key: cord-0067024-g00xv1y1
authors: Birkland, Thomas A; Taylor, Kristin; Crow, Deserai A; DeLeo, Rob
title: Governing in a Polarized Era: Federalism and the Response of U.S. State and Federal Governments to the COVID-19 Pandemic
date: 2021-08-14
journal: Publius
DOI: 10.1093/publius/pjab024
sha: 04ccb987bbdc6d8bf896bd93af1d9773e4733c26
doc_id: 67024
cord_uid: g00xv1y1

How does the state of American federalism explain responses to COVID-19? State-by-state variations to the COVID-19 pandemic illustrate the political dynamics of “kaleidoscopic federalism,” under which there is no single prevailing principle of federalism. In the COVID-19 pandemic, features of kaleidoscopic federalism combined with shortcomings in the public health system under the Trump administration, leading to fragmented responses to the pandemic among the states. Federalism alone does not explain the shortcomings of the United States’ response to the pandemic. Rather, the fragmented response was driven by state partisanship, which shaped state public health interventions and resulted in differences in public health outcomes. This has sobering implications for American federalism because state-level partisan differences yield different and unequal responses to the pandemic.

the country who wonder how one of the wealthiest and most technologically advanced countries could respond so poorly to this pandemic (Wike, Fetterolf, and Mordecai 2020) . But it would be misleading to consider the response to the pandemic as being uniform across the United States. The states' responses-and their outcomes-have shown considerable variation. As Benton (2020) argues, there is evidence that the states' response to the pandemic has worked as we would expect in a federal system, but there is also evidence that it is not working well, particularly when compared with other federal systems, like Canada (B eland et al. 2021) and Australia (Downey and Myers 2020) .

In this article, we argue that the uneven response to the COVID-19 pandemic in the United States can be attributed to two distinct but interrelated factors. The first of these factors is the accelerating trend toward the institutionalization of a style of federalism that has been described as variable-speed, go-it-alone, or kaleidoscopic federalism (Benton 2020) . All these labels for the current style of federalism in the United States share in common the acknowledgement of the increasing trend towards political polarization in both the electorate and among state officials, particularly, for our purposes, among governors. This means that state governments have, since the early 2000s, been more likely than in prior years to make policy decisions based on partisan considerations. We find that this change to the nature of American federalism predated the Trump Administration, with one important caveat: those factions within state governments that were aligned with the president ideologically seized on his disavowing of good public health practices to shirk key crisis management responsibilities.

A second important factor in understanding the uneven response to the pandemic is the presence of Donald Trump as the titular leader of the federal executive branch. Neither Trump nor his advisors had any particular principle of federalism on which they relied to shape their formation of domestic policy generally, and policy related to COVID in particular. Rather, the administration's policy decisions were almost entirely transactional, predicated on how recipients of the administration's attention and largesse could benefit the president politically. Furthermore, most of its staff had little theoretical or practical knowledge of public health or crisis response in a federal system. The Trump administration's sharp departure from the assumption of a strong federal role of the sort that would typically characterize a major pandemic response exacerbated the pandemic. While the Biden administration has taken more effective measures to confront this pandemic than the Trump administration (Stolberg 2021; Blake 2021) , the new administration will need to confront the kaleidoscopic style of federalism that emerged in the last decade, the ramifications of which will likely reverberate for years into the future and may prove injurious to crisis management in the public health as well as other policy domains.

should be most attentive (Foreman 1994; Garrett 2000) . Particularly in the wake of the September 11, 2001 terrorist attacks, the federal government redoubled its efforts to bolster the nation's preparedness capability, investing billions of dollars into a host of federal-state cooperative programs aimed at enhancing the nation's readiness for "all hazards," including public health emergencies stemming from bioterrorism or pandemics.

At the root of the preparedness system is the idea that the federal government, state governments, and private and nonprofit organizations share an interest in promoting public health. As Altman and Morgan (1983, 7) noted, the design of the public health system is a classic example of "marble cake federalism" in which the roles of the federal and state government are intertwined. The federal government employs thousands of scientists in agencies such as the Centers for Disease Control and Prevention, the National Institutes of Health, and the Public Health Service, all under the U.S. Department of Health and Human Services. These agencies undertake research, gather data, share good practices with states and local governments, and develop ideas and recommendations for health interventions. The federal government also has considerable fiscal power to induce states to adopt particular public health policies (DeLeo 2010).

Global disease surveillance is a responsibility of the federal government, led by the Centers for Disease Control and Prevention. The public health literature calls the CDC the "sentinel" for American public health (DeLeo 2014; Etheridge 1992), because it is responsible for identifying and tracking outbreaks of threatening infectious diseases before they make their way to the U.S. In the current COVIDpandemic, the CDC established an Incident Management Structure to begin U.S. response operations on January 7, 2020.

Despite these features of the design of the public health system, observers of the post-9/11 public health preparedness regime have similarly lamented the lack of clarity around public health emergency preparedness, noting "despite anecdotal reports suggesting that progress has been made, it is unclear whether these investments have left the nation better prepared to respond to a bioterrorist attack, pandemic influenza, or any other large-scale public health emergency" (Nelson et al. 2007 ).

As part of CDC's disease surveillance mission, the agency collects and publishes indicators of public health in its Morbidity and Mortality Weekly Report or MMWR. The dissemination of information on novel viruses in the MMWR is central to the national system of disease surveillance. Novel viruses are particularly concerning because the human population has no experience with such viruses, so there is no human immunity, and the disease can spread rapidly. This surveillance function of the CDC relies on scientific excellence for the gathering and dissemination of credible information vital to the public health system.

In this section, we provide a brief review of the state of American federalism prior to the COVID-19 pandemic. We note that the United States has moved away from a collaborative form of federalism to a more contentious form of federalism characterized by significant state-to-state variation in policy that is largely driven by partisanship. However, this move toward a more partisan form of federalism does not fully explain the particularly poor response to the COVID pandemic in the United States. After all, prior administrations, both Republican and Democratic, have taken effective action to prepare for and respond to potential pandemics, even as state-level partisanship has intensified. But cooperative federalism had begun to break down before the COVID-19 pandemic, and this breakdown is partially responsible for the highly variable response to the pandemic in the states and the overall poor performance of the United States as a whole.

But if cooperative federalism has broken down, what has replaced it? It is difficult to find one pithy metaphor to describe contemporary federalism in the United States. As Benton (2020 ) notes, Deil Wright (1978 , 1988 laid out a series of metaphoric phases of American intergovernmental relations, such as layer-cake, marble-cake, and picket-fence federalism, all characterized by, respectively, conflict, cooperation, and competition. The question that confronts federalism scholars in the current era is whether Wright's last phase of intergovernmental relationscontractive federalism-describes the current moment in American politics. This contractive federalism is characterized by fiscal constraint at the federal level. The problems of COVID-19 are not problems of fiscal constraint, primarily, but rather are problems of federal leadership and coordination of federal resources and state responses.

Given all the different labels that have been tried to characterize early-21st century federalism in the United States, we found Benton's (2018) idea of "kaleidoscopic" federalism most apt because this term "can be identified by metaphors such as fragmented, push-back, nuanced, fend-for-yourself, and collaborative" (Benton 2020, 537) . Benton argues that participants' perceptions of intergovernmental relations (IGR) in this period are characterized by "polarization, inaction, indecisiveness, convolution, and collaboration." Collaboration, in this case, means a greater reliance on inter-local and inter-state agreements, with less direction from the federal government. In simplest terms, "federalism and IGR are no longer either 'cooperative,' 'creative,' 'competitive," 'coercive,' or 'calculative.' All of these descriptions and approaches, as well as many others, depict the operation and relations of the federal system, even within the same policy area" (Benton 2018, 19) . There is not a single overarching form of federalism that characterizes all policy domains, so it is difficult, if not impossible, for students of federalism to say that the United States is characterized by one kind of federalism.

Another term in recent use in the federalism literature that describes the response to the COVID-19 pandemic is "fractured federalism," in which cooperative, innovative, and coercive federalism can be observed simultaneously in the same policy domain because of intense political partisanship and polarization (Bowling and Pickerill 2013; Goelzhauser and Konisky 2019) . From the 2010s onward, political polarization has caused the U.S. Congress to become less prescriptive in legislative actions, leaving states the ability to act with minimal coordination or guidance from Washington as federal preemption has waned (Goelzhauser and Konisky 2019; Kettl 2020) .

The effect of this fractured federalism is that states have adopted a patchwork of policy standards and implementation across the country (Bowling and Pickerill 2013) , in what Conlan and Posner (2016) refer to as "variable speed federalism." A notable feature of variable-speed federalism is that there are significant regional variations in politics and policy that align with regional partisan characteristics. Moreover, states push back on implementing federal policies because of political polarization and the associated level of agreement with current presidential administrations (Haeder and Weimer 2013) .

With all of this as background, it is not surprising that state responses to COVID-19 have varied significantly. The diversity of state responses to the pandemic and degree to which Republican-controlled states and Democraticcontrolled states followed similar patterns can be viewed as more of a continuation of recent trends in American federalism than a departure from them. In this regard, the variety of state responses to COVID-19 bear many of the hallmarks of state responses to the Affordable Care Act (ACA), in that under the ACA states have been able to elect to exceed federal standards and request waivers to deviate from baseline federal policy standards, resulting in a patchwork of policies and with significant regional variation (Conlan and Posner 2016) .

But we must not confuse the current state of intergovernmental relations or federalism-no matter how messy it is-with the inability of the federal government under the Trump Administration to act decisively when called upon to do so. The George W. Bush Administration responded to the outbreak of novel H5N1 avian influenza in Southeast Asia by requesting several billion dollars to fund pandemic preparedness programs. In fact, Bush's Secretary of Health and Human Services Michael Leavitt aggressively lobbied for a host of statutory changes aimed at bolstering the nation's capacity to manage a large-scale disease outbreak. These changes were contained in the Pandemic and All-Hazards Preparedness Act (PAHPA) (PL 109-417), which helped centralize public health emergency response activities within HHS, and the Public Readiness and Emergency Preparedness Act (PREPA) (PL 109-148), which aims to help shield vaccine manufacturers from liabilities arising from drugs created to combat a novel disease. All of this activity occurred despite the fact that not a single case of H5N1 avian influenza reached U.S. shores (DeLeo 2018).

The Bush Administration's pandemic planning later served as the foundation for the Obama Administration's response to the 2009 H1N1 swine influenza pandemic. While the pandemic was milder and less virulent than originally predicted, falling short of the kind of crisis represented by COVID-19, the Obama Administration's management of this threat was successful save for some problems in acquiring and distributing a vaccine. The virus is estimated to have killed roughly 12,000 Americans, far fewer than was originally projected. The President's Council of Advisors on Science and Technology estimated the virus could result in upwards of 90,000 deaths in the U.S. alone (McNeil 2010) . Public health officials applauded the Obama Administration for leading a response that was "informed by science and epidemiology" (Huang and Whitehead 2010). The Bush and Obama administrations' response to the threats posed by novel viruses suggest that the United States can prepare for and respond to these threats when there is commitment on the part of leaders in the federal government to prepare for these threats and to respond quickly when the threats first appear.

It is an article of faith in American politics that the federal government will join with state and local governments to support them during crises. The very foundation of federal emergency management policy in general is the idea that state and local governments are best positioned to respond to emergencies. When particularly large disasters happen that overwhelm local government capacity or when such crises span political boundaries, the federal government is generally expected to, at a minimum, provide material support for state and local governments.

Writing about whether federalism was at the root of the poor response to Hurricane Katrina, Birkland and Waterman (2008) asked "were the failures in Hurricane Katrina a result of federalism or of a particular style of federalism that characterizes disaster policy in the US? Or did these failures simply reflect the inherent difficulties in preparing for, responding to, and recovering from catastrophes?" (2008, 693) . They argued that it was not federalism itself that was at the heart of the problems of response to Katrina. Rather, they argued that the prevailing style of federalism, in which policy accreted without regard to the effect of those policies on institutional arrangements or on policy consequences, led to a poor response. Birkland and Waterman concluded that disaster response during Hurricane Katrina occurred during a period of "opportunistic federalism," in which "actors in the system . . . pursue their immediate interests with little regard for the institutional or collective consequences" (Conlan 2006, 667) . After the Governing in a Polarized Era September 11, 2001 terrorist attacks, and the idea of "homeland security" came into vogue, the federal government became even more involved in emergency preparedness, in a way that echoed the civil defense efforts of the 1960s. "Opportunity" in this case was therefore the opening created by the September 11 attacks for the federal government to assume leadership in the design of emergency management that focused on terrorism to the exclusion of at least three decades of experience about how to effectively manage federal-state-local relationships in preparing for and responding to emergencies.

The history of national disaster policy in the United States is increasing federal involvement in disaster preparedness, relief, response, and recovery. Combine this growing tendency for the federal government to become more involved in "homeland security" defined very broadly to include natural disasters and pandemics as well as terrorism, and one could reasonably assume that the federal government would take a strong leadership role, and that the states would, in broad terms, respond in a roughly uniform way to federal leadership and inducements.

In contrast to the Katrina case, this accretion of federal power and leadership is not what has happened in the COVID-19 pandemic, both because federal leadership and inducements were often absent or confusing, and because the sorting of states into "red" and "blue" states with strong ideological commitments meant that response to the pandemic was remarkably divergent. Here we explore these trends in some detail, in terms of three problems that have arisen in the response to the COVID-19 pandemic: incoherent federal policy leadership, partisanship and its effect on policy variation at the state level, and fractured federalism and its implications for state responses. We find is that, as seen during Hurricane Katrina, it is not federalism per se that explains the relatively poor response to the COVID pandemic in the United States. Rather, we find that the current style of federalism, characterized by partisanship that creates greater stateby-state variation in the response to the pandemic, combined with ineffective federal leadership to create a much worse response than might have otherwise been expected.

We examine the role of the federal government through the lens of policy implementation in an intergovernmental system because there were existing pandemic response plans and procedures in place that were designed to coordinate federal and state policy activity. One model of effective policy implementation in the federal system (Goggin et al. 1990 ) holds that, for policies to be effectively implemented, federal policy must be perceived as credible by the states, the federal government must clearly communicate the policy to the states, and the states must be both receptive to the policy and possess the capacity to implement the policy.

In contrast with the Obama and Bush administrations' responses to emerging diseases, the Trump administration's leadership was remarkably poor. If the Trump Administration had any organizing principles about federalism and intergovernmental relations, it was unclear. It is unlikely that Donald Trump had any particular vision of federalism beyond what Bowling et al. (2020) call "transactional federalism," which is characterized by "a market-based model of intergovernmental relationships in which the core feature is a form of market exchange. This approach to federalism has become more obvious during the national response to the COVID-19 pandemic" (Williamson and Morris 2021, 8) . In this case, the exchange is not financial as much as it is political loyalty. This was manifest in Trump's political appointees' interference in the production of the data on the pandemic in the Morbidity and Mortality Weekly Report, where these officials sought to change language that they viewed as politically damaging (Diamond 2020) .

The Trump administration was broadly characterized by executive incompetence (Mayer 2021; Pfiffner 2021; Howell and Moe 2020) . This incompetence was manifest in severe cuts in the number of CDC officials stationed in China whose job it was to detect emerging diseases (Taylor 2020) , even as these cuts were viewed with great concern by the public health community. In this way, even before the pandemic, the Trump administration did not treat global public health seriously.

In terms of the response to the pandemic, Trump and his staff's inexperience in governance, their manifest lack of knowledge of basic constitutional structures and norms, and fundamental managerial incompetence, yielded what could best be characterized, as Ross Baker (2020) puts it, as "laissez faire federalism." Baker notes that the early federal response was marred by "a claim on total authority by Trump paired with a total abdication of responsibility." The former president's relationship with states was particularly confrontational and antagonistic (Harkness 2018) . The result was a policy context for intergovernmental implementation in which states simply refused to implement federal guidance that they did not want to accept (Bowling and Pickerill 2013) , effectively leaving states to "fend for themselves" (Balz 2020) . Indeed, the Trump administration's lack of organizing principles around federalism and its failure to understand the intergovernmental public health system led the federal government to compete with states, and the states compete with each other, for essential equipment and supplies at a time when the federal government should have been coordinating with states, not competing with them (Timmer 2020; Rose 2020) . The absence of federal political leadership generally and coherent policy leadership in the COVID-19 pandemic specifically is reflected in the considerable variation in response among states. States like New York, Massachusetts, New Jersey, California, and Michigan were proactive in implementing policies to control the transmission of COVID-19 during the early stages of the pandemic, thereby reducing the rate of cases and fatalities, while states like Kentucky, Florida, Wisconsin, and South Carolina have been highly restrained in their policy response. In late September 2020, only four states-Colorado, Connecticut, New Jersey, and New York-met at least four of the five public health criteria used to determine whether it was safe to begin reopening economies. Another twenty-five states (including Massachusetts, Michigan, and California) met two or three of the criteria. The remaining states met none or only one of these criteria (Lopez 2020) .

But if the story ended here, this would be a story of initial fumbling by the Trump administration as it sought to get on the right foot and respond to a crisis that grew faster than anyone had anticipated. Unfortunately, the story does not end here. First, the administration, particularly President Trump, never considered the idea of infectious disease as a serious matter. It never developed a national strategy for dealing with this virus (Kettl 2020, 599) ; in contrast, the Biden administration took office with such a strategy in hand (Stolberg 2021) . The Trump administration's failure to act is even more shocking considering that they had been explicitly warned of the likelihood of a significant pandemic. Former national security advisor Susan Rice explained in a New York Times op-ed that the outgoing Obama administration worked through a tabletop exercise focused on a potential pandemic with the incoming Trump officials, whose interest in the topic was variable at best (Rice 2020) . Coupled with the rate of staff turnover in the administration and its remarkably slow pace in appointing officials to key government positions Toosi, Lippman, and Diamond 2020) , this disinterest ensured that the federal government at the highest levels would be understaffed and inexperienced. While civil servants in key federal agencies such as the Centers for Disease Control and Prevention and the National Institutes for Health worked tirelessly to develop public health interventions to slow the progression of this pandemic, their efforts are often frustrated by the malfeasance of the political appointees that are charged with directing these agencies.

Second, federal pandemic response policies were not clearly communicated to the states. The Trump Administration failed to use its bully pulpit to communicate with the public in a consistent and predictable way about the most effective protective measures for stemming the spread of the virus. Consistent, clear, and frequent communication are all essential during a crisis, but particularly when so much uncertainty exists as it does with COVID-19 (Weible et al. 2020 ). Indeed, "With his impatient demands and decrees, Mr. Trump has disrupted efforts to mitigate the crisis while effectively sidelining himself from participating in those efforts" (Burns, Martin, and Haberman 2020) .

Perhaps the most visible manifestation of this failure was the administration's position on mask wearing. Most public health authorities argued that wearing a face covering would have a substantial effect on slowing the spread of coronavirus infections. If, as Richard Neustadt (1990) has argued, the power the presidency is the power to persuade, then one of the most formidable powers the president has is to communicate to the public about efforts it can take to join in a national effort to address a problem. As late as September of 2020, in a town-hall meeting aired on ABC television, President Trump cast doubt on the effectiveness of masks (ABC News 2020), and the previous week his campaign held a rally in Nevada which ignored local rules on mask wearing and social distancing (Sanchez 2020 ). Under such circumstances, even on those occasions when the president drew on expert advice to tell Americans about what to do about the pandemic, his message was often undercut by his own behavior. This is made more complex by the tendency, at the federal level and some state governments, to view the response to the pandemic as an individual responsibility rather than as something for which government could be mobilized (Klein 2020) . President Trump therefore abdicated his responsibility and the power of his bully pulpit to influence positive behavior choices and left the states and career civil servants at the state and federal levels to fill the vacuum of leadership and public communication.

Third, the Trump administration was remarkably lacking in attention to, use of, and concern for the complex actions required during crisis management, even so far as failing to use existing mechanisms to respond to COVID-19. The Trump administration refused to use the authorities given to it by the Defense Production Act to purchase medical equipment or supplies for distribution to the states (Jacobs 2020). Along the same lines, the federal government did not seek to coordinate the acquisition and distribution of needed equipment, but instead actually competed with states to obtain equipment that was then not efficiently distributed to the places that needed it most. This exacerbated the problem by placing the states in competition with both the federal government and with each other in acquiring essential equipment, rather than using the federal government market power to amass as many resources as needed to aid the states and localities in addressing the crisis. All of these problems would have been far less acute had the administration taken seriously the warnings developed in at least two prior presidential administrations of the threats posed by novel diseases.

The Trump administration also failed to develop and implement a wellfunctioning testing strategy. Most public health authorities argued that in the absence of mass vaccination, the best way to "flatten the curve" of the COVID-19 pandemic is to improve testing and contact tracing. Testing was one of the techniques used by South Korea, for example, to control the spread of the virus. The Trump administration stumbled early in the production of such testing kits (Rice 2020) . The CDC attempted to create test kits for distribution of the states, but their effort was unsuccessful. Contributing to the overall lack of access to testing was the Trump administration's go-it-alone strategy wherein it chose not to adopt the World Health Organization's testing kits, leading to an overall lack of testing supplies across the country.

Under the ideal operation of the shared federal-state public health system, we would expect to see some uniformity in the measures taken to address the pandemic, because there is a community of public health experts who share information across states and whose approaches to emerging disease threats would be expected to be reasonably uniform due to similarities in education, causal beliefs, and professional experience (Dunlop 2009 ). However, because of the changing nature of federalism, state policy interventions are inextricably linked to political ideology, creating the pattern of variable speed federalism described above (Conlan and Posner 2016) . While the idea of state political ideology influencing the rate of policy adoption is not necessarily new (Berry and Berry 2017; Gray 1973; Walker 1969) , the earliest empirical studies thus far undertaken of the response to the pandemic find that, notwithstanding the severity of the pandemic or the broad consensus among public health experts about what measures should be taken to address the pandemic, state-level partisanship shapes responses to the pandemic.

The first measure of differences among states comes in the decision to declare a state of emergency. Declaring a state of emergency empowers governors to take extraordinary measures to protect public health and safety, such as stay-at-home orders. Fowler et al. (2021) developed an event history model in which the dependent variable is the date the emergency was declared; they control for other variables such as the proportion of the population over age sixty and population density, both of which are related to higher mortality and morbidity. They found that "Democratic governors were about 2.67 times quicker to declare a state of emergency than Republican governors, which largely conforms to our expectations of a partisan gap in responses to the pandemic" (Fowler et al. 2021, 9) . They also found that Republican governors were quicker to adopt emergency declarations in states where Trump was less popular, which suggests that these governors responded to the specific political climate in their states.

Stay-at-home orders were the second major policy intervention, aimed to limit community transmission of COVID-19 to avoid overtaxing hospitals, and were among the earliest interventions adopted by states. Kettl (2020, 596) argues that the decision to adopt a stay-at home-order was central to efforts to control the pandemic. Gusmano et al. (2020) found that there was a "statistically significant relationship between having a Republican governor and issuing a stay-at-home order" (Gusmano et al. 2020, 384) . Republican states, if they issued such an order at all, did so later than did states with Democratic governors. All eight states that initially failed to issue stay-at-home orders were Republican led. The results of their analysis held when the usual controls were applied, such as population density, hospital capacity, people living in poverty, or people without health insurance. That there is a partisan effect at work here that has its roots in existing policy choices is starkly illustrated by Kettl's finding that twenty-eight of the thirtytwo states that locked down in March were Medicaid expansion states under the ACA, while seven of the eleven states that did not lock down were Medicaid expansion states.

A third public health intervention that received widespread attention was the mandate that individuals wear masks in public. While there was some early confusion about the wisdom of promoting masking wearing, public health officials offered guidance which showed that masks were effective in reducing the spread of the virus, thereby protecting public health, not simply the individual mask wearer (Bai 2020) .These mandates were, like all other interventions, unevenly applied across the states, with some states adopting such a mandate quickly and some adopting late, if it all. Here, again, the party affiliation of the governor is consequential. Adolph et al. (2020, 1) found that "the governor's party affiliation was the most important predictor of state differences in the timing of indoor mask mandates." Using an event history model (Cox proportional hazard model), these researchers found that "the marginal effect of having [a] Republican governor instead of a Democrat was a 29.9-day (95 percent confidence-interval: 24.6 to 35.2) delay in the announcement of broad state-wide mask mandates." They found that this variable has the largest effect of any that they studied; they wrote that "by far, the most powerful predictor of broad mask mandate adoption and timing is the political party of the Governor" (7). The delay in a mask mandate was even greater in states with a Republican governor and a conservative state ideology; such states delayed mask mandates by 38.1 days compared to states with a Democratic governor and a liberal ideology. The result "is robust to many different sensitivity analyses testing a large number of possible confounders" (2) such as the differences in the source of daily COVID-19 indicator data, including sources of data on morbidity and mortality, Trump vote share in the 2016 election, the rate of adoption in neighboring states and in states they identified as "peer" states, as well as measures of population density, relative state wealth, and size of the elderly population.

The findings are consistent across multiple studies: that even when controlling for the usual confounding variables, such as age, degree of urbanization, or state wealth, states led by Republican governors are, by and large, slower and less likely to adopt stringent public health measures that would "flatten the curve" of COVID-19 cases compared with their Democratic counterparts. Of course, none of these state-by-state differences would matter much if there was no discernible effect of partisanship on morbidity and mortality. But early research shows that these partisan differences are related to outcomes. Neelon et al. (2021) found that, in the early stages of the pandemic-March to early June 2020-states led by Democrats had higher case rates of COVID-19 than did Republican-led states, in large part because coastal states led by Democrats, such as Washington, California, and New York, were the most likely places for the emergence of a disease considering that these states admit a large number of travelers from overseas. But after early June 2020 and onward, case rates were higher in Republican-led states. The analysis considered confounding variables relating to population density, size of the elderly population, poverty, and the prevalence of health conditions such as obesity that increase risks of serious illness from . This result was that, while COVID-19 hit Democratic-led states earlier in the pandemic, the death rate in Republican-led states began to exceed that of Democratic-led states on July 4, test positivity in Republican-led states exceeded that of Democratic-led states on May 30, and testing in Republican-led states started to lag behind Democratic-led states on September 30.

What explains this difference in state-by-state handling of the COVID-19 pandemic? We argue that there is a schism in American governance and public health in general, and with respect to the coronavirus in particular. Under this schism, we see many (albeit not all) state public health agencies seeking to work with federal agencies, like the Centers for Disease Control and Prevention, to track disease and to send effective messages to citizens on ways to protect themselves and to slow the spread of the coronavirus. As previously noted, the growing trend in U.S. federalism has been partisan polarization (Bowling and Pickerill 2013; Conlan and Posner 2016; Goelzhauser and Konisky 2019; Kincaid 2017) .

Polarization and partisanship in American governance is manifesting in two ways. First, there is a debate between the governors and the highest levels of the executive branch of the federal government over authority to adopt policies designed to slow the spread of the virus. These state policies, such as mask requirements, restrictions on social gatherings, partial closure of schools, businesses, and other activities, were anathema to the federal political response to the pandemic during 2020. President Trump, who de-emphasized effective government management as unimportant during his time in office, instead focused on the politics of state policies, as opposed to their efficacy as public health interventions. The president's well-known opposition to wearing face masks was at odds with state guidelines, often leading to very public conflict between governors and the president, such as when Trump visited a Ford factory in Michigan without wearing a mask (Karni 2020) . In addition, governors were expected to personally express gratitude and praise the former president for fulfilling basic requests for supplies at the start of the pandemic (Ronayne and Lemire 2020) as a form of transactional federalism. In many ways, President Trump's politicization of mask wearing was more about the president's perception of public support for him personally, than it was a public health measure (Bender 2020) . However, such politicization of public health measures further fractures our federal system's efforts to respond to the pandemic when state requirements to wear masks are transformed into powerful policy symbols of a partisan divide. This is particularly troublesome when federal agencies continue to issue guidance that measures such as mask wearing are effective at slowing the spread of the virus.

Second, there are divisions between states and regions on the best responses to the pandemic. So-called red states and blue states have taken divergent policy positions on the best ways to respond to the coronavirus. In any federal system, particularly in the United States, we would expect differences between states across a range of policy issues, but the COVID-19 pandemic is a transboundary crisis that spans health, the economy, education, housing, and so on, and that affects all the states. While the severity of the pandemic was uneven from March 2020 to March 2021, it appears to have posed similar boundary spanning policy problems in states like New York and Texas alike. Such boundary-spanning issues are not addressed well by disparate state policies that can work at cross purposes. In many ways, the U.S. federal system is at the apex of political polarization between states and the federal government and between states themselves. In this context, the notion of "variable speed" federalism means that conservative red states with conservative ideologies that align politically with President Trump, concluded that masks should not be mandatory, schools should open, and that bars and restaurants should operate at full capacity while the pandemic raged on.

The differences in timing and severity of the pandemic notwithstanding, there are noteworthy differences in state policies to curb the virus, like mask wearing and physical distancing, that can be explained by ideology. States like New York and California have required mask wearing while states like Texas and Florida have lifted mask mandates while the severity of case counts remains similar across all four states. In the absence of federal leadership in the early months of the pandemic, regional groups of states joined COVID-19 alliances (Rogers, 2020) . The early alliance of Pacific Coast states, which later included Colorado and Nevada, formed the Western States Compact (Office of Governor Gavin Newsom 2020). The states pledged not to re-open before science deemed it safe to do so. These compacts evolved to be venues for sharing policy ideas, lessons learned, and mobilizing resources between states. There were similar compacts in the Northeast (Associated Press 2020), Midwest (Van Berkel 2020), and Southeast.

The regional alliances had four broad approaches. First, the early alliances in the West and Northeast, all led by Democratic governors, arose to focus on datainformed policy decision-making, coordination in policy responses, sharing of lessons learned, and assistance in resource mobilization to respond to the pandemic. Second, the regional alliance among bi-partisan governors in the Midwest coordinated reopening of economies after the initial stay-at-home orders. Third, Southern states, led by Georgia Governor Brian Kemp, pushed back against the idea that government restrictions were the best approach to responding to the coronavirus pandemic, seeking to jump start state economies (Trubey, Bluestein, and Capelouto 2020) . Fourth, later in the pandemic response, a bipartisan alliance arose among Governors to tackle specific problems like testing and contact tracing (Kelly 2020) .

Any assessment of the United States' response to the COVID-19 pandemic will inevitably confront the role of federalism in the national response. As we have reviewed in this article, preparedness for and response to a pandemic is a shared responsibility between national and state governments. The current state of federalism has made this response particularly challenging. Recent federalism scholarship has found the since the Obama administration, partisanship has played a greater role in shaping American federalism than it had in the last half of the twentieth century (Conlan and Posner 2016; Goelzhauser and Konisky 2019; Kettl 2020) . The states have been sorted by partisanship, with policy styles in each state that reflect, quite strongly, the national political goals of the parties that dominate each state. In the case of the COVID-19 pandemic, this partisan polarization yielded divergent policy and health outcomes. In this article, we have argued that the severity of the pandemic in different states and at different points in time during the pandemic has highlighted a confounding problem of kaleidoscopic federalism: how to untangle the effects of political polarization from the inherently challenging nature of severe, boundary-spanning policy problems.

In simplest terms, the research so far shows that states that were more ideologically liberal and led by Democratic governors adopted policy interventions earlier, and at a greater level of stringency, than did more conservative states led by Republic governors whose responses were often late and less stringent, if policies were adopted at all. At the same time as state-level responses diverged based on partisanship, these tendencies were exacerbated by a style of top-level federal policy leadership that ranged from incoherent to nonexistent. States were not provided the level of support for managing pandemic disease in the way that the public health system was originally designed. Federal policy goals and objectives for responding to the COVID-19 pandemic were unclear and lacked credibility, thereby leaving the states with insufficient support for mounting effective immediate responses. The politicization of policy responses-most notably mask wearing-further muddied policy messages and intergovernmental relations in a way that undermined states' authority and ability to respond, particularly in states that did not align politically with the White House. We do acknowledge that, in many cases, clear, coherent messages were being sent from federal executive branch agencies, such as the Centers for Disease Control and Prevention and the National Institutes of Health, to the states through normal public health channels. But politicization of the high-level response in the Trump administration exacerbated partisan differences among and between states.

The change in administration in 2021 provides a natural opportunity to assess this extent to which the problems of response were primarily due to feckless national leadership rather than problems posed by contemporary federalism. With the change in presidential administrations in January 2021, the federal government became more supportive of active state policies to stem the pandemic and simultaneously became more actively engaged in pandemic response actions. This support is more firmly grounded in sound science and managerial competence that it was in the prior administration. The messages sent from Washington are therefore more credible than they were only months prior. This return to something more like normal national leadership reveals the degree to which the Trump administration's approach deviated from typical crisis governance in the United States.

In the end, we do not find that federalism itself is at the root of the overall poor response to the pandemic. The fundamental design of the public health system is generally sound, and in prior administrations it has performed reasonably well. Rather, we find that the poor response to the COVID-19 pandemic in the United States is the product of the coincidence of a managerially and politically incompetent presidency and the current kaleidoscopic style of federalism that, to a degree seldom seen in modern American politics, reflects significant differences in policy based on state partisanship. This confluence of partisan polarization and federal managerial incompetence have worked together to yield a response to the pandemic that ranked among the world's worst in the first year of the crisis.

Trump's ABC News Town Hall: Full Transcript. The President Answered Voters' Questions Tuesday in Philadelphia

Governor Partisanship Explains the Adoption of Statewide Mandates to Wear Face Coverings

Anti-Federalist Federalism: American 'Populism' and the Spatial Contradictions of Us Government in the Time of Covid-19. Geographical Review 1-18

The Role of State and Local Government in Health

Joins 6 Other Northeast States in Post-Pandemic Economic Compact. WBUR

Still Confused About Masks? Here's the Science Behind How Face Masks Prevent Coronavirus

Trump's Laissez Faire Federalism has Made for Failed

As Washington Stumbled, Governors Stepped to the Forefront

Social Policy Responses to COVID-19 in Canada and the United States: Explaining Policy Variations between Two Liberal Welfare State Regimes

Economy in WSJ Interview

Intergovernmental relations in the early twenty-first century: Lingering images of earlier phases and emergence of a new phase

Challenges to federalism and intergovernmental relations and takeaways amid the COVID-19 experience

Innovation and diffusion models in policy research

Is federalism the reason for policy failure in hurricane Katrina?

Analysis j Biden's Strong Coronavirus Numbers Reinforce the Folly of Trump's Approach

Seeking Patterns in Chaos: Transactional Federalism in the Trump Administration's Response to the COVID-19 Pandemic

Fragmented federalism: The state of American federalism 2012-13

Flattening the COVID-19 Curves

Cooperative federalism and hydraulic fracturing: A human right to a clean environment

As Trump Ignores Virus Crisis, Republicans Start to Break Ranks

From cooperative to opportunistic federalism: Reflections on the half century anniversary of the commission on intergovernmental relations

American federalism in an era of partisan polarization: The intergovernmental paradox of Obama's "new nationalism

Anticipatory-conjectural policy problems: A case study of the avian influenza

The centers for disease control and prevention: A policy perspective. In Guide to U.S. Health and Health Care Policy

Indicators, agendas and streams: Analysing the politics of preparedness

Trump Team Failed to Follow NSC's Pandemic Playbook

Trump Officials Interfered with CDC Reports on Covid-19

Of babies and bathwater: Why the clean air act's cooperative federalism framework is useful for addressing global warming

Federalism, intergovernmental relationships, and emergency response: A comparison of Australia and the United States

Policy transfer as learning: Capturing variation in what decisionmakers learn from epistemic communities

Sentinel for health: A history of the centers for disease control and prevention

Plagues, products, and politics: Emergent public health hazards and national policymaking

Pandemics and partisanship: Following old paths into uncharted territory

Betrayal of trust: The collapse of global public health

The state of American federalism 2018-2019: Litigation. Partisan polarization, and the administrative presidency

Implementation theory and practice: Toward a third generation

Public health theory and practice in the constitutional design

Innovation in the states: A diffusion study

Partisanship in initial state responses to the COVID-19 pandemic

You can't make me do it: State implementation of insurance exchanges under the Affordable Care Act

Federalism is Broken. Can it be Fixed? Governing

The political economy of cooperative federalism: Why state autonomy makes sense and 'dual sovereignty' doesn't

Implementing modern public health goals through government: An examination of new federalism and public health law

The role of new federalism and public health law

Presidents, populism, and the crisis of democracy

Comparing the H1N1 Crises and Responses in the US and China

Despite Claims, Trump Rarely Uses Wartime Law in Battle Against Covid

Michigan Visit, Trump Forgoes Criticism and Talks About the Economy and the Flood

Seven governors join deal in pursuit of first multistate coordinated testing strategy. CNN

Flattening-the-Curve Associated with Reduced COVID-19 Case Fatality Rates-an Ecological Analysis of 65 Countries

States Divided: The Implications of American Federalism for COVID-19

From Cooperative to Coercive Federalism

Introduction: The Trump Interlude and the States of American Federalism

There are no good choices: In shifting so much responsibility to individual people, America's government has revealed the limits of individualism

Rich Countries Other than the U.S. Have Coronavirus under Control

Just 4 States Meet These Basic Criteria to Reopen and Stay Safe

The Random Walk Presidency

U.S. Reactions to Swine Flu Apt and Lucky

Associations between Governor Political Affiliation and COVID-19 Cases, Deaths, and Testing in the United States

Conceptualizing and defining public health emergency preparedness

Presidential power: The politics of leadership from FDR to Carter

Dying in a leadership vacuum

Colorado & Nevada Join California, Oregon & Washington in Western States Pact

Donald Trump and the Norms of the Presidency

The Government has Failed on Coronavirus, but there is Still Time

State Alliances are Leading the US Fight Against Covid-19, Wired

Flatter or Fight? Governors Seeking Help Must Navigate Trump

A 'War' For Medical Supplies: States Say FEMA Wins by Poaching Orders. National Public Radio

No social distancing and few masks as crowd waits for Trump rally in Nevada

Why Outbreaks like Coronavirus Spread Exponentially, and How to 'Flatten the Curve

Biden Unveils National Strategy that Trump Resisted

Exclusive: U.S. Slashed CDC Staff inside China Prior to Coronavirus Outbreak

It's Becoming Clear Why the US' Response to COVID-19 is Terrible. Ars Technica

Before Trump's Inauguration, a Warning: 'The Worst Influenza Pandemic since 1918

Kemp, Southern Governors Talk coronavirus reopening strategies

Minnesota Gov. Tim Walz Joins Compact with Midwest Governors to Reopen Economy

How Many People has the Coronavirus Killed?

The diffusion of innovations among the American states

COVID-19 and the policy sciences: Initial reactions and perspectives

Image Plummets Internationally as Most Say Country Has Handled Coronavirus Badly. Pew Research Center's Global Attitudes Project (blog)

Cooperative federalism and the clean air act: A defense of minimum federal standards

Lessons from the COVID-19 pandemic for federalism and infrastructure: A call to action

Understanding intergovernmental relations

Understanding intergovernmental relations

Face Masks in the US: Why Guidance has Changed so Much -CNN. CNN

This research was written by the four authors but is also part of a larger team project, the Risk and Social Policy Working Group, https://www.riskandsocialpolicy.org. This research was supported in part by a RAPID grant from the National Foundation, No. 2030316, Rob DeLeo, PI. 1. https://covid.cdc.gov/covid-data-tracker/#cases_totalcases Governing in a Polarized Era