreme levels of inequality in history. The only limits their leaders placed on their concentration of power and resources were designed to ensure that agricultural workers would stay alive.
This indicates that the prevalence of hierarchy is likely reinforced by modern institutions, which simply did not exist in the age of hunter-gatherers. These are “complementary” to hierarchy insofar as they mutually reinforce each other. To provide an example, I will focus on the effects of one particular type of institution, namely meritocracy.
The invention and first use of meritocracy is generally attributed to the Qin and Han dynasties of ancient China. In order to rule over an increasingly large empire, the Chinese emperor came to rely upon an intricate bureaucracy of civil servants. What differentiated this state apparatus from previous administrations was the fact that officials had to pass public tests and be appointed to positions according to their competence. This stands in stark contrast to the medieval European church and royal courts, where people were awarded positions of power because of inherited status or arbitrary nomination.
Meritocracy accordingly emerged through the addition of specific rules to the traditional hierarchical system. Positions are filled through what is proclaimed to be a fair contest. But what is a fair contest? Generally, the nomination process constitutes a competitive “contest” insofar as candidates are selected on the basis of how their performance compares to that of others, according to a same method of measurement: there is comparability. The contest is seen as “fair” when all eligible candidates face the same conditions, so that there is equal opportunity, and when the normative criteria according to which winners are selected are determined in advance, so that there is impartiality.
Whether these rules suffice to make a contest fair really depends on the context. Facing the same “legal” conditions does not mean that candidates start with equal chances, because there are always circumstantial factors that cannot be controlled for – individuals may start with fortunate advantages (e.g. due to personal experience, or superior physical attributes), while others may be subject to disadvantaging conditions (e.g. illness, emotional instability). And while selection criteria may be determined in advance, in practice the slightest possibility of prejudice or conflict of interest may skew the entire contest in favor of some particular candidate.
Yet despite these limitations, people often end up believing that candidates selected through a formally meritocratic contest have superior qualities (motivation, intelligence, socially skill), and that they are selected as a reward for these qualities. Nominations, whether for entry positions or for internal promotions, are believed to reflect the intrinsic competence, or “merit,” of the candidate.
It is thus easy to see how meritocracy can add further legitimacy to the authority of decision-makers in a hierarchy. It is also easy to see how these may end up thinking of themselves as being “better” or “superior” to others. People at the top are overrated. The fastest runner is only slightly better than the second fastest, yet he gets all the attention.
Does technology reduce the need for hierarchy? Technological improvements have tended to remove spatial barriers to communication. During most of human history, people had little mobility and could only communicate with a few people at a time. Starting with the printing press, communication technologies made it increasingly easy for people to share and receive information. This in turn facilitated large-scale deliberation and the diffusion of organizational power, as it became increasingly difficult for elites to monopolize the dissemination of knowledge. The advent of the internet has now made it possible for a potentially unlimited number of people to meet and interact in a single place.
However, technology cannot entirely remove temporal barriers: while we may be able to communicate more efficiently and at a much wider scale than before, it is impossible to restrict discussion to a single point in time, considering that we need time to properly articulate our thoughts, and that dialogue necessarily requires that we take turns in speaking. Moreover, communication technologies have brought their share of negative side effects, such as the incredible quantity of noise and disinformation resulting from the easy access to communication tools.
I would argue that our constant strife for efficiency (often reinforced by technology) has made us neglect the benefits of taking our time, as we constantly assume that “faster is better” and that “time is money.” Unlike what many would think, slow non-hierarchical coordination often provides important side benefits.
First, participatory decision-making processes may be time-consuming, but they have the notable advantage of taking more information into consideration and can thus help avoid managerial “blind spots.” This can be called the epistemic advantage of deliberative groups.
Second, slow non-hierarchical governance can increase the motivation of participants: many people report having a feeling of empowerment when they are involved in a decision-making process which directly concerns them and in which they have an opportunity to contribute. However this can only happen if participants have the desire and the capacity to take on an active role.
Third, adequate time simply allows for more rational decision-making. An essential precondition for rational deliberation is that participants actually listen to each other at length, that they focus on a single topic at a time, and that they be willing to reconsider their own beliefs. This allows well-constructed and nuanced perspectives to be formulated. It was first observed by the philosopher Baruch Spinoza, who argued that “men’s wits are too obtuse to get straight to the heart of every question, but by discussing, listening to others, and debating, their wits are sharpened.” Strategies that cope with complex environments will likely benefit from such argumentative fine-tuning.
The rise of hybrids
The rise of communication technologies have clearly opened the door to flatter (less hierarchical) kinds of organizations. Jon Husband associates the rise of technologically intensive and networked organizations with a new form of social organization which he names the wirearchy. This is the “technologically wired” organization, where people take full advantage of technology in order to perform activities through non-hierarchical social coordination.
While many organizations have progressed toward non-hierarchical modes of coordination, nearly all organizations have maintained some form of authoritative decision-making mechanism. Corporations will always maintain at least two hierarchical levels, because their legal constitution makes managers subordinate to shareholders. Yet even outside the world of corporations, hierarchy persists because organizations inevitably end up facing efficiency, growth, and complexity-related problems.
As such there may be strong moral and pragmatic arguments for moving toward less hierarchical organizations, but there are also good reasons for maintaining some form of hierarchy.
So what kind of organizational coordination should be aimed for? Really, it depends on context. Unlike what many business gurus would say, there is no ideal form of management.
It is a fact that small businesses are particularly suited to cooperative arrangements, which in many cases make them even more flexible and resilient. However, small is not always better. Many organizations provide greater benefits to society when they are allowed to take advantage of large economies of scale – in other words, they ought to be big.
A good example is the power utility: building a hydroelectric dam is both a capital intensive and complex task. It requires an organization capable of quickly coordinating a large number of highly specialized people. No such large organization has survived without having at least some form of hierarchical, top-down, authoritative coordination mechanism.
Nonetheless, even in such big organizations, there remains substantial room for managerial innovation. Not all responsibilities have to be defined in a hierarchical manner. Decision-making can be made more representative and better informed. The organization as a whole can be made more proactive and resilient to external shocks by encouraging workers to take local and collective initiatives. This simply cannot be achieved by confining people to squarely defined hierarchical roles.
As Jon Husband rightly points out, we should take advantage of the benefits that technology gives us in order to take up more responsibility, both individually and collectively.