Checks, Balances, and Emergencies: Tensions Between Emergency Management Acts and Constitutional Governance

The authors of this paper assess how Emergency Management Acts have been used during the COVID-19 pandemic and offer suggestions for how they can be adjusted to better prepare state and local governments for future emergencies.


Braden Boucek
Daniel Greenberg
Kimberly Hermann
Mithun Mansinghani
Clark Neily
Jon Riches
Luke A. Wake
Shoshana Weissmann


The Federalist Society and Regulatory Transparency Project take no position on particular legal or public policy matters. This paper was the work of multiple authors, and no assumption should be made that any or all of the views expressed are held by any individual author except where stated. The views expressed are those of the authors in their personal capacities and not in their official or professional capacities.
To cite this paper: B. Boucek, et al., “Checks, Balances, and Emergencies: Tensions Between Emergency Management Acts and Constitutional Governance,” released by the Regulatory Transparency Project of the Federalist Society, November 17, 2021 (


A fire rages through the city. Home after home burns as the flames spread among the wooden structures of the town’s historic district. People trapped inside need rescue, while those who have escaped look on as their life’s savings go up in smoke. But the catastrophe’s expansion can be stopped if homes at the edges of the inferno are torn down, so that some untouched structures can be saved from igniting. Yet how does the government have the power to destroy citizens’ houses and all their attendant possessions? Or to restrict people’s liberty to travel into the dangerous zone? And who gets to decide such things: must the state legislature meet to introduce a new law, debate it, amend it, and then cast votes? Or can the chief executive make such decisions quickly and unilaterally?

Over the years, governments have accumulated greater and greater powers to restrict and impinge on individual rights in order to protect public health and safety during emergencies. The emergencies giving rise to such authority are diverse: fire, flood, tornado, hurricane, war, riot, earthquake, terrorism, and (perhaps freshest in mind) epidemic. And from the September 11 terrorist attacks and Hurricane Katrina to racial justice protests and the COVID-19 pandemic, we have witnessed remarkable exercises in emergency power.

As the U.S. Supreme Court has reasoned, “the common law had long recognized that in times of imminent peril – such as when fire threatened a whole community – the sovereign could, with immunity, destroy the property of a few that the property of many and the lives of many more could be saved.”1 So too in war, “[w]hatever would embarrass or impede the advance of the enemy, as the breaking up of roads, or the burning of bridges, or would cripple and defeat him, as destroying his means of subsistence, were lawfully ordered by the commanding general.”2 Rural areas may be intentionally flooded to avoid a levee breach that would be catastrophic to cities.3 A factory can be destroyed “because it was thought that the structure housed the germs of a contagious disease.”4

Such extraordinary deprivation of liberty during emergency goes beyond property rights. Lincoln famously suspended the writ of habeas corpus in the midst of the Civil War, asking rhetorically, “are all the laws, but one, to go unexecuted, and the government itself go to pieces, lest that one be violated?” The essential liberty of movement is sometimes restricted by the state: consider forced evacuations from an area about to be slammed by a hurricane or the quarantine of a nurse who had recently treated Ebola patients.5 Courts upheld restrictions on large assemblies, though otherwise protected by the First Amendment, during the Spanish Flu.6 Citizens’ bodily integrity has been violated by mandatory vaccine requirements during an epidemic.7

Emergencies are often used as justifications to sidestep the rule of law in less dramatic fashion, too. State regulations on procuring goods and services, which often require lengthy bidding and approval processes, can be suspended to make rapid purchases that are useful in an emergency, such as buying sandbags during a flood, repairing needed infrastructure after a storm, or obtaining personal protective equipment from disease. Licensing requirements for needed personnel can also be suspended, including those that govern healthcare workers treating the injured or diseased or firefighters arriving from out of state.8

Emergencies raise questions not only about the scope of state power in extraordinary situations, but also who gets to exercise it. Usually, the locus of state power is with the state legislature, with the doctrine of the separation of powers giving the executive only the power to enforce the substantive rules created by the legislative branch. But in emergencies, the legislative process is often considered too slow to make quick decisions. State legislatures also meet infrequently, often only a few months a year or every two years.9 So emergency power often devolves to the executive. Indeed, Alexander Hamilton, in Federalist No. 70, famously extolled, as a principal virtue of the unitary executive, its ability to act with “energy,” “activity,” and “despatch.”

To address such emergency situations, all states have enacted some form of an Emergency Management Act (“EMA”). In general, state EMAs grant additional authority to the executive branch, usually the state’s governor, upon the governor’s declaration of an emergency. The governor’s ability to declare an emergency is usually without check from the legislature,10 though a few states give the legislature some limiting power.11 The emergency declaration often lasts as long as the governor says, but the legislature may also retain the power to terminate the emergency.12 However, the legislature’s authority to terminate is usually all-or-nothing: it cannot disapprove some but not all gubernatorial acts and therefore must either terminate the emergency altogether or allow the governor to continue to exercise full emergency powers. In some states, the legislature does not have even this limited termination power.13 The governor’s power under these laws is often very broad. Enacted EMAs typically confer unlimited discretion to direct state resources and personnel, power to command or restrict the liberty of movement (e.g., evacuations or quarantines), and even the ability to suspend statutes and regulations or to effectively create new ones by emergency order.14 And further, these EMAs sometimes purport to confer all police powers of the State, or otherwise contain catch-all provisions, allowing the governor to exercise all “necessary” powers to protect the population.15

We begin, in Section I, with an overview of existing EMA statutes, observing that many of these statutes provide loose standards for what qualifies as an emergency – and arguing that while some degree of flexibility in defining an emergency may be necessary, statutory definitions should not be construed so amorphously as to invite the executive branch to exercise emergency powers to address matters of social or economic importance when the legislature has had time to consider them. Section II observes that during the coronavirus crisis, the lax nature of existing EMA standards became increasingly evident, and that the exercise of gubernatorial emergency powers was without precedent in American history. Section III then compares two types of EMA statutes – which, respectively, provide “blank check” authorizations and enumerated powers – finding that, although in theory a structure of enumerated powers cabins the authority of the state’s chief executive, in practice the enumerated powers approach can sometimes be almost functionally indistinguishable from a blank-check approach; this is so because the enumerated powers of EMA statutes are sometimes carelessly drafted, broadly construed, or both.

Notably, as illustrated in Section IV, whether the role assigned to the chief executive in an emergency is limited (as under a scheme of enumerated powers) or essentially limitless (as in under a “blank check” scheme), recent history is replete with instances of emergency-related gubernatorial powers being exercised in an arbitrary – and often destructive – fashion. Accordingly, we propose some procedural safeguards to protect individual liberty here; however, such procedural restrictions are no substitute for real political accountability. As such, in Section V the authors argue that temporal limitations on emergency governance are the most effective way of bridling the exercise of political power and increasing political accountability.

This paper next presents several related policy options intended to ensure accountability, preserve constitutional norms, and foster more reasoned decision-making. Section VI argues that courts should be more engaged in the project of protecting individual rights, even during an emergency. Section VII urges states to consider providing additional process for aggrieved individuals or businesses to obtain clarity as what is being required or to get quick decisions as to the propriety of imposed restrictions. Finally, Section VIII concludes with a discussion of the special importance of honest, serious, and open discussion of cost-benefit analysis in emergency management planning – and even in the midst of crisis.

What Constitutes an Emergency?

The first question we must ask in considering whether and what sort of emergency powers we should vest in the executive branch is: what is an emergency? No one questions that the impending hurricane or fire is an emergency. But there is potential for abuse when the chief executive is authorized to invoke emergency powers under a vague and open-ended definition. Such vagueness can open the door to political mischief: for instance, the executive may wish to declare emergencies not merely to impose order on the public, but also for the sake of assuming greater control over the public purse.

Under most states’ EMA statutes, there is an unclear line between an “emergency,” where a legislative response is not practically possible, and a “political priority,” that legislators are perfectly capable of addressing but won’t for conventional reasons. Policymakers should distinguish the two. They should make clear that the latter cannot be an emergency that would justify bypassing the ordinary legislative process.

We must ensure that an emergency declaration can only be employed to address those threats to public health and safety that truly require immediate and decisive action and for which the legislature cannot address on its own accord in a timely manner. Even if we are going to have a flexible definition of emergency, there must still be a limitation rooted in this principle.

A. The Problem of Defining an Emergency

Let us begin with two states’ definitions in order to illustrate the potential for abuse. In Texas, a disaster is an “occurrence or imminent threat of widespread or severe damage, injury, or loss of life or property resulting from any natural or man-made cause ….”16 In New York, a disaster is an “occurrence or imminent threat of widespread or severe damage, injury, or loss of life or property resulting from any natural or man-made causes ….”17 These definitions do not clearly distinguish between two classes of events: those which, by their nature, require immediate executive action and those which, by their nature, could realistically be addressed by a deliberative lawmaking body.

The time to consider the scope of emergency powers acts is before the asserted emergency. If an ambitious executive sees an opportunity in an emergency to achieve signature accomplishments that would otherwise remain stubbornly out of reach owing to political realities, it will be too late to foil this ambition once the declaration is made.

The prospect of achieving policy outcomes through emergency powers that would never be achievable through the democratic process is tantalizing for some. Consider:

  • Climate change: Senate Majority Leader Chuck Schumer has called on President Biden to declare climate change to be an emergency. If he heeds Senator Schumer’s advice, then the decision will be difficult to undo absent a change in leadership. National emergency declarations are generally nonjusticiable because they involve “political questions” that “revolve around policy choices and value determinations constitutionally committed for resolution to the halls of Congress or the confines of the executive branch.”18
  • Immigration: Under an emergency declaration, President Trump was able to divert over $6 billion dollars never allocated by Congress to fund construction of a border wall. As if to illustrate the preceding point, courts ruled that there were no ‘judicially discoverable and manageable standards’ to help the Court determine whether the situation at the border is a ‘national emergency.’”19 This issue is alive at the state level as well. The governors of Arizona and Texas issued disaster declarations in response to surging waves of illegal immigration.
  • Pornography: There is a burgeoning effort to declare pornography to be a public health crisis. As of 2019, 16 states had passed resolutions declaring it to be so. Although they are non-binding legislative resolutions, it is not difficult to imagine a chief executive deeming pornography an emergency.
  • Racism: In June of 2020, in the midst of the pandemic crisis, the American Academy of Family Physicians called on the White House to declare racism a public health emergency.20 According to Pew Research, 20 states and cities have declared racism to be a public health crisis. Do the executives of these cities have the power to act free of legislative restraint until “racism” ceases?
  • Gun violence: A mere two weeks after New York Governor Andrew Cuomo repealed his declaration of an emergency relative to COVID-19, he declared gun violence to be a “public health crisis” and declared a state of emergency. Governor Cuomo’s utilization of emergency powers allowed him to direct over $138 million on crime intervention programs and job creation, all without legislative input.21 Perhaps such spending is wise. Perhaps treating gun violence through a public health issue is an innovative approach to an intractable problem. That is not the point. Directing funds from the public fisc and recasting gun violence as a matter of public health are both quintessentially legislative functions. One can only conclude that such policy does not enjoy adequate support, otherwise the legislature would have enacted it.

B. Policymakers Must Reexamine the Definition of an Emergency

Guidance about what circumstances may justifiably allow a chief executive to assume emergency powers is crucial. The executive should only be able to assume unilateral control if that choice is the only way to ensure an effective public response to an otherwise insoluble problem. Put another way, unilateral executive control must be confined to situations that the legislature is incapable of addressing, not just unwilling.

Therefore, the definition must, above all else, make clear that circumstances only rise to the level of an “emergency” when it is impossible to address them through the ordinary legislative process. Although there is likely a need for some degree of flexibility in the definition of an emergency, legislators should consider providing a more legible standard that clarifies the conditions that necessitate a temporary replacement for their policymaking role.

The most effective way to curb the potential for abuse would be to specify what is not an emergency. Any definition should expressly state that the governor may not declare an emergency to respond to social, economic, or other politically divisive issues which the legislature has had time to consider.

…there is potential for abuse when the chief executive is authorized to invoke emergency powers under a vague and open-ended definition.

No Analog in American History

The executive branch response to COVID-19 was unlike anything we had ever seen before. Whereas in the past governors have exercised emergency powers delegated from the legislature by statute for short periods and in a much more limited fashion, in 2020 we saw the governors of many states asserting power to micromanage every aspect of their state economy for a prolonged time. Never before had a governor sought to shut down entire industries. To be sure, during the influenza pandemic of 1918 there were legislatively imposed business closure or restriction orders – but nothing like the sort of emergency orders Americans faced during the COVID-19 pandemic.

It is true that executive branch actors had historically exercised quarantine powers pursuant to enacted statutes. But the power to quarantine did not authorize industry-wide closure orders. Instead, that power authorized restrictions to prevent the spread of communicable disease only in cases where there was probable cause to believe that a specific person or place had been infected. And quarantine restrictions lasted only so long as was reasonably necessary to prevent the spread of the disease in such a case – meaning that the delegated quarantine power provided no historical analog for imposing restrictions on businesses preemptively, much less for continuing such restrictions for months on end.

Indeed, the far-reaching and continued exercise of emergency powers in 2020 – which now stretches into 2021 – represents a truly unprecedented event in American history. As such, it is little wonder that “[t]he pandemic has resulted in previously unimaginable restrictions on individual liberty.” Indeed, the Founding Fathers warned that liberty would be imperiled where the executive is permitted to wield legislative powers.

Of course, one might find historical analogs from other societies where an individual has wielded far-reaching emergency powers. Recall that the Roman Senate would appoint a “dictator” in times of crisis, wielding total power – but never for more than six months, as it was understood that it was especially dangerous to give such temporally unlimited power to any man. And there have been other examples in history, as when the French gave Napoleon Bonaparte authority as “First Consul” to rescue the nation during a time of military emergency and financial peril.22

But the exercise of that sort of autocratic power has no place in the American legal tradition. Indeed, one-man rule is fundamentally contrary to the republican ideals that underpin the U.S. Constitution and state constitutions as well. Our federal and state constitutions alike rest on the premises that it is essential to divide power between the legislative and executive branches, and that this separation of powers provides an essential structural safeguard for preserving individual liberty.

For this reason, the states vest all lawmaking powers in their elected state legislatures, just as all lawmaking power is vested in Congress at the federal level. This rests on the notion that the legislature is the most democratic organ of government, representing the diverse interests of the state while accountable to their local electorate. This separation of powers also protects liberty by ensuring that the law will not change absent broad-based social consensus.

In our legal tradition, the executive has no power to make law. Absent some express grant of emergency power in the state constitution, governors have no inherent emergency power to make rules either. Yet as explained in Section III, the governors were able to stretch their conferred authority to justify all sorts of emergency restrictions. And with only one notable exception, the governors succeeded in defending broad conferrals of emergency powers – based on the view that the legislature had decided to allow the executive to take a flexible approach in responding to emergencies generally.23 Accordingly, the extraordinary emergency orders imposed through 2020 and 2021 have largely survived constitutional challenge – even as they stand in tension with the precept that the legislature should be making fundamental policy decisions.

…the far-reaching and continued exercise of emergency powers in 2020 – which now stretches into 2021 – represents a truly unprecedented event in American history.

Open-Ended Conferrals of Emergency Power

At the very outset of the COVID-19 pandemic, when governors across the country were shutting down all “non-essential” businesses, few people seemed to question their authority. But as days turned into weeks and then months, more and more Americans began to question the legality of one-man emergency rule. To be sure, the constitutional norm in the United States has always been that laws restricting individual liberty should come from legislative bodies, not from the executive. So, through the course of 2020 and into 2021, at least three major questions loomed large – and were the subject of many lawsuits:

  • What is the legal basis for emergency orders?
  • Are there limits on what a governor can do by emergency order?
  • And for how long can a governor exercise such extraordinary powers?

A. Legal Basis for Emergency Orders

If you look at emergency orders from 2020, you will typically see the state governor invoking emergency powers vested by statute – as opposed to some inherent power vested in their office. That is because state constitutions generally vest governors only with the power to execute law as enacted by the legislature and vest the power to make law exclusively in the legislative branch. To be sure, some state constitutions confer very limited emergency powers to the governor. For example, the Florida Constitution vests the governor with power to move the locus of government during certain catastrophic emergencies, such as an invasion or nuclear disaster.24

But even in the most extraordinary times, the governor’s emergency powers are usually limited to those powers conferred in an Emergency Management Act, which spells out both when a governor can exercise emergency powers and the scope of those powers. Accordingly, the scope and duration of emergency powers differs state by state. As is often the case in the law, one must begin by looking at the language of the statute in question.

B. Two Models of Emergency Management: Blank-Check Authorizations or Enumerated Powers

With few exceptions, state governors exercised their emergency powers to close down all non-essential businesses, to impose mandatory stay-at-home orders on non-essential workers, to close down places of religious worship, and to impose various other restrictions on individual liberty in 2020 – with some extending those restrictions into 2021. Governors invoked emergency orders not just to restrict business operations and to impose limitations on the right to assemble in places of public accommodation, but also to impose eviction moratoria and to issue other orders with only tangential connections to limiting the spread of COVID-19 – for instance, Governor Newsom’s emergency order that temporarily changed workers’ compensation standards. With few exceptions, these emergency orders were upheld.

In those cases where courts upheld emergency orders, they often cited a 1905 U.S. Supreme Court case, which affirmed that it was within the police powers of the state to require individuals to be vaccinated to contain contagious disease. But that case, Jacobson v. Massachusetts, concerned a municipal order enacted pursuant to express legislative authorization in Massachusetts. So Jacobson really was of limited probative value in deciding whether the governor had constitutional authority to issue emergency orders on broad-ranging subjects. That case stands for the proposition that it is generally within the power of the State to enact law to protect the public health. But while we might assume that the legislature could enact all sorts of laws to protect the public health, the question presented in COVID-19 cases was whether such powers could be wielded by the executive branch.

As suggested above, the answer really depends on what powers the legislature delegated to the governor by statute. In some states the delegation is remarkably simple, and yet astoundingly broad on its face. For example, the emergency management acts in Arizona and California expressly grant their governors “all police powers of the State,” which would seem to confer a power literally coextensive with the state legislature’s power to make law.25 To be sure, when a legislative body enacts new law to regulate private conduct, it is exercising “police powers.”

So in these states the only substantive limitations on the governor’s power to issue emergency orders are those constitutional rules that restrict the legislature’s prerogative to regulate employment matters, occupational licensing, zoning, or anything else the State may be interested in regulating in the name of public health, safety, or morals. And because the general rule is that legislation will be upheld so long as there is any conceivable rational basis, there is virtually nothing that the governor could not regulate by emergency order.26

By contrast, other states enumerate the governor’s emergency powers. In vesting the governor with only specific powers, these emergency management acts at least conceptually limit the scope of the governor’s emergency powers to issuing orders on specific subjects. For example, the statute might authorize the governor to force an evacuation from a disaster area, restrict movement within an emergency area, and to commandeer private property during an emergency; however, delegation of those powers could not justify an emergency order imposing price controls on commodities. That sort of order would require separate statutory authorization.

Florida’s Emergency Management Act27 is typical. During a declared emergency, the governor may:

  • Suspend Regulation: “[S]uspend regulatory statutes proscribing procedures for state conduct or the orders or rules of any state agency, if strict compliance with the provisions of any such statute, order, or rule would in any way prevent, hinder, or delay necessary action in coping with the emergency.28
  • Direct State Resources: Direct state resources and personnel, militia, public utilities as necessary in responding to the emergency.29
  • Direct Evacuation: Direct evacuation and provide temporary housing.30
  • Control Movement: “[C]ontrol ingress and egress to and from an emergency area, the movement of persons within the area, and the occupancy of premises therein.”31
  • Control Civilians: “[T]ake measures concerning the conduct of civilians, movement of pedestrian and vehicular traffic, the calling of public meetings and gatherings, as provided in the emergency management plan of the state and political subdivisions thereof.”32
  • Commandeer Property: “[C]ommandeer or utilize any private property… [as] necessary to cope with the emergency.”33
  • Suspend Sale of Alcohol and Weapons: Suspend the sale of alcoholic beverages, firearms, explosives, etc.34
  • Exempt Certain Businesses from Curfews: Authorize certain businesses to operate in excess of otherwise applicable curfews.35

Note that not one of these enumerated powers expressly authorized the governor to issue general business closure orders. Nonetheless, Governor DeSantis was able to assert broad power to control business operations because he had been delegated power to issue orders controlling the movement of people. Specifically, the power to control “the occupancy of premises” within the disaster area gave him latitude not only to require business closures, but also to impose occupancy restrictions for certain industries upon re-opening.

Orders imposing restrictions were likewise justified in other states under similar authorizations to control or restrict the occupancy of premises. But in some cases, governors relied on still more tenuous interpretations of this authority. For example, Governor Polis issued an emergency order prohibiting evictions, in apparent reliance on his conferred authority to “[c]ontrol” the “occupancy of premises” within the disaster area. While technically an eviction moratorium might be said to control the occupancy of a premises within a disaster area, it seems unlikely that legislators intended this authorization to be construed so broadly. Indeed, these sort of authorizations were more likely intended to enable orders prohibiting citizens from entering dangerous areas, as opposed to authorizing a governor to regulate every aspect of activity occurring within a given premises.

North Carolina provides another example of how inartfully crafted statutes may fail to limit a governor’s emergency powers. As an initial matter the state’s Emergency Management Act limits the governor to only a few core powers, like utilizing state resources to respond to an emergency.36 The Act grants the governor some more extraordinary powers – like the power to issue price control orders – with consent from the Council of State, which is composed of other elected constitutional officers. At first blush, this structure would appear to significantly limit the governor’s emergency power authority.37 But another avenue of the Act allows the governor to exercise broad emergency powers – without the need for obtaining consent from the legislature or any other constitutional officer.

A separate section of the state’s EMA grants all sorts of emergency powers to local authorities in the event of a local emergency – including express authority to close down businesses.38 But the governor himself is permitted to exercise these “local emergency powers” upon a finding that “local control of the emergency is insufficient to assure adequate protection for lives and property.”39 But of course, that regime left it entirely to Governor Cooper’s discretion to decide whether and to what extent to regulate North Carolina businesses; the Act contains no limitations controlling his judgment as to whether this standard had been met. Not unlike California’s conferral of “all police powers,” North Carolina’s EMA gave Governor Cooper unfettered discretion.

C. Opportunities for Substantive Reforms

One obvious reform option would be to require the governor to obtain legislative authorization to continue emergency orders after a set time frame, or within so many days of the legislature resuming session. Another attractive option would be to enable the legislature to terminate any given emergency order through a mere resolution, requiring only a simple majority, without declaring the emergency over. At the very least, the legislature must retain some ability to terminate an emergency declaration if it disagrees with the way the governor has used his powers. But at least one state (North Carolina) confers irrevocable emergency powers, meaning the state’s General Assembly must enact new legislation and get the governor to sign it before gubernatorial emergency powers can be terminated.40

Our experience over the past year might also prompt state legislatures to reconsider the scope of powers conferred by emergency management statutes. In a place like Arizona or California, that might mean amending the provisions giving their respective governors “all police powers” of the state; they might be wise to move to providing a menu of options for the governor to choose from. But even where emergency powers are so enumerated, we have seen that there may be some need to make changes to clarify the limits of those enumerated powers. For example, a state might still permit the governor to control the occupancy of premises within a disaster area to prevent assembly within buildings that may be in imminent danger – while making clear that this does not entail a power to regulate private enterprise. And it may be that the legislature wants to consider certain express prohibitions that would prevent the governor from, say, imposing occupancy restrictions on businesses based on industry classification, or from imposing occupancy restrictions on religious gatherings that are not also imposed on commercial entities.

The legislature could also enact an amendment to condition the exercise of powers in various ways. For example, the legislature could authorize eviction moratoriums during a declared emergency, but on condition that the State must provide landlords just compensation for any non-paid rents during this period. More often than not, such conditional authority would mean requiring the governor to make certain predicate findings of fact before issuing orders on certain subjects. For example, the legislature could preserve an emergency power to restrict commercial activity while still channeling the exercise of discretion by requiring the governor to consider various factors in deciding what restrictions will apply on any given business.

In vesting the governor with only specific powers, these emergency management acts at least conceptually limit the scope of the governor’s emergency powers to issuing orders on specific subjects.

Arbitrary Implementation and Enforcement of Emergency Orders

Executive orders appeared increasingly unbounded during the pandemic. In many cases, no clear rationale existed for why some businesses were allowed to stay open and others were not. “Essential business” classifications regularly appeared to be random.

A. Case Studies in Arbitrary Orders

For example, when Governor Ron DeSantis issued a lockdown order in 2020, he later decided to exempt World Wrestling Entertainment (WWE), so long as the wrestlers performed without an audience. Theaters, however, were not granted that same kind of exemption. Meanwhile, in California, Governor Newsom allowed Disneyland to re-open in the spring of 2021, while some businesses remained entirely shuttered still by his Blueprint for a Safer Economy.41

In Connecticut, nail and hair salons were given different reopening dates by the government for no reason – despite the fact that “[t]he Governor’s own Reopening Plan recognizes that hair salons and nail salons pose the same public health risks and both [could] operate under safety guidelines.” This meant that some struggling salons were treated entirely differently. It also meant that Luis and Rosiris Ramirez, owners of Roxy Nail Design in Hartford, Connecticut, had to struggle to get by. As their attorneys, at Pacific Legal Foundation, explained, “Luis and Rosiris were relieved to learn they might be able to reopen on May 20. They scraped together $800 to comply with state-required safety precautions and prepared to immediately open their doors and safely serve customers.” But according to Connecticut’s emergency orders, that wasn’t enough; they had to remain shuttered still longer, without any legitimate explanation.42

Likewise, in North Carolina, Governor Cooper’s executive orders allowed all establishments that serve alcoholic drinks to resume operations in a phased re-opening of the economy, – but he continued to prohibit re-opening for just one class of bars.  Club 519 in Greenville, North Carolina was shuttered for nearly a year under these continuing orders. The bar’s owners, Crystal and Kenneth Waldron, were lucky enough to find pro bono representation from Pacific Legal Foundation, and their lawsuit ultimately prompted the state to allow a limited re-opening in February, 2021.

Yet even when permitted to re-open, businesses often face seemingly arbitrary rules. For example, Governor Cooper’s emergency orders required that businesses serving alcohol also had to serve food, even though a patron could drink all night without ordering anything to eat. These sort of restrictions were common though. In New York, restaurants began selling chips seasoned with anger (dubbed “Cuomo Chips”), due to a rule his administration imposed that required restaurants to serve food with alcoholic beverages for dubious public safety reasons. Shortly thereafter, State Liquor Authority regulators added to the arbitrariness of these rules in stating that chips did not count, but chips and salsa did.

As silly as it might seem to casual observers, the State Liquor Authority actually conducted sting operations to ensure compliance. One Saratoga Springs bar, Pint Sized, was fined over $1,000 for failing to comply. Because the regulations forced restaurants to serve food that customers did not want, the bar cut down on food waste by providing one bowl of canned food per table. Undercover Liquor Authority agents ordered drinks and were given too little food that nobody was eating, so they fined the bar.

B. Procedural Safeguards Minimize the Risk of Arbitrary Rules

The legislature might consider procedural rules to guard against arbitrary orders or favoritism. For starters, one basic procedural rule would be to require advance notice to the public for orders issued after so many days in an ongoing emergency, as well as opportunity for public comment. Notice and comment is infeasible in a quick-moving emergency, but if we are going to allow a continuing exercise of emergency powers for an indefinite time, we should expect transparency and opportunity for public engagement. For example, Governor Newsom pronounced his controversial color-coded system of industry regulation in late September – 177 days into the emergency. Clearly the governor had time to seek public input before implementing this complex regime. Likewise, there was plenty of time for notice and comment for emergency orders as they were renewed in 60- or 90- day increments in many states.

But there are other innovative models to draw from. Earlier this year, North Carolina’s legislature proposed building on the state’s existing requirement that the governor must obtain concurrence from its Council of State – i.e., a council of various elected constitutional officers, like the Secretary of State – before exercising certain emergency management powers. The proposal “[w]ould require the governor to obtain formal support from other elected leaders to enforce long-term statewide emergency orders,” as WECT reported. It would also require the concurrence of the Council for state-wide declarations of emergency longer than thirty days. Without concurrence, the declarations would expire within seven days. And even with concurrence, emergencies longer than thirty days could not continue without further concurrence. Following this model, other states might consider amending their EMAs to require concurrence of other constitutional officers as a sort of check when the governor contemplates such drastic measures as shutting down entire industries.

All of these fixes would be a step in the right direction. But ultimately the only way to preserve political accountability is to ensure that the legislature is making meaningful decisions. That won’t wholly cure the problem of arbitrary governance. But it’s better than continuing autocratic rule.

For starters, one basic procedural rule would be to require advance notice to the public for orders issued after so many days in an ongoing emergency, as well as opportunity for public comment.

Temporal Limitations as a Means of Ensuring Political Accountability

As set forth above, state emergency management laws tend to fall into one of two models. Some states authorize the governor to do virtually anything in the name of mitigating the effects of emergency. Other states implicitly limit the governor’s emergency powers to the issuance of orders on certain subjects; however, we have seen that, in practice, those enumerated emergency powers leave tremendous leeway for governors to pursue whatever policies they deem appropriate. That means the most effective limitation on the exercise of emergency powers are unambiguous time constraints.

States may impose temporal limitations on emergency powers on the assumption that the extraordinary exercise of emergency powers can only be justified on a short-term basis. The default rule in many states is that an emergency declaration automatically expires after a set number of days. But this sort of temporal limitation is of limited practical value if the governor can unilaterally renew the emergency declaration to further extend emergency powers incrementally. Indeed, through 2020 and into 2021 numerous Governors simply renewed emergency orders on a recurrent basis.

A much more effective means of limiting emergency powers is to require the governor to seek and obtain approval from the legislature for any extension. At the outset of the pandemic, at least eight states (as well as the District of Columbia and the Virgin Islands) required this sort of legislative ratification. More recently, several states have amended their Emergency Management Acts to impose this sort of requirement.

Of course, there is room to debate how long a governor should be enabled to exercise emergency powers before having to seek authorization from the legislature. But if the purpose of delegating emergency powers is to enable a swift and coordinated state response to an emergency, then the unilateral exercise of such powers can only be justified when the legislature is incapable of resolving fundamental policy questions. As such, Kentucky’s recent enactment requiring legislative approval for the continuation of emergency orders upon three days of reconvening the legislature makes sense. Perhaps one might argue that a longer period (say two weeks or 30 days) is appropriate to ensure time for extended deliberation. Yet in any event, it stands to reason that there should be some definitive temporal limitation.

It is noteworthy that in many cases the governor is vested with discretion to decide just how long his or her emergency powers will continue. For example, once an emergency is declared in California, the governor can continue exercising emergency powers indefinitely. While the governor is encouraged to end the emergency at the earliest possible date, the governor maintains unabated discretion to decide whether emergency conditions still persist. In these states, the only way to wrestle rulemaking powers back from the governor may be for the legislature to vote to terminate the emergency declaration on its end. But this is starkly different from those states where an emergency declaration ends automatically without a vote to extend it.

The difference is that a requirement to seek legislative authorization to extend emergency powers ensures that the elected legislative body ultimately retains control as to fundamental state policy, whereas the alternative – as in California and other such states – is that the governor retains unilateral control in fact. It is true that there is some theoretical possibility of legislative intervention if the statute allows termination of an emergency by mere legislative resolution; however, in practice there is little incentive for the legislature to intervene, because it would mean owning the political consequences of making decisions on difficult issues. Furthermore, intervention to end an emergency declaration means cutting off emergency funds and perhaps ending some emergency orders that might still seem appropriate – even while legislators might take issue with other orders.

To be sure, such considerations might likewise call for a legislature to vote to further extend a governor’s emergency powers. But at least in that case the legislature can be said to have expressly approved of the governor’s policies; it must then own the political consequences. As such, firm temporal limitations are ultimately for the best if we value political accountability.

…the most effective limitation on the exercise of emergency powers are unambiguous time constraints.

Individual Rights During an Emergency: How Should Courts Scrutinize Emergency Orders Restricting Liberty?

In the midst of the Covid pandemic, state and local governments and their agencies issued orders restricting nearly all forms of social activity – from closing businesses to restricting or forbidding attendance at religious services. When several of those actions were challenged, courts applied various standards of review depending on the activity being regulated. The U.S. Supreme Court’s 1905 decision in Jacobson v. Massachusetts set the baseline for a high level of judicial deference – akin to rational basis review – to state action during a public health emergency.43 The courts relied on this precedent to restrict various forms of business activity. Yet, as the Supreme Court was faced with cases that involved restrictions on First Amendment rights, such as religious liberty, the Court indicated that strict scrutiny applies – requiring the government to show that the government restriction is narrowly tailored to advance a compelling government interest.44 The Court was particularly skeptical of government restrictions that prohibited religious activity but permitted business activity that posed a similar or greater threat to the public’s health and safety.

This dichotomy tracks the general state of judicial review when courts examine cases involving First Amendment protections compared to economic liberties. When laws or regulations restrict people’s freedom of speech or religious freedom, courts examine a challenge to that government requirement under what is called “strict scrutiny.”45 In other words, the court will require the government to prove that the restriction is narrowly tailored to accomplish a compelling government interest.46 Under this standard of review, a regulation that undermines a constitutional right is susceptible to being struck down.

The opposite is true for laws or regulations that restrict economic freedom and the right to earn a living. Courts examine these restrictions under a much more lenient “rational basis” test, under which a court will presume the law is constitutional and require the regulated party to disprove every imaginable justification for the law.47

Under this relaxed standard of review, regulations will be upheld in all but the most exceptional circumstances.48

But the Court’s skepticism of regulations that discriminate between religious activities and secular activities should apply in both directions. In other words, our system should presume in favor of the rights of job seekers and business owners, requiring regulators at least to provide some good reason for interfering with a person’s ability to work. This is particularly true during a declared state of emergency, when popularly elected officials may be more inclined to disregard the structural restraints on their power. Thus, whether a government regulation interferes with the free exercise of First Amendment liberties, or economic liberties, the government should bear the burden of justifying restrictions on any activity by pointing to a sufficient government interest that is carefully tailored to achieve that interest. This is true whether an emergency exists or not.

A. The Jacobson Standard

The test articulated in the U.S. Supreme Court’s decision in Jacobson has been the standard by which courts have evaluated the constitutionality of emergency health measures for more than a century. In Jacobson, the Supreme Court rejected a Fourteenth Amendment challenge to a compulsory vaccination law enacted during a smallpox epidemic.49 The Court said that a community enjoys the right “to protect itself against an epidemic of disease which threatens the safety of its members.”50 It went on to observe that the “pressure of great danger[],” like a global pandemic, justifies the reasonable restriction of constitutional rights “as the safety of the general public may demand.”51 The Court then set out two separate tests, either of which a plaintiff must satisfy when challenging a government restriction during a health crisis: (1) that the government orders have “no real or substantial relation” to protecting public health, or (2) that they are “beyond all question, a plain, palpable invasion of rights secured by the fundamental law.”52 This looks somewhat like rational basis review during non-emergencies, although the language may seem more sweeping and deferential. Under the standard articulated in Jacobson, all but the most sweeping or irrational government restrictions that do not involve a fundamental right will be upheld during a health emergency. 

B. Standard of Review for Business Closures and Other Restrictions on Economic Liberty

Application of the Jacobson standard in business shutdown cases has generally led courts to uphold government health orders. For example, consider the recent case in which several bar owners challenged Louisiana Governor John Bel Edwards’s order banning on-site consumption of food or drink at bars while allowing restaurants with bars to remain open.53 The bar owners alleged that the order lacked a rational basis and was a violation of due process, equal protection, and freedom from unlawful takings.54 Relying on Jacobson, the court denied the request for a preliminary injunction, finding that that the bar owners had not met their burden to show that the closure orders did not “bear[] a ‘real or substantial relation’ to the goal of slowing the spread of COVID-19 and were ‘beyond all question’ [in] violation of the bar owners’ constitutional rights.”55 Thus, even a restriction that appeared arbitrary in its treatment of certain activities that commonly occur in a restaurant was upheld under the Jacobson standard.

Yet, despite the significant level of deference afforded to government action during a health crisis under Jacobson, some courts have still found that government restrictions, including economic restrictions, went too far during the Covid pandemic. For example, one federal district court judge in Pennsylvania struck down public health orders issued by Governor Tom Wolf and health secretary Rachel Levine.56 While recognizing that “the Jacobson Court unquestionably afforded a substantial level of deference to the discretion of state and local officials in matters of public health, it did not hold that deference is limitless.”57 Id. at 896. The court found no evidence that the specific numerical limit on public gatherings was necessary to achieve public health goals; it concluded that the order was overly broad and not narrowly tailored to meet a compelling governmental interest.58 The orders to close businesses were found to lack a rational relationship to a legitimate governmental end, and the court concluded that the design, implementation, and administration of the closure categories were arbitrary.59 The court also held that the business closure orders violated the Equal Protection Clause, since they did not keep consumers home from businesses that could be open. Id. at 927.60 In the Butler case, the court applied a rational basis review but still struck down the restrictions. Thus, as least in this instance, it was perhaps the irrationality of the restriction rather than the standard of review that determined the outcome.

C. Standard of Review for Restrictions on First Amendment Rights Subject to Strict Scrutiny

During the Covid pandemic, the U.S. Supreme Court was most suspicious of restrictions on First Amendment rights, including religious freedom, particularly when restrictions on these rights were more onerous than restrictions on similarly situated secular entities. In Roman Cath. Diocese of Brooklyn v. Cuomo,61 the Catholic Diocese of Brooklyn and two Orthodox Jewish synagogues filed suit to block enforcement of Governor Cuomo’s emergency orders that imposed occupancy restrictions on houses of worship during the pandemic. The religious communities claimed that attendance restrictions violated their right to the free exercise of religion guaranteed by the First Amendment, particularly as secular businesses in the same areas remained open. The Court found that the Diocese and the synagogues made a “strong” showing that the challenged restrictions violated a “minimum requirement of neutrality” by specifically naming religious entities for restrictions, while allowing secular businesses to be categorized as “essential.”62 The Court also found that “[t]he loss of First Amendment freedoms, for even minimal periods of time, unquestionably constitutes irreparable injury.”63 And it held that the government had not demonstrated that the requested relief would harm the public, as the government did not claim that attendance at the petitioners’ services resulted in the spread of disease.

Justice Neil Gorsuch authored a concurring opinion that referenced Jacobson but argued that the pandemic should not prevent the Court from applying “our usual constitutional standards.”64 Justice Gorsuch wrote that “Rational basis review is the test this Court normally applies to Fourteenth Amendment challenges, so long as they do not involve suspect classifications based on race or some other ground, or a claim of fundamental right. Put differently, Jacobson didn’t seek to depart from normal legal rules during a pandemic, and it supplies no precedent for doing so.” Instead, Jacobson merely applied “the traditional legal test associated with the right at issue.” That test – strict scrutiny – was properly applied. Thus, in Roman Catholic Diocese and future cases, according to Justice Gorsuch, that means the government must “treat religious exercises at least as well as comparable secular activities unless it can…show[] it has employed the most narrowly tailored means available to satisfy a compelling state interest.”65 Government restrictions that impair First Amendment rights – even during an emergency – will not be sustained if they cannot satisfy strict scrutiny, particularly when the restrictions single out the free exercise of religion or treat religious groups differently from similarly-situated secular organizations.

D. Toward a Model Standard of Review

The standards of review that apply when courts are reviewing emergency orders that restrict business, when compared to orders that restrict First Amendment rights, led to an obvious question about whether it is arbitrary to apply differing standards at all. As Justice Gorsuch astutely observed in Roman Catholic Diocese, “there is no world in which the Constitution tolerates color-coded executive edicts that reopen liquor stores and bike shops but shutter churches, synagogues, and mosques.”66 But shouldn’t that rationale work in both directions? In other words, when government health restrictions permit political protests or church services but restrict a business from safely operating, the judiciary should apply equal skepticism. As Judge Willett, then on the Texas Supreme Court, noted in a case that applied heightened judicial scrutiny for economic regulations, “Self-ownership, the right to put your mind and body to productive enterprise, is not a mere luxury to be enjoyed at the sufferance of governmental grace, but is indispensable to human dignity and prosperity.”67

The forces of majoritarian rule over individual liberty and constitutionally limited government are just as strong whether elected representatives seek to suppress free speech or restrict free enterprise. As Judge Willett put it, “a written constitution is mere meringue if courts rotely exalt majoritarianism over constitutionalism.”68 This is particularly true during declared emergencies, when the popular pressure to depart from constitutional norms in the name of safety or exigency is often strong. Just as in religious liberty or First Amendment cases, when the government restricts people’s rights to provide for themselves and their families, it should be required to show that there is a public safety requirement for that restriction and that the restriction is appropriately tailored to the public safety requirement—or at least that it does something to substantially advance the cited government interest. What’s more, the courts should not allow disparate treatment among activities that the government alleges pose similar threats to the public’s health and safety. The local small business that can safely operate with a ten-person capacity should not be shuttered if hundreds are permitted to gather at a rally.

E. A Role for State Legislatures

As observed above, state legislatures also have a crucial role to play in restoring constitutional boundaries when state and local officials exercise emergency powers. In addition to imposing objective temporal limitations on emergency powers or amending existing emergency management statutes so as to require the governor to call the legislature into session during a statewide emergency, the legislature can also minimize the risk of arbitrary emergency orders by enacting statutes establishing heightened standards of review. For example, the legislature can create a cause of action that directs state courts to enjoin any emergency order (or for that matter any other state action) that is not appropriately tailored to protect the public’s health and safety.

…our system should presume in favor of the rights of job seekers and business owners, requiring regulators at least to provide some good reason for interfering with a person’s ability to work.

The Need for Clarity, Lenity, and Prompt Guidance

At this point, it comes as no surprise that emergency orders add significantly greater complexity to the thicket of regulatory issues that small businesses already face. Even in the best of times, entrepreneurs face a daunting task when trying to navigate complex regulatory regimes. As such, regulatory compliance eats up a disproportionate share of time, above and beyond the ordinary tasks that come with operating a business. And of course, good faith mistakes happen, even for businesses striving for 100 percent compliance.

That is especially true when businesses must pivot, nearly overnight, when subjected to emergency orders. The burdens created by policymakers during an emergency are not confined to state-level rules. Local authorities complicate matters by imposing additional conditions, restrictions, and mandates on top of statewide emergency orders. As we’ve explored previously, America’s entrepreneurs faced a bewildering tangle of overlaid regulations before the pandemic. Those regulations are often repetitive or of marginal value. Sometimes they are even contradictory. To busy entrepreneurs, regulations appear to be coming from all sides; they impose real costs, both in dollars and innovation.

The extraordinary regulatory ambiguity that the nation’s businesses faced in 2020 is instructive. Emergency restrictions were narrow in scope at first and were aimed at specific types of businesses that appeared to create the highest risk for spreading the virus. But on March 19, 2020, Governor Gavin Newsom issued a stay-at-home order for California that required closure of non-essential businesses. The governors of New York and Illinois followed suit on March 20. By April most of the nation was on lockdown, with most local and state governments allowing only “essential businesses” or “essential employees” (also referred to as “life-sustaining” and “non-life-sustaining”) to continue operations.

The question for our country’s businesses and workforce soon became: what is an essential business? But no one knew, not even those imposing the restrictions. This was unprecedented. As one court noted, “there has never been an instance where a government or agent thereof has sua sponte divided every business . . . into two camps – ‘life sustaining’ and ‘non-life-sustaining’ – and closed all of the businesses deemed ‘non-life-sustaining.’”69

To make matters even more confusing, definitions varied from state to state; within states, many local government definitions contradicted one another. For example, in Georgia, nearly every single one of its 538 cities and 159 counties defined “essential” business differently and imposed different restrictions. The effect of this patchwork was that the very same business could simultaneously be shut down completely, be allowed to operate freely, and be allowed to operate if it closed by 9 p.m. because the applicable city, county, and state emergency orders differed.70 For the average small business owner just trying to survive financially during an unprecedented emergency, it could be overwhelming, confusing, and terrifying.

As the country started opening up, states and local governments issued even more emergency orders – including standards for social distancing, health screening practices, face coverings, and return-to-work protocols. We may never know if medical and scientific evidence justified all of these standards, but what we do know is that many of these additional emergency orders made matters even more confusing – they continued to differ by industry, size, and location.

When state and local governments issue emergency orders, they produce multiple layers of uncertain and ambiguous state and local orders on top of an already confusing regulatory labyrinth. During the COVID-19 emergency, the only way to resolve that confusion was through lawsuits.71 And there were many lawsuits. But lawsuits like those we saw in response to the thousands of COVID-19 emergency orders across our country are expensive. And, by the time small businesses got resolution and clarity, it was frequently too late to save them.

This is precisely why state and local governments must issue timely guidance to explain their emergency orders. It has been 19 months since America shut down in an emergency response to the pandemic. Nonetheless, a number of states still do not define “essential” workers, but still seek to enforce many provisions set forth in emergency orders that will soon be two years old.

In fairness, no one could have anticipated COVID-19, but that is true for any emergency. That is why our local and state governments should get to work now in creating a process by which businesses could have an opportunity to seek clarification of emergency orders. Then, after the government provides clarification, businesses could have an opportunity to correct mistakes before facing enforcement actions and/or lawsuits. To be clear, this isn’t a proposal to turn a blind eye: rather, it encourages businesses to obtain, and the government to provide, necessary guidance surrounding ambiguous, confusing, and hastily drafted emergency orders.

Additionally, as we have suggested before, in times of emergency, it is important to offer regulatory safe harbor provisions for businesses that make technical mistakes. In other words, unless businesses are willfully violating a regulation, regulatory agencies enforcing emergency orders should take a remedial approach. Conduct should be deemed willful only if (i) the person charged with violation of a rule knew that his or her conduct was prohibited by the rule, or (ii) a reasonable, similarly situated person would have known that his or her conduct was prohibited by the rule. Just because a state or local government has published a rule should not suffice to demonstrate that a person knew, or that a reasonable person would have known, of the existence or content of the rule. The demonstration of a willful violation should be an independently required element of proof that must be established in addition to any other requirements imposed by law or rule. This should serve as a basic framework for liability.

When state and local governments issue emergency orders, they produce multiple layers of uncertain and ambiguous state and local orders on top of an already confusing regulatory labyrinth.

The Importance of Benefit-Cost Analysis in Times of Crisis

No discussion of emergency management orders would be complete without considering how the authority they imply should be exercised. At one end of the spectrum, governors or other executive-branch officials might do nothing at all, choosing instead to stand back and let the crisis run its course in the face of purely unofficial responses from private individuals and firms. At the other end of the spectrum, officials might exercise every bit of power at their disposal on the premise that doing something always beats doing nothing, even in the face of challenges that are not well understood.

In between those extremes lies a happy medium that we suggest will almost always be the best course. That alternative rests on an analysis of (1) the overall magnitude of the potential harms, (2) the net benefits (if any) of various governmental responses, and, perhaps most importantly, (3) whether there are any relatively low-impact responses that seem likely to deliver significant net benefits while imposing comparatively light burdens and restrictions on individuals, markets, and other sectors of the economy. That process is typically described as a “benefit-cost analysis” or BCA.72

A. The Nature of Benefit-Cost Analysis

A BCA “is a simple and sometimes controversial technique for thoroughly and consistently evaluating the pros and cons associated with prospective policy changes. Specifically, it is an attempt to express in dollar terms all of the effects of proposed government policies projects.”73 At the heart of the BCA is the conviction that “the effects of a policy change on society are no more or no less than the aggregate of the effects on the individuals who constitute society.” As Cass Sunstein has explained, “[i]f we care about human welfare, we will insist on exploring the costs and benefits of various approaches, because the resulting numbers give us indispensable information.”74

Public opinion sometimes appears uninformed by the insights of BCA. Thus, for example, following record-setting low temperatures and rolling blackouts during February 2021, many people were quick to heap scorn on state officials responsible for failing to winterize the Texas power grid. However, as a study by Texas A&M’s Private Enterprise Research Center noted, Texans enjoy some of the cheapest electricity in the nation, and mandated winterization represents a double-edged sword in the form of higher electricity costs and discouraging power generators from building new power plants, leading to reduced capacity, lower “reserve margins,” and a correspondingly greater likelihood of blackouts.75 In this instance, BCA illuminated the opportunity costs of policy alternatives.

A collateral benefit of BCA is that the process itself can help alert us to certain irrational biases that incline us towards suboptimal policies. Those biases include the “identifiable victim effect” (the tendency to discount “statistical” deaths of faceless strangers compared to people we know personally); identifiable cause effect (prioritizing efforts to save lives at risk from known causes); present bias (preferring to save fewer lives in the short term than more lives over a longer period of time); anchoring bias (adhering to an initial hypothesis despite evidence that disproves it); and escalation of commitment bias, which entails “investing more resources into a set course of action even in the face of evidence [that] there are better options.76

B. Practical Problems of Benefit-Cost Analysis

The COVID pandemic prompted a blizzard of BCAs and a substantial secondary literature about the challenges of performing BCA in that context.77 The picture that emerges from these studies is scarcely reassuring and perhaps even a bit discouraging. Some researchers assess with substantial confidence that the benefits of social distancing, lockdowns, and other public-health responses undertaken before vaccines became available provided somewhere between $5.2 trillion78 and $8 trillion79 in net benefits, whereas others estimate the net benefits at somewhere between $0 and $800 billion.80 Still another highly experienced BCA researcher expressed serious doubt whether it was even possible to perform a meaningful BCA for COVID given that “[a]ny such attempt must pin down several great unknowns,” including the extent to which people would or would not comply with government mandates and “how different types of people react to having different vulnerability to the disease.”81 As one journalist aptly summarized:

“The cost-benefit analysis approach to Covid-19 shutdowns clearly needs some honing. A hodgepodge of closure and reopening policies among populations with wildly different risks of infection and death does not lend itself to balancing a cost in dollars against a cost in blood. What researchers would like to know is which specific interventions are most successful stopping the virus and have the least impact on people’s economic lives.”82

The challenge of performing a useful BCA is further compounded when the values at issue are categorically different and appear less commensurable. For example, the Federal Emergency Management Agency (FEMA) provides a “Benefit-Cost Analysis Toolkit” that enables policymakers and other decision makers to complete the BCA that the agency often requires in order to demonstrate the cost-effectiveness of various disaster risk reduction strategies that may be eligible for agency subsidies.83 To streamline the grant application process, FEMA provides pre-calculated analysis for various eligible projects, including residential and non-residential hurricane wind retrofits, tornado safe rooms, and responses to threats from floods and wildfires.84

A BCA that examines the cost of retrofitting homes in order to make them more hurricane-resistant and then compares those expenses to the projected cost of repairing or replacing those same homes from the number and strength of hurricanes to which they are likely to be subjected during their useful lifetimes represents a fairly straightforward dollar-versus-dollar comparison. But of course, hurricanes impose plenty of costs that are not purely monetary, including everything from human deaths and injuries to the loss of pets, photo albums, and cherished family heirlooms, all of which have significant but potentially highly subjective value that is extremely difficult to express in dollars and cents.

In the midst of a dynamic and rapidly unfolding crisis, analysts faced a host of uncertainties that substantially complicated the necessary calculus. In the early stages of the COVID pandemic, for example, these unknowns included the transmissibility of the virus itself, the lethality of COVID-19, and the nature and extent of “iatrogenic collateral harms” that might flow from lockdowns, quarantines, school closures, and other potential policy responses. It turns out those included “major detrimental effects on childhood vaccination programs, education, sexual and reproductive health services, food security, poverty, maternal and under five mortality, and infectious disease mortality.”85 Even when they can be identified with some confidence, those kinds of harms are extraordinarily difficult to quantify and assign dollar-value cost figures to. One researcher memorably characterized this as “The Corona Dilemma,” akin to the famous “Trolley Problem” from moral philosophy, and argued that faced with a choice between pulling the metaphorical lever to divert the COVID-19 train or doing nothing, “[t]he world pulled the lever, and the unintended health consequences of these [lockdown] measures did not play a part in modeling or policy.”86

Another confounding challenge is figuring out how to account for the effect of people’s behavioral changes in response to new information. Again, with COVID this included widespread use of hand sanitizer (which appears to have accomplished very little) and voluntary masking (which we can now say with confidence substantially reduced the spread of coronavirus in enclosed spaces). Simply put, people have strong incentives to act on new information about potentially life-saving new practices, and a statistical model that assumes continuity from an initial set point will do a poor job of predicting actual outcomes.

C. The Relevance of CBA in a World of Limited Information

The point of this paper – and specifically this section – is not to examine the limits of BCA analysis in general, nor to assess the overall quality of the COVID-19 response so far, including the various benefit-cost analyses that informed at least some of those responses. Rather, the point here is to make the case that the exercise of emergency management powers should always be undertaken on the basis of at least some effort to perform a rational assessment of the likely cost and benefits of any given course of action as compared to its alternatives – including no action.

Performing a reliable BCA can, as noted above, present considerable challenges. According to one COVID-related BCA, the end goal for policymakers in deciding how to respond to a crisis “is to maximize the sum of years lived by the population, weighted by the health quality of those years…or the wellbeing quality of those years – the latter being defined as “the value of anything that makes life enjoyable…measured by life satisfaction.”87 While assigning a dollar value to something as amorphous and value-laden as “wellbeing quality of years lived,” may seem inherently subjective, assessing the value of human lives saved or the “value of a statistical life” is a routine fixture of economic analysis and public policymaking, with respect to everything from how much money it is worth spending to fix a hazardous curve on a highway to decisions about how to allocate scarce medical resources. So, the mere fact that we cannot achieve absolute precision does not mean that BCA has nothing to offer policymakers.

Indeed, BCAs can provide useful analysis to policymakers even when they make persistent errors in their assumptions about various costs and benefits. Thus, two scholars conducted an ex-post evaluation of BCAs used to guide decision-making on road projects in Norway and discovered that even though the BCAs reflected certain persistent inaccuracies – including systematically underestimating traffic growth rates and construction costs – they still shed useful light on whether to undertake various road projects.88

Notably, some commentators have argued that, in the context of COVID, the extent of the unknowns and unquantifiables necessarily renders BCA analysis imprecise and even dubious. Two responses seem especially salient here. First, it is axiomatic that policymakers must never allow the perfect to become the enemy of the good. Imprecision is a fact of life when dealing with large numbers of more or less autonomous, imperfectly rational people. But the goal of BCA generally is not to identify the perfect or most optimal policy; instead, it often has the more modest, but still extremely useful, function of helping policymakers identify and eliminate the worst options – whether that means complete inaction or blindly careening from one panacea to another as the public becomes increasingly jaded and noncompliant.

Another reason to employ BCA even when precision is impossible to achieve is that doing so makes it more difficult for policymakers to cloak their actions in the mantle of the public interest when their true motives are more narrowly self-interested – currying favor with particular constituencies, for example, or incurring the political cost of making vital but politically unpopular demands on the electorate. Of course, the mere fact that a BCA identifies certain courses of action as being clearly more beneficial than others certainly does not assure that policymakers will embrace those policies, but it may well make it more difficult and politically costly to pursue socially suboptimal policies in pursuit of narrow self-interest.

In the end, what BCA enables us to do, even under conditions of great uncertainty, is estimate the costs of the extremes – that is, doing nothing at all on the one hand versus full-blown lockdowns in the context of COVID – and help us focus on the fact that “the real policy work is on the intermediate solutions,” including “targeted policies” that are likely to be highly productive.89

what BCA enables us to do, even under conditions of great uncertainty, is estimate the costs of the extremes – that is, doing nothing at all on the one hand versus full-blown lockdowns in the context of COVID – and help us focus on the fact that “the real policy work is on the intermediate solutions…”


Emergencies require rapid responses. The executive must be allowed authority not only to mobilize state resources to respond to an emergency, but also some latitude in issuing the emergency orders that are needed to protect the community – including matters involving public health and private property. But we have seen that existing emergency management statutes are flawed primarily because they give emergency powers for the governor that are unfettered across multiple dimensions.

No one doubts the need for an efficient and effective state response to emergencies; however, we must also ensure that we are protecting our basic liberty to be free of autocratic rule. One lesson of the pandemic is that unfettered emergency powers present an existential threat to our constitutional order – at least where a governor is permitted to wield those powers on an ongoing basis. Policymakers should consider the solutions we have described above in the course of planning for the next inevitable emergency.


  1.  U.S. v. Caltex, 344 U.S. 149, 154 (1952).
  2.  Id. at 153-54 (quoting United States v. Pacific R. Co., 120 U.S. 227, 234 (1887)). But see Ark. Game & Fish Comm., 568 U.S. 23, 36 (2012) (holding that the government may incur liability under the Takings Clause to provide just compensation when intentionally flooding private property).
  3.  Campbell Robertson, Areas Will Be Flooded to Protect Louisiana Cities, N.Y. Times (May 13, 2011),
  4.  Caltex, 344 U.S. at 154.
  5.  See Hickox v. Christie, 205 F. Supp. 3d 579, 585 (D.N.J. 2016); see also Compagnie Francaise de Navigation a Vapeur v. Bd. of Health of State of La., 186 U.S. 380 (1902) (upholding quarantine).
  6.  Benson v. Walker, 274 F. 622 (4th Cir. 1921).
  7.  Jacobson v. Commonwealth of Massachusetts, 197 U.S. 11, 25 (1905); Phillips v. City of N.Y., 775 F.3d 538 (2d Cir. 2015); Reynolds v. McNichols, 488 F.2d 1378, 1381-83 (10th Cir. 1973).
  8.  See
  9.  See Jim Rossi, State Executive Lawmaking in Crisis, 56 Duke L.J. 237, 246 (2006).
  10.  E.g. Cal. Gov’t Code § 8558; Tex. Gov’t Code § 418.014(a); Ariz. Rev. Stat. §§ 26-301, 26-303(D).
  11.  E.g. Conn. Gen. Stat. Ann. § 28-9; Kan. Stat. Ann. § 48-924; Okla. Stat. tit. 63, § 6405; S.C. Code Ann. § 1-3-420.
  12.  E.g. Cal. Gov’t Code § 8629; Tex. Gov’t Code § 418.014(b)-(c); Ariz. Rev. Stat. § 26-303(F).
  13.  E.g. Haw. Rev. Stat. Ann. § 127A-14. Other states that lack such legislative authority include Ohio, Vermont, and Washington.
  14.  E.g. Cal. Gov’t Code §§ 8627, 8627.5, 8628; Tex. Gov’t Code §§ 418.012, 418.016(a), 418.017, 418.018; Ariz. Rev. Stat. § 26-303(E); N.H. Rev. Stat. § 4:47; Okla. Stat. tit. 63, § 683.9.
  15.  E.g., Ala. Code § 31-9-8(a)(5).
  16.  Tex. Gov’t Code § 418.004(1).
  17.  N.Y. Executive Law § 20(2)a.
  18.  Japan Whaling Association v. American Cetacean Society, 478 U.S. 221, 230 (1986).
  19.  Ctr. for Biological Diversity v. Trump, 453 F. Supp. 3d 11, 32 (D.D.C. 2020).
  20.  Echoing this sentiment, the American Public Health Association also declared racism to be a public health crisis, and called a declaration to this effect “an important first step in the movement to advance racial equity and justice [to be] followed by allocation of resources and strategic action.”
  21.  Ibid.
  22.  For further discussion on these historical analogs, see Amicus Br. of Goldwater Inst., Ghost Golf v. Newsom, No. F082357, available at
  23.  In re Certified Questions From United States Dist. Court, W. Dist. of Michigan, S. Div. No. 161492, 2020 WL 5877599 (Mich. Oct. 2, 2020) (holding an emergency management statute in Michigan violated separation of powers in conferring a substantial portion of the state’s police power to the Governor); but see Beshear v. Acree, 615 S.W.3d 780 (Ky. Nov. 12, 2020) (upholding a broad delegation of emergency powers); Casey v. Lamont, No. 20494, 2021 WL 1181937 (Conn. Mar. 29, 2021) (same).
  24.  Fl. Const., Art, Section 2 (providing that the seat of government will be in Tallahassee, except that “in time of invasion or grave emergency, the governor by proclamation may for the period of the emergency transfer the seat of government to another place.”).
  25.  See A.R.S § 26-303 (E)(1); Cal. Gov’t Code § 8626.
  26.  The only exceptions would be with regard to regulations impinging First Amendment rights, or other “fundamental rights.” But unfortunately, modern precedent does not recognize economic liberty as a fundamental right. For this reason, regulation of business activity is almost invariably going to survive scrutiny.
  27.  Florida Emergency Management Act (EMA), Fla. Stat. Ann. § 252.
  28.  Id. at § 252.36(5)(a).
  29.  Id. at §§ 252.36(5)(b)-(c), (j), (l), (n).
  30.  Id. at § 252.36(5)(e)(f)(i). The EMA requires the governor to “ formulate and execute plans and rules for the control of traffic in order to provide for the rapid and safe movement or evacuation over public highways and streets of people, troops, or vehicles and materials for national defense or for use in any defense industry…” Id. at § 252.36(10).
  31.  Id. at § 252.36(5)(g).
  32.  Id. at § 252.36(5)(k) (emphasis added).
  33.  Id. at § 252.36(5)(d).
  34.  Id. at § 252.36(5)(h).
  35.  Id. at § 252.36(5)(m).
  36.  E.g., NC Gen. Stat. § 166A-19.30(a)(1).
  37.  See Id. at § 166A–19.30(b)(2).
  38.  Id. at § 166A-19.31.
  39.  Id. at § 166A-19.30(c).
  40.  Note that Governor Cooper vetoed several bills that would have limited his emergency orders or powers in 2020 while continuing to assert a unilateral authority to decide how long to keep industries shuttered.
  41.  In a similar vein, the Kansas Division of Emergency Management allowed golf courses to remain open so long as they followed certain safeguards, but denied the same opportunity to photographers and personal trainers. All three businesses can incorporate social distancing and operate safely and effectively outside. So the distinction is not based on risk. Personal trainers faced such disparate treatment in other states as well. So it’s no wonder that one survey reported that, during 2020, “58 percent of personal trainers lost some or all of their income.” These exemptions and acts of favoritism have real economic consequences.
  42.  Hair salons were also subject to arbitrary and inconsistent rules in many states during the pandemic. For example, in California salons were temporarily required to operate outdoors, even as there were concerns their cosmetology licenses would not permit them to perform services outside.
  43.  197 U.S. 11 (1905).
  44.  Roman Cath. Diocese of Brooklyn v. Cuomo, 141 S. Ct. 63 (2020).
  45.  Reed v. Town of Gilbert, Ariz., 576 U.S. 155, 164 (2015) (content-based speech restrictions are subject to strict scrutiny); Espinoza v. Montana Dep’t of Revenue, 140 S. Ct. 2246, 2257 (2020) (laws that discriminate based on religious status are subject to strict scrutiny).
  46.  Reed, 576 U.S. at 171.
  47.  City of New Orleans v. Dukes, 427 U.S. 297, 303 (1976) (economic regulations will be upheld if they are “rationally related to a legitimate government interest.”).
  48.  City of Cleburne, Tex. v. Cleburne Living Ctr., 473 U.S. 432, 440 (1985) (when economic regulation is at issue, the state gets “wide latitude.”).
  49.  197 U.S. at 26.
  50.  Id.
  51.  Id.
  52.  Jacobson, 197 U.S. at 31.
  53.  4 Aces Enterprises, LLC v. Edwards, 2020 WL 4747660 (E.D. La. Aug. 17, 2020), at 315, aff’d sub nom. Big Time Invs., L.L.C. v. Edwards, 2021 WL 118628 (5th Cir. Jan. 13, 2021).
  54.  Id. at 327.
  55.  Id. See also Grisham v. Romero, 483 P.3d 545 (N.M. 2021) (finding that the governor’s executive orders for temporary closure of indoor dining imposed upon restaurants and breweries were neither arbitrary nor capricious, and that the unique risks of indoor dining and the increased COVID-19 cases among New Mexico restaurant staff showed a “real and substantial relation” between the order’s temporary prohibition and the object of controlling and suppressing the spread of COVID-19”); McCarthy v. Cuomo, 2020 WL 3286530 (E.D.N.Y. June 18, 2020) (denying plaintiff’s petition for preliminary injunction of Governor Cuomo’s coronavirus-related executive orders–an action to immediately re-open businesses–because plaintiffs failed to demonstrate a likelihood of success on the merits of any of their claims under the Jacobson standard).
  56.  Cty. of Butler v. Wolf, 486 F. Supp. 3d 883 (W.D. Pa. 2020).
  57.  Id.
  58.  Id. at 907.
  59.  Id. at 922.
  60.  The U.S. Court of Appeals for the Third Circuit granted a stay of the trial court’s injunction pending appeal, Cty. of Butler v. Governor of Pennsylvania, No. 20-2936, 2020 WL 5868393, at *1 (3d Cir. Oct. 1, 2020), and later dismissed the appeal as moot, after Pennsylvania General Assembly restricted the Governor’s powers to enter the same orders. Cty. of Butler v. Governor of Pennsylvania, 8 F.4th 226 (3d Cir. 2021).
  61.  141 S. Ct. 63 (2020).
  62.  Id. at 66.
  63.  Id.
  64.  Id. at 71.
  65.  Id. at 70.
  66.  Id. at 72.
  67.  Patel v. Texas Dep’t of Licensing & Regul., 469 S.W.3d 69, 92 (Tex. 2015).
  68.  Id. at 95.
  69.  County of Butler v. Wolf, Civil Action No. 2:20-cv-677, *49 (W.D.Pa. Sept. 14, 2020).
  70.  The rules were also inconsistent from state to state. Pennsylvania deemed hotels essential, while Nevada ordered all hotels to shut down.
  71.  This is assuming that the jurisdictional court was even operating, because many shut down themselves for a period of time.
  72.  “Benefit-cost analysis (BCA), and cost-benefit-analysis (CBA)…are generally regarded as equivalent terms” and will be treated as such in this section. Richard O. Zerbe & Tyler Scott, A Primer for Understanding Benefit-Cost Analysis, University of Washington.
  73.  Paul R. Portney, Benefit-Cost Analysis, Library of Economics and Liberty.
  74.  Cass R. Sunstein, This Time the Numbers Show We Can’t Be Too Careful, Bloomberg (March 26, 2020).
  75.  Dennis Jansen, et al., Trade-Offs in Winterizing the Texas Power Grid, Private Enterprise Research Center (Texas A&M University) (March 17, 2021).
  76.  James Odeck & Anne Kjerkreit, The Accuracy of Benefit-Analyses (BCAs) in Transportation: An Ex-Post Evaluation of Road Projects, 2019.
  77.  See, e.g., Society for Benefit Cost Analysis, Resources for Benefit Cost Analyses to Inform COVID-19 Policymaking, March 21, 2021.
  78.  Linda Thunstrom, et al., The Benefits and Costs of Using Social Distancing to Flatten the Curve for COVID-19, Journal of Benefit Cost Analysis, May 21, 2020.
  79.  Michael Greenstone & Vishan Nigam, Does Social Distancing Matter?, University of Chicago and NBER, March 2020.
  80.  James Broughel & Michael Koutros, The Benefits of Coronavirus Suppression: A Cost-Benefit Analysis of the Response to the First Wave of COVID-19, Mercatus Working Paper, June 2020.
  81.  Giles Wilkes, The Doubtful Case for an Impossible COVID-19 Cost-Benefit Analysis, Institute for Government, Nov. 27, 2020.
  82.  Adam Rogers, How Much Is a Human Life Actually Worth?, Wired, May 11, 2020.
  83.  See, e.g., FEMA, Benefit Cost Analysis.
  84.  Id.
  85.  Ari R. Joffe, COVID-19: Rethinking the Lockdown Groupthink, Frontiers in Public Health.
  86.  Id.
  87.  Id.
  88.  James Odeck & Anne Kjerkreit, The Accuracy of Benefit-Analyses (BCAs) in Transportation: An Ex-Post Evaluation of Road Projects, January 2019.
  89.  Casey Mulligan, et al., Some Basic Economics of COVID-19 Policy, Apr. 27, 2020.
Skip to content