A hierarchical organization needs two things from its communication structure: fast coordination downward (so everyone aligns quickly) and reliable information upward (so decisions at the top reflect conditions at the bottom). These are not the same requirement, and optimizing for one degrades the other.
The model uses continuous-time dynamics on hierarchical networks where nodes update based on their neighbors' states. Linear systems theory and perturbation analysis show that strong inter-layer coupling speeds convergence — orders from the top propagate rapidly through middle management to the floor. But the same strong coupling that carries commands downward suppresses the signal that would carry information upward. A frontline observation must swim against the current of top-down coordination to reach the top, and in networks optimized for coordination speed, it drowns.
The tradeoff is structural, not managerial. It is not that bad organizations ignore their employees. It is that the communication architecture which makes an organization responsive to commands from above is the same architecture that makes it deaf to signals from below. The bandwidth is shared. Optimizing the downward channel occupies the network's capacity, leaving less for the upward channel.
This has a formal shape. Organizational breadth (how many direct reports per layer) and inter-layer coupling strength (how tightly layers are connected) are the control parameters. For any fixed network, there exists a Pareto frontier: you can be fast at coordination, or good at information sharing, or somewhere in between. You cannot be both. The tradeoff is not a design flaw to be solved but a geometric constraint of hierarchical structure. The same pipe carries both signals, and they flow in opposite directions.
(arXiv:2603.18701)