Topic 3.12 — Crisis, Emergency, and Resilience Management

Why this matters in public value

Every public body now operates inside a longer run of overlapping shocks: extreme weather arriving more often and more severely, pandemics, cyberattacks on critical infrastructure, supply chain shocks, and conflict-driven displacement. Citizens judge governments and social-sector organisations less on any single decision than on how those organisations behave when the ground moves — whether the hospital keeps functioning, whether the evacuation order comes in time, whether the food bank can scale up overnight. That judgement is durable: a single well-handled or badly handled crisis can reshape public trust in an institution for a generation (see Topic 4.2 — Trust in Government and Civic Engagement).

The budgetary logic works against readiness. Standing capacity to respond to a crisis that may not happen this year competes, in every annual budget round, against visible service delivery that voters and funders can see today. A well-maintained flood defence or an exercised continuity plan produces no headline when it works — its value shows up only as the disaster that did not happen. This is the central tension of this topic: resilience is expensive to build, invisible when it succeeds, and unforgiving when it is absent.

Crisis, emergency, and resilience management is therefore not a specialist annexe to public management; it is a test of whether an organisation's public value proposition — its legitimacy, its operational capacity, its stated value to citizens — holds up under the one condition that most reveals whether it was ever real (see Topic 1.2 — The Strategic Triangle: Legitimacy, Value, and Capacity). This topic treats resilience not merely as risk mitigation but as a public value in its own right, one that societies should be willing to pay for in advance, in the same way they pay for defence or public health.

Core concepts

A hazard is a potential source of harm — a flood, a pathogen, a cyber intrusion, a failure of a critical supplier. A hazard becomes a disaster only when it overwhelms the coping capacity of the people and systems it affects; the same flood that is a minor inconvenience in a well-drained, well-insured city can be catastrophic without those defences. Risk is the combination of a hazard's likelihood, its severity, and the exposure and vulnerability of what stands in its path — the analytical apparatus for assessing that combination is covered in full elsewhere, and this topic builds on it rather than repeating it (see Topic 3.10 — Regulation and Public Risk Management).

Emergency management is the discipline of reducing vulnerability to hazards and coordinating the response when they occur, organised around a cycle of four phases. Prevention and mitigation reduce a hazard's likelihood or severity before it strikes — building codes, flood barriers, vaccination programmes, cyber-hardening. Preparedness builds the capacity to respond well when prevention fails — plans, trained staff, stockpiles, exercises. Response is the operational phase during and immediately after an event, and recovery is the longer task of restoring communities and systems, ideally to a state less vulnerable than before rather than an identical rebuild.

Crisis leadership is the human dimension of the broader field of crisis management, and differs from routine management in three ways. Decisions must be made under severe time pressure, often in minutes rather than the weeks a normal policy decision allows; the information available is always incomplete and frequently wrong in the first hours, so leaders must act on a working picture they know will need revision; and scrutiny is immediate and public, so a leader's visible conduct becomes part of the public's assessment of whether the institution is in control. Crisis leadership therefore combines two jobs at once: making the best operational call available, and providing the sense-making that keeps public trust intact while the evidence is incomplete.

An all-hazards approach builds planning, command structures, and core capabilities that work across many hazard types — flood, pandemic, cyberattack, industrial accident — rather than maintaining a separate, siloed plan for each one. Because capabilities such as communication, logistics, and mutual aid are common to nearly every hazard type, an all-hazards framework with hazard-specific annexes copes far better with the hazard nobody planned for than a library of single-hazard plans does.

Business continuity planning is the organisational counterpart to emergency management: keeping an organisation's own critical functions running, or restoring them quickly, when disrupted by an incident that may have nothing to do with a public emergency — a fire in a data centre, the loss of a key supplier, a strike. The reference standard is ISO 22301, published by the International Organization for Standardization (ISO), which specifies requirements for a management system that protects against, prepares for, responds to, and recovers from such incidents. The two disciplines overlap but are not identical: an organisation can have a superb continuity plan for its payroll system and no emergency plan for a flood affecting its whole service area.

Resilience is the capacity to anticipate a shock, absorb it, adapt during it, and — ideally — emerge stronger than before. Engineering resilience concerns how a physical system withstands damage and returns to function; community resilience concerns how households, neighbourhoods, and institutions cope and recover, often drawing on social capital as much as physical assets. A technically resilient bridge is of little use to a community with no resilient way to reach, evacuate across, or fund the repair of it.

Resilience has the structure of a public good (see Topic 1.4 — Public Goods, Merit Goods, and Value Pluralism): once a flood defence or early warning system exists, its protective benefit is largely non-excludable (everyone in its footprint benefits, whether or not they funded it) and non-rival (one household's protection does not diminish a neighbour's). Left to individual or market provision, resilience is systematically under-supplied, because no single actor captures enough of the benefit to justify paying the full cost — precisely the public value case for collective, advance-funded investment in it.

Finally, many of the most damaging failures are not single hazards but cascading or compound crises, where one failure triggers or masks another — a storm knocks out power, which disables water treatment, which forces a hospital onto backup generators that then fail. The sociologist Charles Perrow's Normal Accidents, his analysis of complex, tightly coupled systems, argued that in such systems some accidents are close to unavoidable regardless of how careful any operator is, because failures propagate through interdependencies faster than anyone can intervene. Planning that only ever tests one hazard at a time will not find the failure that comes from two or three arriving together.

Best practices

Treat the emergency management cycle as a standing capability, not a project. Prevention and mitigation, preparedness, response, and recovery are continuous, resourced functions with named owners, revisited on a fixed cycle rather than only after an incident forces the issue.
Plan on an all-hazards basis, with hazard-specific annexes layered on top. Build common capabilities — command, communication, logistics, mutual aid — once, then add the detail a flood, a pandemic, or a cyber incident each require, avoiding the trap of being superbly prepared for the last crisis and blind to the next.
Invest in monitoring and early warning before investing in response capacity. A day of warning time is often worth more than a week of extra response capability, because it lets people and systems move out of harm's way rather than be rescued from it. Early warning systems are also among the clearest examples of resilience's public-good character: cheap to extend to one more household, and useless if withheld from any.
Rehearse before the capability is needed, not after. Tabletop exercises, functional exercises, and full-scale simulations surface the gaps a written plan hides — the phone tree that is two years out of date, the authority nobody remembers exists. Exercise across organisational boundaries, not only within one, because most real crises do not respect an organisation chart.
Pre-negotiate mutual aid and surge capacity agreements before a crisis, not during one. Agreements on shared equipment, staff secondment, and financial backstops take weeks to negotiate in normal times and are nearly impossible to negotiate under crisis pressure. Local, regional, and cross-border arrangements should be signed, funded, and tested well in advance.
Define command tiers and decision authority in writing, and test that the authority actually works. Separate strategic (objectives and priorities), tactical (coordinating the response), and operational (executing on the ground) decision-making, with clear escalation triggers between them, following a standardised model such as the Incident Command System where one is in use. Know, and rehearse, exactly who can declare an emergency and activate the plan at three in the morning on a public holiday.
Build and test a business continuity plan for the organisation's own critical functions. Identify the functions whose loss would be unacceptable within hours, not weeks, and design recovery arrangements — alternative sites, data backups, delegated authority — for those functions specifically, following a recognised standard such as ISO 22301 rather than an ad hoc checklist.
Communicate early, honestly, and through a single trusted channel. In the information vacuum of a crisis's first hours, rumour and misinformation fill whatever space official communication leaves empty. Say what is known, what is not yet known, and when the next update will come, even if the substantive answer is still "we do not yet know."
Design recovery to reduce the next disaster's risk, not merely to rebuild what existed before. "Build back better" is a discipline, not a slogan: recovery funding is the single largest pool of money most public bodies will ever have available for resilience investment, and spending it to recreate yesterday's vulnerability is a wasted opportunity that will be paid for again.
Fund resilience as a standing line in the base budget, not only through emergency and disaster funds. Reactive, after-the-fact disaster funding is typically many times more expensive than the mitigation spending that would have reduced the need for it, yet base budgets are the first place resilience investment gets cut in a tight fiscal year. Protect a minimum allocation the way a defence budget protects standing readiness.
Run a structured after-action review after every significant incident and exercise, and track whether its findings were acted on. A lessons-identified report that nobody implements is indistinguishable, from the next crisis's point of view, from never having reviewed anything at all. Assign owners and deadlines to each recommendation and report on completion.
Protect the welfare of the people who staff the response. Protracted crises exhaust responders physically and psychologically, and burnout among experienced staff is itself a resilience risk that hollows out the capacity the organisation is trying to protect. Build rotation, rest, and post-incident support into the plan from the outset, not as an afterthought once people are already depleted.

Questions to discuss with your team

Who has the authority to declare a crisis, and have we tested that it actually works at three in the morning on a public holiday? Authority that exists only on an organisation chart is not the same as authority that functions when the usual chain of command is asleep, travelling, or unreachable. Walk through, concretely, who is contacted first, what happens if that person cannot be reached within 15 minutes, and who can practically make the first consequential decision. Many organisations discover, only during a real incident, that the named authority has changed jobs, the backup contact list is out of date, or no one below the top of the hierarchy believes they may act without permission. Resolve this before the test is a real one.

What is our resilience budget, and would we recognise it if someone asked us to point to it? Most organisations can name their disaster response budget but struggle to identify what they spend, in an ordinary year, on prevention, mitigation, and preparedness. If resilience spending is invisible even to the people managing it, it will be the first thing cut when budgets tighten, because nobody will notice its absence until the next crisis arrives. Discuss whether resilience investment is a protected, labelled line, and who would have to approve reducing it.

When did we last exercise our business continuity and emergency plans, and what actually failed? A plan never tested against a realistic scenario is a hypothesis, not a capability. Review the most recent exercise or real incident honestly: what assumption turned out to be wrong, what contact could not be reached, what dependency had not been mapped. If nothing failed, that is itself worth examining — it may mean the exercise was not demanding enough.

How do we make decisions when the evidence is incomplete and the clock is running? Ordinary policy-making can wait for better evidence; crisis leadership cannot. Discuss what authority and default options exist for the moment a team must choose between acting now on partial information and waiting at the cost of time. Agree in advance which decisions are reversible enough to make quickly and which are consequential enough to warrant deliberate delay even under pressure.

Whose voice does the public trust in a crisis, and have we deliberately built that person's capacity to lead? The most senior person on the organisation chart is not automatically the person the public will trust, or the person best suited to the sense-making role crisis leadership demands. Discuss who that person or persons should be, whether they have had real media and crisis-communication training, and whether the organisation has built redundancy so the role does not depend on one individual's availability.

What did we learn from our last significant incident, and can we point to something we changed because of it? Many organisations produce thorough after-action reports that are read once, filed, and never revisited. Discuss the last three recommendations from a past review and trace, specifically, whether each was implemented, partially implemented, or quietly dropped. An organisation that cannot answer this question is not actually learning from experience, however good its written reviews look.

In practice: a public value example

The following scenario is fictional and illustrative. It does not describe any real institution.

The City of Rivermouth, a fictional mid-sized coastal municipality, runs its own municipal water utility serving 340,000 residents. In late autumn, a severe storm system stalls over the region, producing the heaviest rainfall in the utility's 40-year operating record, and river levels exceed the design threshold of the treatment plant's flood defences within six hours. At the same time — unrelated to the storm, but exploiting the distraction it has caused — an intrusion is detected in the utility's supervisory control and data acquisition (SCADA) system, the software that remotely operates pumps and valves, and a substation serving both the plant and a nearby hospital loses power.

The utility's emergency operations plan, built on an all-hazards template with a flood annex and a cyber-incident annex, is activated by the on-call duty manager under authority pre-delegated in writing, without needing to reach the chief executive first. A strategic command group convenes by video within 20 minutes to set objectives — protect drinking water safety and hospital power above all else — while a tactical cell, staffed jointly by utility engineers, the municipal emergency management office, and the regional power operator, manages the moment-to-moment response. Because the cyber annex assumed exactly this kind of compounding failure, the SCADA system is isolated from remote access within 30 minutes, and operators fall back to a manual control procedure exercised twice in the previous 18 months — slower, but not improvised.

The utility's business continuity plan had already identified drinking water treatment and hospital-adjacent power as the two functions whose loss was unacceptable within hours, which had driven the earlier investment in a mobile backup generator contract and the manual fallback procedure now proving decisive. A single trained spokesperson gives hourly public updates stating what is confirmed, what is not, and when the next update will come, which keeps rumour from filling the gap. Within 36 hours, water service is confirmed safe and hospital power is restored via a temporary line; full network stability takes five more days.

The recovery board, reviewing the incident three months later, does not simply rebuild the plant to its original specification. It raises the flood defence to a new threshold informed by revised climate projections, the kind of adaptation this topic leaves to the specialist literature (see Topic 4.3 — Climate, Sustainability, and Environmental Public Value), funds a second, separate SCADA control centre, and requires that future infrastructure upgrades be assessed against cascading, multi-hazard scenarios rather than single-hazard ones alone. None of this spending shows up as a service citizens see day to day; its entire public value lies in the disaster that does not happen next time.

Four sector lenses

Local government

Local government is almost always the first responder and the level of government closest to where a hazard lands, which gives it operational knowledge no higher tier can replicate but also the thinnest fiscal capacity to absorb a major shock alone. Local resilience typically depends on partnerships across neighbouring municipalities, emergency services, utilities, and voluntary organisations, because almost no single local authority holds every capability a serious incident requires in-house. Its genuine challenge is sustaining specialist readiness — trained staff, exercised plans, maintained equipment — inside a workforce and budget structure built around routine service delivery (see Topic 3.6 — Public Sector Workforce and Labour Markets).

National government

National government sets standards, holds strategic reserves that no single locality could justify maintaining alone, and is usually the only actor with authority to declare a national emergency, coordinate a whole-of-government approach across regions and departments, and fund large-scale recovery. Its distinctive risk is the opposite of local government's: a national ministry can produce excellent doctrine that never translates into tested capability where an incident occurs, because implementation depends on lower tiers it does not directly control. Effective national resilience depends heavily on how national and sub-national responsibilities are divided and funded (see Topic 3.9 — Intergovernmental Relations and Federalism).

Nonprofit and community organisations often hold resilience capacity formal government cannot easily replicate: pre-existing trust and reach into populations hardest to serve in a crisis, and the flexibility to improvise delivery — food, shelter, outreach — faster than a procurement-bound public body. Their characteristic vulnerability is funding volatility: donations and emergency grants spike immediately after a visible disaster and dry up during the unglamorous preparedness phase, precisely when planning and volunteer infrastructure most need investment (see Topic 3.8 — Social Sector and Nonprofit Management). Volunteer burnout during protracted or repeated crises is a recurring risk here specifically, because surge capacity depends on goodwill that is not infinitely renewable.

Multilateral and international

Multilateral bodies add value no single national government can supply alone: coordinating cross-border hazards such as pandemics or transboundary flooding, pooling technical expertise smaller states cannot maintain individually, and operating pre-arranged financing mechanisms — contingency funds, catastrophe-linked insurance, pooled reserves — that spread risk across many countries. Their characteristic constraint is sovereignty: an international body can recommend, convene, and fund, but cannot compel a national government to prepare, and its effectiveness depends on national authorities acting on shared standards such as the Sendai Framework for Disaster Risk Reduction, coordinated through the United Nations Office for Disaster Risk Reduction (UNDRR) as the leading global reference for disaster risk reduction.

Common failure modes

The plan as compliance artefact. A written emergency or continuity plan exists, satisfies an audit requirement, and is never rehearsed, updated, or read by the people who would need to use it. The cost surfaces only during a real incident, when contact numbers, assumptions, and authorities turn out to be years out of date.

Unclear or untested command authority. Nobody is confident who may actually declare an emergency, activate mutual aid, or authorise emergency spending, so the first hours of a real incident are spent establishing authority rather than exercising it — spending the one resource a crisis cannot recover: early time.

Single-hazard tunnel vision. Planning and resourcing focus on refighting the last crisis rather than a different or compound one, so an organisation superbly prepared for a flood is blind to the cyberattack that arrives during it. Perrow's account of cascading failure in tightly coupled systems describes exactly this pattern, and it recurs across sectors that never planned for hazards arriving together.

Heroic improvisation mistaken for preparedness. A crisis is handled well because one exceptionally capable individual improvised under pressure, and the organisation concludes the system worked — when it worked despite the system, not because of it. That fragility is invisible until the capable individual is unavailable next time.

The communication vacuum. Official silence in the first hours of an incident, whether from caution or process, is never actually silence — it is filled by rumour and misinformation that is far harder to correct later than it would have been to pre-empt (see Topic 4.4 — Social Media, Misinformation, and Public Communication).

Rebuilding the vulnerability rather than the resilience. Recovery funding restores what existed before an incident rather than reducing the risk of the next one, because "build back better" costs more up front than "build back the same," and recovery timelines rarely reward that patience.

Pro-cyclical resilience cuts. Standing investment in mitigation and standby capacity is the first line cut in a tight budget year precisely because its absence is invisible until a crisis exposes it — so resilience spending falls fastest exactly when fiscal pressure makes future shocks more damaging, not less.

Maturity model

Dimension	Initiate	Develop	Standardise	Manage	Orchestrate
Emergency planning cycle	No documented plan; response is entirely improvised when an incident occurs.	A written plan exists for the most familiar hazard but is rarely reviewed or shared widely.	An all-hazards plan with hazard-specific annexes is documented, adopted, and owned by named roles.	The plan is reviewed on a fixed cycle, tested against performance measures, and updated from exercise and incident findings.	Planning is coordinated across neighbouring organisations and jurisdictions and adapted proactively as hazard patterns change.
Crisis decision-making and command	No defined command structure; authority is assumed informally during an incident.	Command roles are named on paper but have not been tested under pressure.	Strategic, tactical, and operational tiers and escalation triggers are documented and understood by those who hold them.	Command authority is exercised regularly, decision timeliness is measured, and gaps trigger corrective training.	Command structures interoperate smoothly with partner organisations, including across sectors and borders, and adapt tier design as incidents demonstrate need.
Business continuity planning	Critical functions are not identified; disruption is handled case by case.	Some critical functions have informal continuity arrangements, inconsistent across teams.	A continuity plan aligned to a recognised standard identifies critical functions and recovery arrangements organisation-wide.	Continuity arrangements are tested regularly, recovery times are measured against targets, and shortfalls are resourced.	Continuity planning is coordinated with suppliers, partners, and the wider sector so a shared disruption does not cascade across the network.
Resilience investment and financing	Resilience spending is not identified separately from routine budgets.	Some resilience spending occurs but is the first item cut under fiscal pressure.	A protected resilience budget line exists and is defended in the ordinary budget cycle.	Resilience spending is measured against risk reduction achieved and reported to oversight bodies.	Resilience financing is pooled or co-ordinated across jurisdictions or partners, including pre-arranged mechanisms such as contingency funds or risk pooling.
Learning and after-action review	Incidents and exercises are not formally reviewed.	Reviews are written after major incidents but recommendations are rarely tracked to completion.	A structured after-action review process is standard practice with assigned recommendation owners.	Recommendation completion is tracked and reported, and recurring findings trigger systemic changes, not just local fixes.	Lessons are shared proactively across organisations and sectors, contributing to and drawing on a wider body of shared practice.

Checklist

Written, all-hazards emergency plan exists, with hazard-specific annexes, and is owned by a named role.
Command authority — who can declare a crisis and activate the plan — is documented and has been tested outside normal working hours.
Business continuity plan identifies the organisation's critical functions and their maximum tolerable period of disruption.
Mutual aid and surge capacity agreements with neighbouring organisations or jurisdictions are signed and current, not merely discussed.
At least one exercise (tabletop, functional, or full-scale) has been run in the past 12 months and its findings tracked to completion.
A single trusted spokesperson or communication channel is identified, trained, and ready to activate within the first hour of an incident.
Resilience investment appears as an identifiable, protected line in the base budget, not only in disaster or contingency funds.
Recovery planning includes an explicit "build back better" requirement rather than defaulting to like-for-like reconstruction.
After-action reviews follow every significant incident and exercise, with recommendations assigned owners and deadlines.
Responder welfare — rotation, rest, and post-incident support — is built into the plan, not added after people are already exhausted.
Dependencies on suppliers, utilities, and partner organisations are mapped, so a compound or cascading failure is not a surprise.
Cross-references to risk management (see Topic 3.10 — Regulation and Public Risk Management) and, where relevant, climate-specific resilience (see Topic 4.3 — Climate, Sustainability, and Environmental Public Value) have been checked for consistency with this plan.

Key sources

This topic draws on the international emergency management literature's four-phase cycle (prevention/mitigation, preparedness, response, recovery), the United Nations Office for Disaster Risk Reduction's Sendai Framework for Disaster Risk Reduction as the leading global policy reference for disaster risk reduction, the International Organization for Standardization's ISO 22301 as the reference standard for business continuity management systems, the Organisation for Economic Co-operation and Development's risk governance and resilience work, and Charles Perrow's analysis of cascading failure in complex, tightly coupled systems. National examples are drawn from the United States' Federal Emergency Management Agency's National Preparedness Goal and the United Kingdom's Cabinet Office Resilience Framework, cited as illustrations of national patterns rather than as universal models.

References

Emergency management — Wikipedia — https://en.wikipedia.org/wiki/Emergency_management
Business continuity planning — Wikipedia — https://en.wikipedia.org/wiki/Business_continuity_planning
Crisis management — Wikipedia — https://en.wikipedia.org/wiki/Crisis_management
Disaster risk reduction — Wikipedia — https://en.wikipedia.org/wiki/Disaster_risk_reduction
Sendai Framework for Disaster Risk Reduction — Wikipedia — https://en.wikipedia.org/wiki/Sendai_Framework_for_Disaster_Risk_Reduction
Resilience (engineering and construction) — Wikipedia — https://en.wikipedia.org/wiki/Resilience_(engineering_and_construction)
Community resilience — Wikipedia — https://en.wikipedia.org/wiki/Community_resilience
Incident Command System — Wikipedia — https://en.wikipedia.org/wiki/Incident_Command_System
Normal Accidents — Wikipedia — https://en.wikipedia.org/wiki/Normal_Accidents
Public good (economics) — Wikipedia — https://en.wikipedia.org/wiki/Public_good_(economics)
Whole-of-government approach — Wikipedia — https://en.wikipedia.org/wiki/Whole-of-government_approach
What is the Sendai Framework for Disaster Risk Reduction? — United Nations Office for Disaster Risk Reduction — https://www.undrr.org/implementing-sendai-framework/what-sendai-framework
ISO 22301:2019 Security and resilience — Business continuity management systems — International Organization for Standardization — https://www.iso.org/standard/75106.html
Risk governance — OECD — https://www.oecd.org/en/topics/sub-issues/sustainable-and-resilient-infrastructure/risk-governance.html
National Preparedness Goal — Federal Emergency Management Agency (US) — https://www.fema.gov/emergency-managers/national-preparedness/goal
The UK Government Resilience Framework — Cabinet Office (UK) — https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1125536/6.8213_CO_Resilience_Framework_FINAL_1_.pdf