Emergency Powers: Why Temporary Power Tends to Stay

A crisis is not the same as a formal state of emergency

The first important distinction is both legal and practical. Not every major disaster or public-health threat requires a formal state of emergency. States already possess many powers under ordinary law to protect life, public health, safety and infrastructure. A formal emergency regime becomes justified only when ordinary legal tools are insufficient for a genuinely exceptional threat.

Slovenia's constitutional framework illustrates the distinction clearly. The Constitution links a formal state of emergency to a great and general danger threatening the existence of the state; as a rule, the National Assembly decides on the declaration, urgent measures and their termination. If the Assembly cannot meet, the President may act temporarily, but those decisions must be submitted for confirmation as soon as the legislature reconvenes.

For THY-REALITY the principle is straightforward: the existence of a crisis is not itself permission for an unlimited exception. The first question is whether the goal can still be achieved through ordinary, narrower and already supervised powers. Emergency government should be the last escalation, not the first reflex.

“How Do We Make Decisions Without a Permanent Ruler?” already established a basic principle: in a dangerous situation we can pre-authorise someone to act quickly without granting a permanent right to rule. This article extends that principle. A sound emergency power is functional: tied to a particular threat, a particular task and a particular time window.

That means the boundaries should be as clear as possible. If the purpose is evacuation from an endangered area, that does not imply general authority over all aspects of public life. If the purpose is temporary stabilisation of supply, that does not create a right to redesign ownership, political processes or the information order on a permanent basis.

Speed is a real value during a crisis. But speed and breadth of authority are not the same thing. An institution can act very quickly while remaining narrowly bounded. That combination is safer than a model in which urgency automatically opens the door to every measure later described as useful.

Emergency powers are an exception within law, not a legal vacuum

International standards begin from a similar core. The ICCPR allows temporary derogations from some obligations during a grave public emergency, but requires such measures to remain exceptional, temporary and strictly limited to what the situation requires. The European Convention on Human Rights uses a similar test: derogation is permitted only to the extent strictly required by the exigencies of the situation, while some rights remain specially protected.

Slovenia's Constitution likewise does not create a zone without rules. Temporary limitations of rights are tied to the duration of war or emergency and to the extent required by that situation; some rights cannot be suspended even then. This is an important boundary between an exception inside law and the idea that crisis suspends the rule of law itself.

The project's rule is therefore simple: if a measure cannot explain its legal basis, necessity, proportionality and connection to the concrete danger, the fact that circumstances are stressful is not by itself a sufficient argument.

Emergency authority without a foreseeable ending has a weak brake. The Venice Commission therefore stresses that a state of emergency should be declared for a specified period and that unlimited declarations are not acceptable. Extension can be legitimate when the danger genuinely continues, but the longer the exception lasts, the stronger the need to show why ordinary legal tools are still insufficient.

The phrase “for as long as necessary” sounds reasonable, but it hides the central question: who decides that it is still necessary, by what criteria, and who can review that conclusion? If the same body declares the threat, defines the measures, evaluates their success and decides when they end, it has a strong institutional advantage in validating its own assessment.

A better model therefore specifies a review or expiry date at the beginning, the criteria for renewal and the body that must authorise the extension again. The exception then continues because of a fresh positive decision, not because of inertia.

Why temporary power sometimes tends to remain

The title of this article does not imply that every emergency power inevitably becomes permanent. Comparative research finds both a ratchet effect, where exceptional authority becomes normalised, and a rollback effect, where authority is genuinely returned after the crisis. The key question is which institutional conditions push the system toward one outcome or the other.

Temporary power can persist for several understandable reasons. Crises create new offices, contracts, databases, technical systems and career responsibilities. People assigned new tasks acquire expertise and an organisational interest in continuing them. Political leaders may also face greater blame for withdrawing a measure too early if the danger returns than for extending it somewhat too long.

Malice is therefore not required. Institutional inertia can preserve power without a conspiracy. That is why sound design should not rely only on the goodwill of officeholders; it should build an automatic mechanism of contraction into the emergency architecture itself.

A sunset clause does one crucial thing: it changes the default. Without it, a measure remains until someone actively removes it. With it, the measure ends unless someone actively demonstrates why it should continue.

But an expiry date alone is not enough. If a measure is automatically renewed every thirty days without meaningful discussion, the sunset becomes a calendar ritual. Its purpose is reauthorisation: a fresh assessment of the threat, a new justification, review of results and harms, the possibility of narrowing the scope, and a clear vote or other accountable decision.

OECD work on emergency regulation emphasises temporary measures, sunset provisions and post-implementation review in combination. This article turns that into a general rule: at renewal, the burden should reset. Citizens should not have to prove why exceptional power should be removed; the holder of that power should show why it is still needed.

Oversight must not disappear precisely when it is needed most

Crisis often increases the executive's information advantage. It gains faster access to experts, security information, logistics and operational reporting. That may be necessary for an effective response, but it also deepens the principal-agent problem from “When Does a Delegate Become a Ruler?”: outsiders have a harder time knowing whether the breadth of the measures remains justified.

A sound system therefore preserves several channels of oversight. Parliament or an assembly reviews the mandate and extensions, courts review the legality of individual measures, independent oversight bodies inspect records and implementation, while opposition actors and the public receive enough information to challenge the official account without compromising genuinely necessary operational secrecy.

“Who Guards the Guardians?” already showed that we do not need one supreme guardian. The same is true in a crisis: oversight should remain layered and should not collapse into the same centre that holds the emergency power. If urgency requires faster review, review can be accelerated; it does not have to be abolished.

Emergency authority must not quietly change its purpose

A major risk in long crises is scope creep: a measure introduced for one task begins serving another. A system built for infection tracing may later appear useful for different forms of monitoring; an emergency procurement channel may become a convenient way around ordinary competition; a temporary restriction on assembly may become a broader tool for managing political space.

This does not mean every later use of a technical capability is wrong. It means that a new purpose requires a new legal and political justification. The original crisis should not function as a permanent licence for future, unrelated objectives.

A sound emergency framework therefore defines not only a deadline but also a purpose. If the purpose changes, the old mandate ends and the new measure must pass through the ordinary process or a newly explicit authorisation.

A distributed emergency system can be fast if it is prepared in advance

The strongest argument for centralisation during crisis is often speed. But “When the System Fails: Local Preparedness for Disasters and Major Disruptions” showed that much of real speed comes not from central command but from pre-event preparedness: known contacts, supplies, redundancy, exercises, interoperable standards and clear responsibilities.

A distributed emergency system therefore does not mean everyone acts alone. It means local units have pre-authorised tasks and resources, while wider levels enter when the problem exceeds local capacity. Escalation boundaries are known in advance: when broader support activates, who coordinates shared resources, what can be done without new approval, and when the authority returns downward.

This model combines subsidiarity with urgency. Local actors do not wait for permission for every basic protective action, while wider coordination does not absorb all authority simply because it was needed for one task. Distributed preparedness reduces the amount of power that must be centralised in haste during a crisis.

After the crisis, review the measures — do not merely celebrate the ending

When danger subsides, organisations often want to move on quickly. Yet that is exactly when retrospective review becomes most valuable. Which powers were truly necessary? Which did not help? Which produced side effects? Where was action too slow and where too broad? Which databases, contracts, registers or temporary bodies still exist, and why?

Post-crisis review is not a hunt for blame. It is institutional memory. Without it, the next crisis begins with the same improvisation and with the tools left behind by the previous one. OECD guidance treats post-implementation review as especially important when normal ex ante assessment was shortened or bypassed because of urgency.

This article therefore requires an exit plan as well: who closes the temporary body, who terminates the power, who deletes or returns data, who winds down contracts, and who publicly confirms that ordinary governance has actually been restored.

Practical audit: is the emergency power still exceptional?

Before introducing, renewing or retaining emergency authority, a community can ask twelve questions:

1. Does the threat genuinely exceed the capacity of ordinary legal and operational mechanisms?
2. Is the power tied to a clearly defined purpose rather than the generic word “crisis”?
3. Which authorities are necessary and which are merely convenient?
4. When does the power automatically expire?
5. What evidence must be presented for renewal?
6. Who can independently verify that the conditions for the exception still exist?
7. Do the legislature, courts or other oversight bodies have real access to the necessary information?
8. Can the measure be used for a purpose that was not part of the original authorisation?
9. Which tasks can remain local and which genuinely require wider coordination?
10. Is there a clear process for returning authority to ordinary levels?
11. What happens after the crisis to temporary data, contracts, bodies and technical infrastructure?
12. Who will conduct a public after-action review and publish what should change before the next crisis?

If several of these questions have no clear answer, emergency authority is no longer merely a crisis-management tool. It is becoming an institutional problem in its own right.

This article has shown that crisis centralisation is not necessarily an abuse. It can be rational, necessary and life-saving. But it has also shown why the ending must be built into the beginning. Without that architecture, a temporary apparatus can survive the problem that created it.

The next question is therefore broader than emergencies. What happens when an organisation — emergency or ordinary — begins defending its own budget, powers, personnel and status even when doing so is no longer the same as serving the community? “Institutional Capture: When the System Starts Serving Itself” will examine institutional capture: the point at which a system created for a function begins reshaping the function around itself.

The common thread from “Monopoly, Plutocracy and the Concentration of Economic Power” to “Institutional Capture: When the System Starts Serving Itself” remains the same: distributed power does not remain distributed by itself. It requires boundaries, feedback loops, independent verification and a genuine ability for exceptional or delegated authority to end.