Introduction
The COVID-19 pandemic served as a global stress test for political systems, revealing stark differences in how governments respond to existential threats. While conventional political science often emphasizes the "efficiency" of centralized power—the idea that a single, strong leader can move faster than a complex bureaucracy—this research challenges that narrative. By applying complex systems theory and formal mathematical modeling, the authors argue that federalist systems, characterized by overlapping layers of authority, possess a structural "policy rescue" mechanism that enhances responsiveness and resilience compared to centralized or autocratic regimes.
The Problem of Signal Failure
In an existential crisis, the primary danger is not merely slow movement, but "signal failure": the risk that a central decision-maker misses, ignores, or misinterprets critical data. Centralized systems have a single point of failure; if the top organ fails to recognize a threat, the entire nation remains unprotected.
The authors treat governments as "decision-making automatons" that sample evidence from a noisy environment. In this framework, the topology of a political system—specifically the number of independent "organs" capable of making policy—determines its ability to overcome signal errors.
The "Policy Rescue" Mechanism
The core of the paper's argument lies in the concept of "redundancy" or "degeneracy," terms borrowed from biology and engineering (specifically the work of John von Neumann). In a system with $m$ independent layers of government (e.g., federal, state, and local), the probability of a citizen not being protected is $\epsilon^m$, where $\epsilon$ represents the probability of a signal error at any single level.
As the number of independent policy-making layers ($m$) increases, the probability of protection ($1 - \epsilon^m$) rapidly approaches 1. This "policy rescue" occurs when subnational units act independently of a stalled or dismissive national government. While having multiple layers of government might seem "inefficient" during normal times, it acts as a critical fail-safe during emergencies.
Bayesian Updating and Independent Learning
Beyond simple redundancy, these systems facilitate faster "belief updating" through a network of independent learning. Using Bayesian decision theory, the model shows that policymakers observe the actions of their peers in other jurisdictions.
If a governor in one state observes a neighboring state implementing a "shelter-in-place" order, they receive a secondary signal that the threat is severe, even if their own primary data was ambiguous. This allows late-adopting jurisdictions to learn from the successes and failures of early-adopters, potentially leading to long-term cost reductions and more effective policy solutions.
Comparative Evidence: The US vs. The UK
The authors illustrate this logic through the early 2020 responses of the United States and the United Kingdom. Despite both national leaders initially adopting dismissive stances toward the virus, their institutional structures led to different outcomes:
- United States: The federal structure allowed states like California and Washington to trigger strong policy responses independently of the White House. This subnational action provided the "policy rescue" predicted by the model.
- United Kingdom: As a unitary system with limited subnational policy authority, the national response closely mirrored the central leadership's initial delays. Without independent "organs" to initiate a rescue, the entire nation remained tethered to the central government's signal failure.
Conclusion
In the face of existential threats, redundancy is a feature, not a bug. The "inefficiency" of federalism provides a robust defense against the catastrophic risks of centralized signal failure. By ensuring multiple access points to policy-making, redundant systems increase the chances of crafting and replicating the "correct" response to a crisis, ultimately protecting citizens more reliably than centralized alternatives.
Summary updated