Tensions in State–Local Intergovernmental Response to Emergencies: The Case of COVID-19

Intergovernmental Responses to Emergencies

The intergovernmental response to emergencies and disasters in the United States was designed to operate from the bottom up (Rubin and Barbee 1985; Schneider 1995, 2008; Zhang, Liu, and Vedlitz 2020). Policies for emergency response and preparedness are based on the assumption that those closest to the emergency have the best perspective on what help is needed. Accordingly, the response to an emergency begins with local governments, typically led by the county, and follows a prescribed series of steps as the emergency escalates to include both the surrounding local governments and the state, and eventually the federal government as needed (Kapucu 2008; Waugh 1994). This bottom-up structure places local governments at the center of emergencies and natural disasters. By starting emergency responsibilities at the local level, officials can address the needs and specific conditions of the community of which others may be unaware. In anticipation of an event, local governments develop emergency preparedness and response plans that adhere to certain state and federal guidelines that outline how the government will respond to a crisis and how it will coordinate its response with all levels of government (Kapucu, Lawther, and Pattison 2007).

As higher levels of government become involved, their intended role is to assist in the coordination of services and to support the local response rather than supersede or replace it (Kapucu and Hu 2016; Schneider 2008). Much like the local governments before them, states maintain a response plan that lays out how state-based resources will be deployed and how they will support and coordinate with their cities and counties in times of need. Though there is some variation on the structure of the plans from state to state, these plans identify the responsibilities assigned to state officials during an emergency to minimize confusion and maximize the effectiveness of the state’s response.

The bottom-up approach to emergency response does not mean states do not play an important role. As the chief executives of their states, governors play key roles in emergency responses (Waugh 2007). Governors are typically given a wide breadth of powers during emergencies, allowing them to declare states of emergency, order evacuations, and mobilize the National Guard. It is only after both local and state resources are exhausted that governors issue requests for federal assistance during a response. This approach to emergency response allows for what Schneider (2008) refers to as a “‘pull’ system of intergovernmental relief” (p. 718).

This emergency response structure has faced some challenges due to concerns surrounding homeland security. Since September 11, 2001, Birkland (2009) notes that the federal government has preferred the top-down approach allowing the federal government to use its experts in times of need rather than relying on local expertise during events that have national implications. Such events are rare, however, allowing the default bottom-up structure to continue.

In the context of a pandemic, the structure of an emergency response remains uncertain due to the novelty of the situation. National disasters are typically geographically bound, but the ease with which the coronavirus spreads makes the issue a national concern rather than a local one. Research into bioterrorism preparation indicates the federal government might push for a top-down response due to the national interest inherent in a terrorist-released bio-agent and the capacity of the federal government’s expertise to track and treat those sickened due to the release (Sutton 2001). Limitations of authority, however, hinder the ability of the federal government to lead the response, as Gibbons v. Ogden, 22 U.S. 1 (1824) reaffirmed the authority of state control, which includes the capacity to regulate public health and impose quarantines. Without planning and clarification by local, state, and federal governments on how the response might differ in an emergency, the default response to the emergency is the bottom-up structure previously outlined. Within the emergency response community, support for this approach is based on the Center for Disease Control and Prevention’s strategic plan for bioterrorism preparedness and response, which focuses on assessing local capacity and then augmenting that capacity as needed to achieve positive outcomes (Koplan 2001). This approach is further supported by the literature emerging out of the crisis in terms of what local and state governments expected to happen (see Berthod et al. 2020; Dzigbede, Gehl, and Willoughby 2020; Kettl 2020; Maher, Hoang, and Hindery 2020). The President proclaimed a National Emergency on March 13, 2020, but otherwise there was no comprehensive federal response to the pandemic (Kettl 2020). In response to this federal policy vacuum, sub-national governments developed their own policy approaches.

City and State Responses

Despite the planning and preparation for emergencies that all levels of government conduct, actual governmental policy responses to the COVID-19 pandemic have been more complicated. To examine the actions municipalities are taking, we turn to the National League of Cities’ (2020) COVID-19 Local Action Tracker. This website tracks actions taken by municipalities in response to the pandemic. These data are self-reported by city officials and citizens. Overall, this database represents 1,731 policy actions taken by 494 cities in 48 states plus the District of Columbia beginning in late February and ending on May 26, 2020. While not a representative sample, these data represent 179 of the 250 largest cities in the U.S. and account for 91.3 million residents. ¹ The median city size is 74,238 residents (average size is 184,806).

Figure 1 plots the number of city-level policy adoptions as a function of date. Policy adoptions begin in late February (February 28 being the earliest recorded adoption) and reach a peak in the week beginning March 16. These early policy changes are nearly all from cities on the U.S. west coast; however, there are outliers in states such as Michigan and Texas (see Figure 2 for a state distribution). The majority of city-level implementations were in the weeks beginning March 16 and March 23, during which time nearly 2/3rds of all policies were implemented. It is important to note local policy adoption largely occurred while daily case counts (the solid line in Figure 1) were low, and the adoption of additional policies is inversely related to the growth in COVID-19 cases.

OPEN IN VIEWER

Figure 1.

City policy adoption by date. Note: Bars = Number of local policies enacted; line = daily COVID-19 cases per million people.

States have their own responses, but these tend to be in the coordination realm (Kapucu and Hu 2016; Schneider 2008). These include but are not limited to official emergency declarations (e.g., activating state response plans), major disaster declarations, activation of the state’s National Guard, and in some instances placing limitations on large group gatherings and closing schools. ² Because states and local governments are somewhat autonomous, each can act without the other, leading to discrepancies in the timing of local-level and state-level policy adoption. We explore this in Figure 2, where local-level policy implementation by state is graphed in a manner similar to Figure 1, and vertical lines are introduced for when a state closed schools (dashed line) and when a state implemented a statewide “stay at home”/“shelter in place” order (solid line). ³ We chose these two actions because the former is a relatively uniformly implemented limitation on large group gatherings (as opposed to actual limits on large groups that vary considerably by state in their stringency) and the latter is on the more extreme end of policy action. Both are also ordered via gubernatorial executive order. The distance between these two actions can be seen as the time it took for a state to begin limitations on activity to fully implementing such a limitation. ⁴ We do not intend these two policy actions to summarize the entire state response; however, these are two signaling events for local governments about whether they should continue their own locally tailored responses or step back and accept uniform state-level intervention. ⁵ Whether this actually happened is a question we will return to in the conclusion.

OPEN IN VIEWER

Figure 2.

City and state policy adoption by state.

Consistent with our expectations, the timing varies between local and state action. In many states, local action predated the first large-scale state action (statewide school closures). However, in a number of states, local policy action increased after this first state-level restriction was imposed. We take this as evidence of states signaling to local governments that additional policy action was necessary. As the first wave of the pandemic progressed, many states imposed “stay at home”/“shelter in place” orders and in many cases, local response tailed off after these orders were announced. This is particularly evident in large states such as California, Florida, Illinois, and New York. However, each of these states has its own trajectory with California and New York having very little time between the two orders (zero and one days, respectively), Illinois having a somewhat larger gap (five days), and Florida having a significant gap (seventeen days). Local responses filled in the gap between the two state responses. In several states (those with either dashed or solid boxes in Figure 2), these “stay at home”/“shelter in place” orders were accompanied by broad preemptions of local governments’ ability to respond.

Executive Orders and State Preemption

As shown in the figures above and in line with the expected intergovernmental response to emergencies and disasters, many U.S. cities passed laws to protect public safety and health in response to the COVID-19 pandemic. However, many governors also responded to the pandemic with executive orders (more than 1,000 executive orders and/or agency orders have been issued about COVID-19, see Curley and Federman (2020) for more information), in some cases preempting their cities from taking action. In the 43 gubernatorial executive orders ordering states to stay-at-home or shelter in place, we observe 20 orders preempting local ordinances that order local sheltering in place or locally defined essential activity/business. ⁶

In Figure 2, states with floor preemptions have dashed boxes and states with ceiling or floor and ceiling provisions have solid boxes. Forty-five percent of these preemptions set policy minimums, called floor preemptions, in which the state establishes a baseline and local governments are free to set their own policies at higher levels (Goodman, Hatch, and McDonald 2020; Wagner et al. 2019). Alabama enacted a floor preemption when the State Health Officer stated specific local governments may be authorized to adopt more stringent response measures. Another quarter of preemptions set policy maximums, called ceiling preemptions, which prevent local governments from exceeding a state-imposed policy or regulation. Georgia’s Executive Order 03.14.20.01 is a ceiling preemption in that it says no local government can adopt a similar ordinance. The remaining 30 percent contained both floor and ceiling provisions. As a result, there is significant variation in the number and scope of policies aimed at addressing the crisis across states. This poses considerable challenges for local government managers in both their immediate and long-term emergency responses.

States have preempted their cities since at least the latter half of the nineteenth century (Zimmerman 2012). Goodman, Hatch, and McDonald (2020) identify four preemption epochs: tax and expenditure limitations, unfunded mandates, public health, and the new preemption. State preemptions in response to COVID-19 have elements of the penultimate epoch. During the 1990s and early 2000s, public health preemptions often limited the extent to which states could regulate behaviors such as smoking. Mowery et al. (2012) emphasize that tobacco regulation preemptions hindered social norm development surrounding the lack of desirability of smoking. Preemptions around COVID-19 responses, such as Mississippi’s Executive Order 1463, which disallowed local government actors from imposing additional social distancing limitations, may have a similar effect on social norms. Just like the earlier public health preemptions, other state COVID-19 preemptions may increase public health. In Maine, Executive Order 28 preempted local governments from adopting less restrictive stay-at-home orders. While the majority of COVID-19 preemptions to date have not taken this form, there is nothing inherently dangerous to public health about them; the content and form of the preemptions are the important features.

While COVID-19 preemptions are structurally and substantively similar to the public health preemptions in the 1990s and early 2000s, the format is different. The earlier preemptions were mostly enacted legislatively. In contrast, COVID-19 preemptions are predominantly issued via executive order, reflecting the emergency powers of governors. Governors in various states have used the wide power and discretion issued to them during emergencies to determine what local governments can and cannot do (Waugh 2007). It is generally thought that strong home rule powers serve as a counterbalance to state preemption (see Swanson and Barrilleaux 2020); however, it is unclear how this legal relationship operates when governors are afforded broad emergency powers. For example, it is the opinion of the South Carolina Attorney General that while municipalities retain their general home rule powers in emergencies, they are forbidden to exercise emergency powers because these powers are reserved for the governor (Cook 2020). In emergency situations, it is possible that, though legally unclear if, gubernatorial power could override home rule protections, defaulting to something more similar to Dillion’s Rule.

Such preemptions create path dependencies, limiting the ability of local governments to respond to emergencies in the expected bottom-up manner (Rubin and Barbee 1985; Schneider 1995, 2008). In times of crisis where policymakers’ decisions are literally questions of life and death, this is a dangerous void. Fast action, particularly by local governments, appears to be associated with a significant decrease in the rate of new COVID-19 cases (Dave et al. 2020). Previous responses to state preemptions, such as lobbying the state legislature, working with interest groups, and turning to constituents for support (Rutkow et al. 2019; Wagner et al. 2019), are unlikely to be feasible in the limited time environment of a pandemic. Policy path dependency suggests that without a long-term strategy, local governments will face similar constraints from their states in any future emergencies.

Conclusion

The COVID-19 pandemic produced unprecedented policy responses as officials at all levels of government reacted to the crisis. Traditionally, disaster management in the United States has operated from the bottom up (Rubin and Barbee 1985; Schneider 1995, 2008). Many governors, however, used their emergency powers to issue executive orders aimed at coordinating their state’s response to the pandemic and limiting what their local governments can do. These types of preemptions are reminiscent of the public health preemptions in the 1990s and early 2000s, but in a form not commonly seen. State preemptions are often legislative (Riverstone-Newell 2017) or judicial (Swanson and Barrilleaux 2020), but the COVID-19 pandemic has shown how the executive branch also participates in limiting the ability of local governments to make locally responsive policy. City actions fell off after states imposed shelter in place orders.

Further research should explore whether, and to what extent, state preemptions had a role in stemming local responses. In order to answer this question, more complete data is needed on the number and content of state preemptions and local responses to the pandemic. Information on local responses should focus on both policies and local government managers’ perceptions (Schuster et al. 2020). As the second wave of the pandemic emerges, a better understanding of the intergovernmental relationship between state and local governments, as well as the emergency response systems that help guide the relationships during a pandemic, is needed.

In addition to why states preempted (or did not preempt) their cities, an open question remains what consequences path-dependent preemptions will have in the future. Preemptions create uniformity within states, perhaps to the detriment of innovation and context-specific adaptation. We do not yet know what effect, if any, the preemptions had on the spread and mortality rates of COVID-19.

While we focus on emergency response and state preemptions, this is part of a broader pattern of states limiting local governments’ policymaking ability. Both the specific example of executive preemptions in the public health realm and broader trends of legislative and judicial preemptions should be a concern for public administrators and scholars. In the short term, these preemptions may spend valuable state relational capital vis-a-vis local government, relational capital that could be even more valuable in the future as the pandemic progresses. In the longer term, preemptions shift the relationship between states and their local governments and administrators will need to adjust. Local government managers have a responsibility to their constituents, and restrictions limit their responsiveness (Sances 2019). For these managers, balancing these responsibilities and constraints might be one of the most difficult administrative challenges during the pandemic and beyond.