In 1995, I was a staff member at Japan's Fire and Disaster Management Agency when the Great Hanshin-Awaji Earthquake struck. I spent those days coordinating support for overwhelmed municipalities behind the scenes. Sixteen years later, in March 2011, I was working at the city of Sapporo when the Great East Japan Earthquake and tsunami devastated the northeastern coast. This time, my job was to organize the dispatch of Sapporo's own personnel to affected towns and cities.

From both vantage points—one national, one local—I witnessed the same structural failure playing out at different scales: Japan's disaster response system is built around the municipality as its primary unit, but municipalities, as currently constituted, are not capable of functioning as self-sufficient crisis management entities.

1.  The 1995 Earthquake

The Great Hanshin-Awaji Earthquake struck directly beneath one of Japan’s most urbanized corridors. Within hours, the administrative capacity of affected municipalities had largely collapsed. City halls were in chaos. Information was contradictory and incomplete. Many of the civil servants responsible for managing the emergency were themselves survivors.  Around-the-clock demands for shelter registration, supply distribution, and damage assessment rapidly exhausted organizations that had never been designed with redundancy in mind.

And yet the 1995 disaster, while devastating, remained geographically contained. The affected zone, while densely populated, was geographically limited. Neighboring municipalities were functional. National-level support, though slow to mobilize, eventually arrived. The system held—not because it was well-designed for crisis, but because enough of the surrounding infrastructure survived to compensate for local failure.

The lesson I drew from 1995 was clear: municipalities are the front line of disaster response in Japan, but they are not—and cannot be—self-sufficient.

2.  The 2011 Earthquake: When Everything Collapses at Once

The Great East Japan Earthquake and the tsunami that followed it demolished that assumption entirely. Along the Tohoku coast, multiple municipalities were struck simultaneously and catastrophically. In some towns, city halls were physically swept away by the tsunami. Personnel files, land records, and citizen registration databases were lost. A significant number of municipal employees—the people whose job it was to manage the crisis—were among the dead.

This was something qualitatively different from 1995. It was not a single municipality in crisis while others around it remained functional. It was simultaneous, multi-jurisdictional administrative collapse across an entire coastal region, compounded by a nuclear emergency that added a layer of national-level crisis management.

Japan's disaster response framework, as codified in law, assigns primary responsibility for emergency operations to the municipality: shelter management, damage assessment, the issuance of disaster certificates, reconstruction support. The system is predicated on the municipality being able to function. In 2011, that predicate was false for dozens of jurisdictions at the same time.

This is the structural flaw. It is not a question of municipal competence or dedication. It is a question of system architecture. A framework built on individual municipalities as self-sufficient units will fail when multiple municipalities fail simultaneously—and large-scale disasters, almost by definition, tend to cause exactly that. 

3.  Personal Networks Over Institutions

When the earthquake struck on March 11, 2011, I was in Sapporo. Official information in those first hours was fragmentary and unreliable. The scale of the disaster was evident, but the specific needs of specific places were not.

Then a colleague—someone I knew personally from years of working across agencies—called from one of the affected prefectures. “We don’t have enough people,” he said. “Can you get anyone out here fast?” That single call clarified the situation in a way that no official report had. Within hours, I had organized a rotation of Sapporo staff to deploy in two-week shifts to affected municipalities. Around the same time, a senior officer in Sapporo’s fire bureau—someone who had also lived through the 1995 earthquake with me—did not wait for formal authorization. He secured ferry capacity from Hokkaido to the affected area to transport fire vehicles, personnel, and relief supplies all at once.

Formal mechanisms for inter-municipal staff deployment existed on paper. But translating those mechanisms into action required specific, reliable, real-time information about where help was needed and in what form—information that the formal system was not providing. What filled that gap was trust accumulated over years and a shared memory of the previous disaster.

Resilience in a crisis depends not only on institutions, but on the relationships between the people who staff them. Yet a system that relies on personal networks is not a sustainable one.

4.  A Model That Already Works: The Emergency Fire Response Team

Japan has, in fact, already solved this problem—but only for one sector.

The Emergency Fire Response Team (Kinkyu Shobo Enjotai) is a standing national framework under the Fire Organization Act, activated at the request of the Minister of Internal Affairs and Communications (the Commissioner of the Fire and Disaster Management Agency). When a major disaster occurs, the Fire and Disaster Management Agency can dispatch fire units from across the country to the affected region. Critically, this deployment does not require a formal request from the affected municipality—the system is designed to activate automatically, before the municipality has the capacity to formulate a request. In the 2011 disaster, deployment orders went out on the day of the earthquake itself. Tens of thousands of firefighters were mobilized in the days and weeks that followed.

The system works because it has the right architecture: pre-registered units, a clear chain of command, standardized training and equipment, and an automatic trigger mechanism that bypasses the bottleneck of formal inter-municipal negotiation.

In other words, Japan already operates what might be called an automatic regional surge model—but only for fire response. The same logic has never been applied to general municipal administration.

5.  The Missing Piece: Administrative Surge Capacity

The functions that proved most acutely understaffed in the years following the 2011 disaster were not emergency services. They were the grinding, technical, long-duration tasks of post-disaster administration: processing disaster certificates (a prerequisite for survivors to access insurance and government support), land surveying for reconstruction, urban replanning in areas where entire neighborhoods had been destroyed, and the sustained case management required to help displaced residents rebuild their lives.

For all of these functions, there is no standing rapid-response system. Staffing for these roles has depended on bilateral coordination between municipalities, brokering by national bodies like the Ministry of Internal Affairs and Communications or the national associations of mayors and village heads, and the patchwork of mutual-aid agreements that municipalities have voluntarily entered into with one another.

The reason fire response was institutionalized while administrative response was not comes down to a categorization problem. Firefighting has long been understood as an emergency function requiring dedicated, standing capacity. Post-disaster administrative work, by contrast, has been treated as an extension of ordinary municipal operations. The earthquakes showed that rebuilding lives and communities is itself a prolonged crisis response—and one that demands equivalent institutional recognition.

6.  Toward a Redesign of Regional Governance

The conclusion I draw from these two disasters is not that municipalities need to be larger, better funded, or more capable in isolation. It is that the unit of governance for crisis management—and, increasingly, for routine public administration—needs to shift from the individual municipality to the region.

This is not merely a disaster-preparedness argument. The mismatch between administrative boundaries and actual lived geography is already stark. Economic activity, healthcare networks, infrastructure systems, and daily commuting patterns all operate across municipal lines. The disasters of 1995 and 2011 made that mismatch violently visible.

Regional coordination cannot remain a matter of goodwill between neighboring governments. It must be a structural feature of the governance system itself.

What would a redesigned system look like in practice? At minimum, it would include a standing regional pool of specialized administrative personnel—staff with expertise in disaster certification, land administration, and reconstruction planning—available for rapid deployment across a defined geographic region. It would also include an automatic activation mechanism for administrative surge support, triggered by disaster scale rather than by a formal request from an overwhelmed municipality. And it would require a reconsideration of how fiscal resources are allocated across levels of government, so that regional coordination is not a favor that well-resourced municipalities extend to their neighbors, but a funded and institutionalized function of the system as a whole. The administrative framework of the future must move away from a design premised on self-sufficient individual municipalities toward one built around regional governance as its foundational unit.