A Seven-Point Checklist for Judging Any Governor's Emergency Powers

Nothing in government lasts longer than a temporary measure. A tax that was "just for the war" outlives the war. A fee that was "just to fix the bridge" outlives the bridge. And an emergency, if nobody sets a clock on it, can last until the governor who declared it has a library named after him.

Now, I am not against emergency powers. When a hurricane is bearing down on the coast, you want somebody who can order an evacuation, open the shelters, wave through the fuel trucks and call out the Guard before the committee finishes its coffee. When a river floods a county, you want a governor who can act at two in the morning. The question was never whether a governor should have power in a crisis. The question is what keeps a crisis power from becoming a permanent one, and whether we hold our own side to the same test we hold the other side to.

That second part is where conservatives need to be honest. If the standard is "I like this governor, so his emergency order is fine, and I dislike that governor, so hers is tyranny," then it is not a standard. It is a team jersey. So here is a standard. It has seven points, it is grouped so you can use it on a napkin, and each point comes with the reason it matters. Run it on any governor of any party, including the one you voted for.

How to use the list

For each of the seven points, give the state two marks for a clear yes, one for a partial yes, and zero for no. Seven points times two marks is a top score of 14. A state that scores 12 or better has a sound emergency system. A state that scores 6 or less is handing its governor a blank check, and the voters should say so. You will not find a perfect state, and that is fine. The point is to know where the gaps are before the next storm, not after.

Nothing here is about medicine or science. This is about who holds power, for how long, and who can say stop.

Group one: where the power comes from

1. Is the authority written down, in a statute the legislature passed?

A governor's emergency power should come from a specific law, passed by the legislature, that says what can be done and when. It should not come from a vague claim that the governor is "the chief executive" and the situation is serious.

Why it matters: in our system the legislature writes the rules and the executive carries them out. When an emergency statute is vague or enormous, the legislature has quietly handed over its job. The Michigan Supreme Court said exactly that in October 2020, in a case about Governor Gretchen Whitmer's orders. The court held that a 1945 law, which let the governor declare emergencies and issue orders with the force of law for as long as the governor judged necessary, was an unconstitutional delegation of legislative power to the executive under the state constitution. Whatever you thought of the orders, the court's point was structural: a statute that lets one person make binding rules, with no standard and no end date, is the legislature abdicating.

Check: read your state's emergency management statute, which is on the legislature's website. Does it list what the governor may do? Or does it say "any measure necessary"?

2. Is the power limited to a real emergency, with a definition?

A good statute defines what counts. A hurricane, a flood, a wildfire, an attack, a collapse of a public system. A bad one lets the governor declare an emergency about anything the governor says is one.

Why it matters: a definition is a fence. If every problem is an emergency, the fence has no gate. The same governor who handles a flood honestly can, in the next term, find an emergency in an issue the legislature declined to act on. A fair definition protects the next governor from temptation, including the one you like.

Group two: the clock and the check

3. Is there a time limit, and does the legislature control the extension?

This is the single most important point on the list, and the one most states get wrong. The best systems say an emergency declaration lasts a set number of days and can be extended only if the legislature agrees.

Look at what two states did. Michigan's Emergency Management Act of 1976, which the state's Supreme Court said was the only valid source of authority in the 2020 case, limited an emergency to 28 days unless the legislature approved an extension. The governor's view was that she could keep declaring new emergencies under the 1945 law; the court rejected that reading. And in Pennsylvania, voters in May 2021 approved constitutional amendments that capped a governor's disaster declaration at 21 days unless the General Assembly extended it by a concurrent resolution, and that let the General Assembly end a disaster declaration by concurrent resolution without the governor's signature. In other words, the people of Pennsylvania wrote a clock into their constitution.

Compare the federal system. The National Emergencies Act of 1976 says a presidential emergency expires on its anniversary unless the President renews it with a published notice. That is a clock, though a weak one, because Congress's power to end an emergency by itself was undermined after the Supreme Court's 1983 decision in INS v. Chadha, which struck down the legislative veto. A termination resolution now needs the President's signature or a two-thirds override. A good state law is stronger than that.

Why it matters: a clock forces a conversation. At day 21 or day 28 the governor has to walk into the legislature and explain why the emergency should continue. If the answer is good, the legislature says yes. If the answer is thin, it says no. Either way, the people's representatives, not one office, decide.

Check: how many days can a declaration last? Who has to approve the extension? Can the legislature end it alone?

4. Can the legislature convene itself and end the emergency on its own?

This is a sibling of point three. A clock does no good if the legislature cannot act when the clock runs out. In some states, only the governor can call a special session, which means the person whose emergency is being reviewed controls whether the reviewers meet. Where that is so, ask whether the legislature can call itself into session, and whether its resolution ending the emergency needs the governor's signature, which is a veto in disguise.

Why it matters: the check has to be a check you can use. If the only way to end an emergency is for the governor to agree, you have not checked anything.

Group three: the referee

5. Is there prompt judicial review?

An emergency order is law, and law can be challenged. The courts should hear the challenge quickly, and the standard of review should not be "trust the governor." A challenger should be able to get into court fast, with an expedited schedule, because an emergency order that expires before it can be reviewed is an order no court ever touches.

Why it matters: the legislature is the check on duration; the courts are the check on legality. The Michigan case went all the way to the state's highest court in a matter of months. A federal example from the same period: in November 2020, in Roman Catholic Diocese of Brooklyn v. Cuomo, the U.S. Supreme Court enjoined New York's numerical limits on attendance at houses of worship, because the state was treating comparable secular gatherings more leniently. The Court acted within weeks of the filing. Good review is fast review.

Check: does your state have a rule for expedited appeals of emergency orders? Does the statute say how the courts should examine them?

Group four: the edges of the order

6. Does the order stay within its scope, and does it protect property and speech?

Emergency orders should be tied to the emergency. A hurricane evacuation order has a reason to move people; it has no reason to restrict what a newspaper may print or what a preacher may say. A fuel allocation order has a reason to touch trucks and pumps; it has no reason to seize a family's home without compensation. The order should be as narrow as the emergency, ending where the emergency ends.

Why it matters: the Bill of Rights is not suspended in a flood. The First Amendment protects speech, worship and assembly, and the Fifth Amendment says private property shall not be taken for public use without just compensation. Your state constitution almost certainly has similar clauses. If an emergency order touches property or liberty, the statute should say how a person gets paid or heard. A good law includes a compensation rule for seized property and a bar on using emergency power to regulate speech or worship beyond what the courts have always allowed for safety.

Check: does the statute address compensation for property used or taken? Does it say anything about speech and assembly? If it is silent, the Constitution still applies, but you want it in writing.

7. Does the order have a sunset built in?

The last point is a cousin of the clock, but it works at the level of each order. Every emergency order, not just the declaration itself, should say when it expires and what must be shown to renew it. A sunset on each rule keeps the emergency from piling up like an attic.

Why it matters: declarations end, but the orders issued under them have a way of surviving. A state that requires each order to carry its own end date and to be reviewed at each renewal has the surest guard against drift.

Check: does a rule issued under emergency authority lapse when the emergency does? Can the governor convert an emergency order into a permanent regulation without the ordinary rulemaking process?

Worked example: scoring two imaginary states

Let's use two hypothetical states so nobody's feelings get hurt.

State A has a clear statute listing the governor's powers (2), defines "disaster" as storms, floods, fires and similar events (2), caps a declaration at 30 days with legislative approval needed to extend (2), lets the legislature meet on the call of its own leaders and end an emergency by resolution without the governor (2), provides expedited review in the state supreme court (2), has a just-compensation clause and a bar on speech limits (2), and says each order lapses with the declaration (1, because renewals are routine). Score: 2 + 2 + 2 + 2 + 2 + 2 + 1 = 13. A good system.

State B has a short statute that says the governor may take "such measures as the governor deems necessary" (0), defines nothing (0), has no cap on duration (0), lets only the governor call a special session (0), has ordinary court review with no expedite rule (1), says nothing about property or speech, though the constitution applies (1), and has no sunset on orders (0). Score: 0 + 0 + 0 + 0 + 1 + 1 + 0 = 2. That state is one bad year away from a crisis of confidence.

Notice something. You can score State B without knowing the party of its governor. That is the point. The list does not care about the jersey.

The best objection

Here is the strongest argument against all this, stated fairly. Emergencies are, by nature, unpredictable. A rule that limits a governor to 21 or 28 days may be too short for a disaster that takes months to recover from. A legislature that has to approve every extension may be slow, divided or out of town. In the worst case, a governor who must ask permission to continue an evacuation or a rebuilding order may find the answer is "no" for reasons of politics, not safety. A flexible system, the argument goes, is better than a rigid one, because disasters do not read statutes.

That deserves a serious answer, and here it is. First, the list does not say the answer must be "no" at day 21; it says the answer must come from the legislature. A hurricane recovery that needs 90 days gets extended, and a legislature that refuses a real extension answers to voters for it. Second, the rule can be built to fit: a short initial clock with renewals, and a fast-track procedure, such as an emergency joint committee, for continuing the orders that matter. Third, the real danger is not the governor who handles the storm. It is the declaration that outlives the storm and becomes the vehicle for an unrelated agenda. A clock fixes that without hurting the first case.

What to do next

Print this list. Take it to your state representative, or to a county party meeting, or to the next town hall, and ask for the answer to each point for your state. Do not make it partisan. If the governor in your state is a Republican, the test is more convincing when you say, "I'm asking this about our own governor, too." That is how a movement earns credibility.

Related reading in this library: What a Governor Can Actually Do and What Only the Legislature Can explains the office, State Constitutions Versus the U.S. Constitution, Side by Side shows why your state's text matters, How a Bill Becomes State Law, From Idea to the Governor's Desk covers how to get a fix passed, and Madison Built a Government Where Ambition Fights Ambition, and It Works gives the philosophy behind a check on every branch. For the same logic at the federal level, see Justice Jackson's Three-Box Test for Presidential Power, Put to Work, and for the limits on delegating lawmaking power, Congress Should Write the Laws, and Other Radical Ideas From Article I.

The bottom line

• Emergency powers are a necessary tool. They become a danger when they have no end date, no check by the legislature and no fast route to a court.
• Judge any governor by the same seven points: authority in a statute, a definition of emergency, a time limit, a legislature that can end it, quick court review, scope and rights protections, and a sunset on each order.
• Michigan's 28-day rule and Pennsylvania's 21-day constitutional limit show that a clock can be built, and that voters and courts will back it.
• Apply the test to the governors you like, because a standard that only binds the other side is not a standard.
• Take the list to your own legislator and ask for the score of your state. Then ask what it would take to fix the zeros.