Flat hierarchies: They are just another step in the wrong direction

Organizations should not be flat, but decentralized. For a simple reason

Lofty talk about making organizations flatter persists. We got accustomed with efforts to reduce middle management, with skimming the amount of hierarchical layers (supposedly), or with scrapping internal bureaucracy in order to achieve more efficiency, more effectiveness, and more enterprise agility. We hardly notice the continued futility of such efforts any more. Flat means we keep barking up the wrong tree. Trying to flatten organizations, we perpetuate the wrong system: Because in flat organization, command-and-control and steering from the top remain.

By now, many people understand that hierarchical steering of organizations is a lose-lose proposition. Even though it was once a pretty good idea: That was during the industrial age, an era that ended in the 1970s or 1980s. At that time, the ability of markets to surprise us increased significantly. It was a paradigm shift and the dawn of a new age: Value creation in the knowledge age is far more dynamic and more prone to surprise than it was in the industrial age. The importance of services, customization, individualized or high-variety production, uncertainty and highly competitive markets rose dramatically. That means: in every organization today, the periphery, or the outer part of the organization must be put in charge. The logic of ‘top-down‘ and ‘bottom-up‘ has become a trap.

Organizations outsourced the steering
to markets fifty years ago, during the 1970s.

As complexity and dynamics reign today’s markets, these confront organizations with vicious market-pull and ever-more constant levels of surprise. Consequently, any kind of centralized counter-push, or steering from "above" in the form of command and control ceases to function properly. In a complex world, internal steering collapses. Whereever we continue to steer from above (or “from the inside-out”, more precisely), such steering inevitably leads to mis-leading actions. Internal steering will cripple value creation, instead of enabling it.

Value creation flows from the inside out and towards the market: This is a key fact that applies for any kind of organization, anywhere in the world. Which means that “flat” is profoundly flawed metaphor, to describe organizations

It is for this reason that flat hierarchies are thus not a solution at all, but a continuation, or perpetuation of management traditions that have since long turned into a mistake. Today, if an organization does not turn to the market consistently, but remains in the mode of hierarchical steering, then you can try to reduce bureaucracy, and slash hierarchical layers as much as you like: middle management will always grow back. The same goes for bureaucratic steering rituals like target negotiation, micro-management, budgeting, planning processes, allocations, cost management, and excessive rules and policies. We cannot make organizations lastingly “agile” while we keep steering them from the top down, or inside-out!

Those who talk of layers of resistance in the middle
have not yet properly understood the problem yet.

It is only in the mode of "decentralization” that the need for having a middle management and an organization’s appetite for centralized steering disappears entirely. In decentralization, or federalization, self-organization and highly effective steering from the outside become possible. In a federalized organization, hierarchy becomes recognized as a somewhat “trivial" phenomenon. Value creation towards the market, from center to periphery to client can become the dominant principle. Value creation can flow from the inside out - unhindered by internal steering.

It is only in the mode of decentralized that the reason for steering and for having a middle management will disappear entirely.

You better stop trying to make your company "flatter", because in complexity, an organization must be federative - not flat. When outside markets rule, it is the part of the organization we call the periphery that earns the money. It is the periphery that learns from the market easiest. It is the periphery that can best adapt to and respond to markets. Remember: Market complexity already provides the steering: Organizations must learn to respond quickly, intelligently, consistently and with discipline. Teams in the periphery are the only ones that can absorb market complexity well, as they “touch the market” directly and can learn from their client interactions at the highest rate.

Market complexity and globalized competition have consequences for the organization’s center, too. Here, the center loses its information monopoly – its competence advantage, compared to the periphery: The center can hardly issue any meaningful commands anymore. The linkages between periphery and center must now be designed in a way that enables the organization to absorb and to process market dynamics. For that, the periphery must steer the center through market-like mechanisms and it must own the monetary resources. Not the other way around. (But hey, the periphery earns the money by serving clients, anyways, right?)

Decentralization as devolution of autonomy and decision-making power to the periphery

In the face of complexity, aspiring to organizational flatness is a joke. Decentralization, or federalization works according to laws different than those of hierarchy. Where the principle of decentralization is applied, the devolution of autonomy and of decision-making power to teams in the periphery can go on, and on, and on forever. This way, teams in the periphery can act more and more independently, assume an ever-increasing amount of roles and responsibilities. Levels of disciplined self-organization and entrepreneurial capability in periphery teams can keep increasing forever. Decentralization never ends.

---

Further reading

This article was originally published in German language in July 2015, and in English in October 2016. It has been updated, edited and modified. A version of this article appears in Niels Pflaegings book Essays on Beta, Vol. 1.