Building the Plane While Flying: Informal State Capacity,Policy Autonomy,and Public Health

Abstract

Public health infrastructure varies widely at the local, state, and national levels, and the COVID-19 response revealed just how critical local health authority can be. Public health officials created COVID policies, enforced behavioral and non-pharmaceutical interventions, and communicated with the public. This article explores the determinants of public health capacity, distinguishing between formal institutional capacity (i.e., budget, staff) and informal embedded capacity (i.e., community ties, insulation from political pressures). Using qualitative data and interviews with county health officers in California, this article shows that informal embedded capacity—while difficult to measure—is essential to public health capacity. It concludes by relating public health capacity to broader issues of state capacity and democracy.

On February 28, 2020, the Bay Area of Northern California recorded its first instance of COVID-19 community spread—and just over a week later, recorded its first COVID-19 death. At the same time, a small band of Bay Area government officials worked quietly to develop a large-scale, public health intervention that would test the extent of state capacity in California counties. Importantly, these public health officials were not simply crafting a public health response, but a whole of government response—a collaborative effort across government agencies to address the impending public health emergency. To maintain order and protect the general welfare of populations, states and counties would have to marshal a level of state and public health capacity not needed in recent memory, and this capacity would need to be activated and engaged for years rather than months.

Over the following months and years, health officers implemented an extraordinary range of interventions including shelter-in-place orders, testing and tracing programs, quarantine and isolation protocols, masking requirements, processes and procedures for the opening and closing of businesses and public spaces, policies to address income, food, and housing insecurity, as well as infrastructures and procedures for vaccine allocation and delivery. However, the ability of local health agencies to develop, implement, and enforce such policies was far from automatic and far from consistently present. Local jurisdictions were often building the capacity for these state duties as they went. As public health capacity varied across time and space, so too did the effectiveness of policy development and implementation, and, ultimately, the scale of cases, hospitalizations, and death. This article examines how and why public health capacity and the COVID-19 policy response varied across California counties.

It is critical to understand the elements of public health capacity and the factors that facilitated or hindered the development and expansion of public health capacity during the COVID-19 pandemic. The limited understanding that existed about the elements that constitute public health capacity is, perhaps, mostly clearly demonstrated by the Johns Hopkins University Global Health Security Index (GHSI), which in 2019 ranked the United States as the country best prepared to “prevent, detect, and rapidly respond to” epidemic and pandemic threats. Despite a deeply flawed response to COVID-19, the United States maintained its number one ranking in the GHSI in 2021. What was, therefore, thought to be important in understanding and measuring public health capacity prior to the COVID-19 pandemic proved incomplete. In addition to focusing primarily on formal, institutional measures, including staffing and financial resources, national indices also fail to capture and explain policy development and implementation that takes places largely at the state and local level. This is particularly problematic for developing an understanding of public health capacity because much of the COVID-19 policy response, like public health policy, more generally, was developed, implemented, and enforced at the state and local level. The pandemic highlighted how slippery the concept of capacity can be for scholars interested in understanding how bureaucrats formulate and implement policies. This is true both of the different measures of good governance or preparedness that existed prior to the pandemic, as well as conceptualizations and measurements that have been created or refined in an effort to understand variations in a state's ability to meet its fundamental goal of protecting public health.

This article examines public health capacity at the county-level in California. Through qualitative analysis and in-depth interviews with public health officers about the COVID-19 policy response across California, we find that the focus on formal, institutional measures of capacity fail to explain county-level outcomes and miss critical factors that both helped and hindered an effective policy response. Instead, we argue that public health capacity contains both elements of formal and informal capacity. Formal capacity includes more commonly considered elements such as staff size, expertise, and physical infrastructure, but we also highlight the importance of informal and semi-formalized structures that facilitated communication and learning within and across government entities, as well as between state and non-state actors.

In this article, we argue that a focus on formal, institutional measures of capacity capture only one dimension of state action. More specifically, we show that a focus on more formalized measures undervalue the importance of three factors critical to effective public health policy outcomes: within-government cooperation, across-government cooperation, and social embeddedness. The partnerships established within and across units of government, and between government and non-state actors, can in themselves produce state capacity. We demonstrate how previous considerations of public health capacity neglect to consider how a state's relational, iterative engagement with non-state actors creates vital partnerships—or, alternatively, undermine trust—between states and communities that are essential to a robust public health response. Furthermore, we show how collaboration and partnerships both within and across government units is critical in policy learning, policy development, and implementation. Such collaboration is not an automatic part of or product of formal bureaucratic institutions. Rather, such collaborative capacity within and across units of government is often a product of past shared experiences with policy formulation and implementation. Informal capacity is not, therefore, only something that is generated outside of government through partnerships with non-state actors, but also refers to the informal collaborative structures between government actors.

A siloed approach to measuring and understanding public capacity, one that looks only within a public health department, is incomplete. While there are any number of ways to study whether or not capacity matters for pandemic outcomes, this article aims instead to study how capacity matters and how it is generated or realized. We hope to better understand local governments as a unit of analysis: given that localities have the most intimate and direct contact with communities, inductive theorizing based on local governments can inform broader scholarly debates about the relationship of states and societies. A focus on local capacity can also help us understand policy areas that are highly decentralized, such as public health. ¹

This article is organized as follows. The “Beyond Capacity: Embeddedness and Public Health” section discusses theories of state capacity, with attention to formal and informal concepts of capacity. For the study of public health, formal indicators of capacity often fail to capture the actual dynamics of public health policy development and implementation. “The Structure of Public Health in California” section describes the structure of public health in California, with particular attention to the statutory and administrative powers of the public health officer. In section “Informal Capacity and County Public Health Officers,” we turn to informal capacity itself, highlighting three ways informal state capacity matters: (1) collaboration within government; (2) collaboration across government units (i.e., across counties); and (3) embeddedness with local communities. Drawing on in-depth interviews with health officers across California, we show both the limitations of many current conceptions of public health capacity as well as the promise of thinking more rigorously about embeddedness and engagement as critical components of state capacity. ²

Beyond Capacity: Embeddedness and Public Health

State Capacity as Formal, Institutional Capacity

State capacity refers, very generally, to the ability of states to make and implement decisions. Michael Mann's seminal work on infrastructural power describes state capacity as the ability to “penetrate civil society, and to implement logistically political decisions through the realm,” ³ while Skocpol refers to capacity as a state “implement[ing] official goals, especially over the actual or potential opposition of powerful social groups.” ⁴ Other definitions of state capacity reference governing, and the need for states to devise policies or deliver services. ⁵ There is a gap, however, between concepts of state capacity and operationalization of those concepts. Measuring state capacity at the national level usually involves examining bureaucratic institutions, particularly their formal duties, their fiscal resources, and their insulation from politicization. The World Bank Governance Indicators, for example, disaggregate governance into components such as the quality of a country's public services, the quality of its civil service, and the independence of the civil service from political pressure. ⁶

Many of these indicators of state capacity point to the potential for capacity by describing whether or not administrative institutions are given the rules, autonomy, and resources to act. County Health Officers in California, for example, are given significant statutory power, are situated within organizational structures meant to provide autonomy, with many offices possessing significant financial and personnel resources, yet public health officers have been reluctant to, or even barred from, implementing policies they know to be necessary and effective. In other words, the institutional position and resources of county health officers in California gave health officers tremendous potential for capacity, but they still often found themselves, in the words of Mann and Skocpol, unable to implement policies across their realm. The organizational structures and personnel resources were not sufficient to develop and implement a robust public health response.

Martin Williams (2021) highlights how capacity and implementation are often elided, with implementation taken as a sign of capacity. In such instances, if a policy is implemented or a particular outcome achieved, it is retrospectively determined that the government or agency had capacity. It can be difficult, therefore, to analyze the reasons for poor policy implementation in states identified as having high levels of formal capacity, particularly when the indicators of capacity are aggregated and national. Hypotheses of capacity, Williams argues, overlook “how complexity and contingency can be even more powerful determinants of policy implementation,” as evidenced by numerous studies of implementation at the local level. ⁷ There is, however, a danger to explaining away important elements of capacity by improperly categorizing potentially replicable factors as “contingent.” The ability, for example, for a single health officer to draw on a cross-national professional network to produce early COVID-19 modeling, or the importance of the unique characteristics of a particular county counsel to the success of a Bay Area public health officer, were important in shaping public health capacity directly. While these seemingly one-off and unique explanations of capacity might be dismissed or minimized as contingent, it is, we argue, possible to draw broader lessons and insights from these episodes about the mechanisms that facilitate borrowing of capacity or the organizational structures that foster collaboration and the realization of capacity. What becomes evident in the examination of county-level public health responses in California is that what might at first be categorized as a “contingent” factor in the realization of state capacity may, in fact, be a replicable aspect of informal capacity or a semi-formalized organizational structure that can be built and enhanced to help fully realize the potential capacity of a public health officer or department. One implication of our argument is that if a source of informal capacity can be replicated, it may also be possible to make it more formal and more evenly distributed across counties.

It is especially difficult to tease out the conceptual ways that capacity matters, particularly the relationship between capacity and implementation. Does strong capacity matter if states cannot implement policies, or can effective implementation mask weak capacity? This literature is occasionally teleological, inferring capacity with respect to the outcome (i.e., a state was strong because it accomplished a certain goal). It is occasionally endogenous, when capacity is built up in response to an outcome. And it may also be latent, with capacity being a necessary but not sufficient condition to bringing about an outcome.

As Kavanagh and Singh have noted, cross-national variation in capacity did not explain early pandemic responses, and the focus on administrative agencies and institutional design neglected to consider how factors such as federalism or partisanship might affect policymaking. ⁸ Further, Bosancianu et al. note any number of challenges associated with applying theories of state capacity to a pandemic:

…classic accounts of state effectiveness typically assume that governments know, or can know, what policies are optimal (or at least sensible). Without such knowledge, the importance of the state's ability to act becomes less clear. Second, some of the relevant policies in this case—such as implementing distancing provisions—require compliance but not state strength as we often understand it. ⁹

Importantly, Bosancianu et al.'s observations highlight the need to include the consideration of policy formulation in addition to policy implementation when conceptualizing capacity. As they write, this is particularly true when operating in a novel policy environment rife with many unknowns. In addition, Bosancianu et al. highlight how a conceptualization of capacity like Mann's may be incomplete when applied to a policy realm like public health in which educational campaigns are a key tool and where state enforcement is often lacking or non-existent. While traditional capacity indicators may not tell us everything about a public health agency's ability to respond effectively to a crisis, they are nonetheless important. States and localities with dedicated public health departments—rather than just health departments—are more likely to be able to formulate effective pandemic responses. Further, public health departments with bigger budgets, more epidemiologists and scientists on staff, and in-house resources, like laboratories, are more likely to be effective, all else being equal. Finally, public health departments that can make decisions independently, and whose policies are not subject to rollback or undermining by elected officials, are more likely to be effective. What has become evident, however, in the policy response to COVID-19, is that the existence of such formal measures of capacity, for example, the number of epidemiologists on staff or the salary of top administrators, is not a guarantee that formulating effective policies will result in the implementation of those policies. What the experience of California's counties demonstrates is that there are important informal and semi-formal mechanisms by which the potential capacity contained in such formal measures can be fully actualized in formulating and implementing public health policies.

Informal Capacity and Embeddedness

State capacity includes not only the formal institutions of the state, but also relies on the relationships of state and non-state actors. In his consideration of developmental states, Evans emphasized the importance of embedded autonomy, and described the need for bureaucrats to set long-term goals while working with critical stakeholders, like businesses, to achieve them. ¹⁰ In public health, it is particularly critical for departments to work well with social partners in order to both formulate and implement community-level policies and shape individual behavior.

When a whole government response is not evident or is met with broad dissatisfaction, studies have also shown the importance of community capacity. Community capacity is characterized by strong civil society networks that operate without formal leadership, organizations, or centralized resources. In a study of Hong Kong's initial COVID-19 response, Hartley and Jarvis show that a combination of non-governmental actors, including social movement organizations and non-profits, helped to coordinate policies in the absence of effective state action. ¹¹ Citizens distributed masks, disinfected buildings, and provided information about the virus through social media, reflecting a capacity built at the community level. One lesson from California is that overall public health capacity can be enhanced when a county invests in community capacity, thereby helping to build up civil society organizations that can become an independent source of capacity. In contrast to Hartley and Jarvis and the case of Hong Kong, however, the case of California demonstrates how community capacity and civil society organizations can also synergize with and enhance state capacity—not just act independently of or in the absence of state leadership.

Outside the context of public health, embeddedness is seen as an important component of a bureaucrat's ability to provide public goods ¹² or to implement the policies of the executive at a local level. ¹³ These studies view embeddedness as the degree to which a bureaucrat shares traits, such as co-ethnicity, with the community they serve. They find that bureaucrats who are from the localities they govern, or who share the same ethnicity, are better able to implement policy, in large part because the community is more likely to trust these bureaucrats and to see the bureaucrats' decision-making as aligned with the needs of the public. There are important parallels between the enhanced trust gained from this type of embeddedness and the trust that can be gained from embeddedness in public health. The state-society relationships between public health departments and community-based organizations help generate a similar type of trust that is critical to effectively communicating and implementing public health measures at the community-level in the United States. Further, local bureaucrats can sometimes outsource policy implementation to non-governmental agents or organizations in ways that create what Jessica Rich terms “shadow capacity,” particularly when bureaucrats are trying to avoid red tape or restrictions on state action. ¹⁴ These concepts help us understand that there are factors beyond formal institutions that facilitate service delivery, which is particularly important in the realm of public health.

Public Health Capacity

As the COVID-19 crisis has shown, the United States' capacity to do things effectively at the national level has little impact when the federal government delegates policy formulation and implementation to the states. ¹⁵ The Constitution's silence on public health, together with the tenth amendment, has long-placed public health responsibilities and authority at the state level, leaving specific interventions on a vast set of policies that includes communicable disease management, violence prevention, emergency preparedness, food safety, and substance use and abuse to the localities. ¹⁶ The federal government can create guidelines, priorities and best practices, as well as provide resources, but it is the responsibility of the states and localities to craft and implement policies. The role of policy implementation and development becomes even more difficult for states when the federal government is slow to provide resources and guidance, or worse, actively undermines public health efforts.

At a minimum, public health capacity does require formal state capacity. Local public health agencies need clearly-delegated powers, funding and resources, and professional staff. At the start of the COVID-19 pandemic, the decades of underinvestment meant counties faced a public health deficit both in terms of resources and in the trust and collaborative relationships that can only be built through prior policy programs. This type of state capacity—both the formal and informal capacity—at the local level helps to explain why counties had more successful policy response and better public health outcomes than others.

In a measure of local public health capacity, the Centers for Disease Control and Prevention (CDC) designates six determinants that begin to identify the existence and importance of both formal and informal public health capacity, but the CDC does not explicitly use such labels or differentiate between the different forms of capacity. The determinants include preparedness planning and readiness assessment, surveillance and epidemiology capacity, laboratory capacity, health alert network (communications and information technology), risk communication and health information dissemination, and education and training. ¹⁷ While some of these relate to formal state capacity, others relate instead to the ability of the state to work with non-state actors, particularly when disseminating information or working through partners. This type of measure does not go far enough in explicitly identifying the informal aspects of capacity and largely miss the political autonomy that enables the full realization of the powers delegated to public health officers and the full use of their resources.

While public health capacity relies extensively on the state's relationship with other actors in order to achieve its stated policy goals, the structures or factors that facilitate such relationships are not often explicitly included in considerations of capacity. Public health services are divided among the state and external groups, including private health providers and non-governmental organizations. Public health agencies must be able “to inform, influence, communicate, and collaborate” with other actors in order to be successful. ¹⁸ Local public health capacity is informed by a local jurisdiction's existing medical and health services, ¹⁹ as well as community health partnerships. ²⁰ Empirically, public health systems that show high levels of formal capacity—that is, the public health agencies that have expertise to formulate policy, and oversee policy implementation—combined with high levels of coordination with local organizations outperform systems that share capacities across governmental and non-governmental groups. ²¹

Unlike health systems alone, public health relies on partnerships (with communities, and among sectors) and communications when implementing policy. ²² These communities and community-based organizations are not just enablers of effective public health service delivery; instead, they are crucial elements of public health capacity. Institutions are often more effective when underlying community social capital is strong, ²³ but informal capacity also captures the endogenous development of effective policy implementation assisted by intergovernmental cooperation and community engagement.

In a review of public health capacities at the national and regional levels, Aluttis et al. lay out reasons that partnerships are just as important to capacity as factors like financial resources or workforce- and knowledge-development. ²⁴ They emphasize the interorganizational relationships between governments and health care providers, academic institutions, the private sector, and socio-economic groups. They also emphasize collaboration with other public health bodies across different levels of government, and coordination across all these actors to address health inequities in particular. The partnerships are more effective when they are formalized across the public and private sectors, through policies and legal mechanisms.

Despite recognition that partnerships between governments and non-state actors are important, however, most studies of public health still focus on observable characteristics of relevant public agencies and the populations they serve. Studies show that the level of centralization of a public health department matters (i.e., how much authority is contained within the agency, versus spread across state and non-state actors). This type of focus can, however, overlook the very nature of public health responses, namely that an effective public health policy response requires a whole government response and partnerships with non-state actors. In assessing the public health policy response in California, we identify key institutional features and relationships that promoted successful organizational and collaboration across single units of governance (i.e., across offices and departments within a county), across units of governance (i.e., between public health offices of different counties), and between state and society (i.e., between public health and community-based organizations or academic centers).

Constraints on Formal Public Health Capacity

There are reasons to expect low levels of state capacity around public health even in affluent liberal democracies. Public health is chronically underfunded, and often suffers from conflicting mandates and priorities. Without a robust public health policy environment preceding the pandemic, public health officers, at times, began the pandemic with limited connections to their communities, and, therefore, limited relationships and trust, as well as populations with limited public health education. The National Academy of Sciences declared the state of American public health “in disarray” as far back as the 1980s, ²⁵ with tensions existing between politicians and public health officials. Public health itself is also subject to boom-and-bust funding cycles in response to crises, shifting agendas, and interest group concerns. Joshua Sharfstein and Georges Benjamin noted that public health infrastructure in the United States is notoriously weak, since it relies on collective action that is hard to mobilize. ²⁶ The successful operation of public health agencies often goes unnoticed, while its failures can have significant consequences.

On capacity indicators, there were reasons to think that the United States could respond effectively to the pandemic, and also signs that the United States was vulnerable to one. The CDC was, until very recently, considered one of the most effective agencies of the federal government and the foremost public health agency in the world. But when the COVID-19 pandemic hit, instead of marshaling the public health resources that existed, the president and administration burned precious capital by sidelining CDC experts, amending CDC reports, and politicizing the science behind and response to the pandemic. The former director of the CDC, Rochelle P. Walensky, described the “dramatic” mistakes made by the CDC in handling the COVID-19 pandemic and called for the agency's reorganization. This move by Walensky is, itself, a fairly direct acknowledgment of disconnect between perceived or potential capacity and the real ability to implement and enforce effective policy.

After the 2008 financial crisis, public health budgets were further slashed nationwide. Epidemiologists, public health leaders, and medical professionals warned of staffing shortfalls and constrained budgets. ²⁷ As the pandemic began, even the largest and most robust public health departments were drawing on expertise from well-outside government and well-outside California. The early pandemic response was characterized by extensive borrowing of capacity and reliance on professionals who were drafted into the service of county public health. Counties, for example, relied on volunteer and retired nurses to do contact tracing and staff phone lines. The deficit that counties found themselves in at the beginning of the pandemic was exacerbated by the simultaneous reduction of resources and expansion of responsibilities. The responsibilities of public health departments, for example, grew to include Medicaid expansion, preventative medicine, chronic disease management, violence prevention, and responding to crises like the opioid epidemic.

Finally, these departments are often poorly insulated from county politics and the pressure of elected officials. The public health response was immediately politicized, and elected officials intervened in the decisions of public health officers at all levels of government. Informal capacity may therefore be critical to effective implementation when elected officials target bureaucrats.

The Structure of Public Health in California

While the COVID-19 pandemic is unlike previous public health crises in the United States, we use a study of local public health capacity in California to better understand the dynamics of public health capacity. ²⁸ The institutional features of local health departments are fairly consistent across California's fifty-eight counties. The structures of local governance are also similar; counties are governed by elected boards of supervisors. Supervisors appoint the health officers of each county department of public health. The county-level health departments in California are all designed to be autonomous; health officers “have immense power to act independently in the interest of public health,” since their “legally binding directives” do not require consent from political officials. ²⁹

By controlling, to a large degree, the institutional design of local public health departments, we can better understand the extent to which formal capacity—such as budgetary resources and staff expertise—explains variation in public health responses. We can also examine the relationship of local departments and social partners, or the degree of cooperation or conflict between departments across county lines. In other words, an investigation of California counties allows us to more clearly identify the factors that transform potential capacity into actual capacity and the implementation of effective policies.

California is the most populous state in the United States, home to the world's fifth-largest economy and 39 million people. Three of its metropolitan areas have populations over four million people, including the Los Angeles metropolitan area, which is the nation's second largest. California is also racially and economically diverse, with no racial group constituting a majority of the population. ³⁰

The California Department of Public Health (CDPH) sits under the California Health and Human Services Agency. It has an annual budget of $3.5 billion, and runs six centers of health operations, including the Center for Infectious Disease. The history of public health governance in California is local, as it is in many other states. In the nineteenth century, the state delegated responsibility over health and care of the indigent to counties, many of which built public hospitals to care for the poor. California was the second state to establish a Board of Health, which it did in 1870. Together, state and local counties responded to disease outbreaks, including bubonic plague in 1900 and the aftermath of the 1906 San Francisco earthquake.

In 1995, California reorganized its health and safety code. The legislature passed SB 1360, which streamlined public health administration by delineating the responsibilities of the state and local health departments. The bill was the result of a yearlong review of public health, including a collaboration with health officials in the public and private sectors. SB 1360 created California's State Department of Health Services to oversee public health and health policy. This reorganization received support from the County Health Executives Association of California, the California Coalition of Local Health Department Nursing Directors, the Western Consortium for Public Health, the California Public Health Foundation, the Office of Statewide Health Planning and Development, and the California Conference of Local Health Officers (CCLHO). ³¹

Local public health authority is vested in fifty-eight counties and three city health departments, totaling sixty-one local public health departments. Each county board of supervisors is responsible for appointing its own health officer. Health officers must be physicians, and they have 171 enumerated duties—one of which includes “broad authority to take action to prevent disease.” ³² They can issue isolation and quarantine orders, and can declare local health emergencies. The Health Officers Association of California, in a guide to counties on appointing health officers, recommends appointing physicians with expertise in preventive medicine or public health.

On matters related to public health, the statute stipulates that the “board of supervisors of each county “shall take measures as may be necessary to preserve and protect the public health.” ³³ The mandate for public health is broad; the responsibilities of the health officer include managing leaks from gas pipelines, ensuring sanitation of prisons, and designating nonprofit food agencies to coordinate food donation efforts. ³⁴

The main duties vested in the health officers are to give orders related to public health and sanitation. Further, the county health officer “may take any preventative measure that may be necessary to protect public health” during any state of emergency or local emergency. While these preventive measures are not directly specified by the statute, they include “abatement, correction, removal or any other protective step.” ³⁵ Again, in looking across California counties, all of which are theoretically given immense formal power, we are able to identify and examine the less formalized structures and relationships that explain the actual public health capacity exhibited.

While local health officers operate independently of one another, they are all part of the CCLHO. They meet with the state-level CDPH on a monthly basis, and have two Continuing Medical Education sessions per year that focus on public health. ³⁶

Funding for public health in California comes from the federal government, state government, and local taxes. Since the 2008 financial crisis, the state has cut funding for public health. Per-capita spending on public health declined 18% between 2008 and 2019, from $97 to $79 per person. ³⁷ Per capita public health spending ranges from a low of around $40 to a high of $951, in low-population Alpine and Sierra counties. The median is $98 per person per county, with a mean of $150. Expenditures on public health as a percentage of a county's total budget range from a minimum of 1.8% to a high of 11%, with an average of 3.9% of a county's budget going to public health.

At the time of the first shelter-in-place order, there were 258 reported cases of COVID-19 in the Bay Area, with four deaths. ³⁸ The CDC had documented the first case of community spread of COVID-19 in Santa Clara County in late February; on February 25, San Francisco Mayor London Breed declared a state of emergency. By March 11, Santa Clara and San Francisco counties banned large gatherings. The events leading up to the shelter in place order involved constant discussions between the health officers of the Association of Bay Area Health Officers (ABAHO) about the best way to respond collectively to the impending pandemic. The story of the Bay Area's early shutdowns was extensively covered by the national press, often in comparison to how the virus was being managed on the East Coast. ³⁹

Informal Capacity and County Public Health Officers

While formal capacity is critical to the ability of bureaucrats to do their jobs—to create and implement policy guided by specialization and expertise—formal capacity alone does not ensure that policies can be formulated and then implemented with any degree of certainty or success. In order to move beyond a conception of the potential for capacity, which Martin Williams argues is what formal measures capture, we define three dimensions of public health capacity that make it more likely bureaucrats can not only develop policy, but also implement it. These are (1) within government collaboration; (2) across government collaboration; and (3) social embeddedness. We draw on interviews with fifteen health officers who served across seventeen counties in California during the COVID-19 public health emergency, as well as reporting by the press and nonprofit health organizations to lay out what these dynamics are, and how they contribute to successful policy implementation. Our interviews included broad regional variation, with interviews conducted with health officers from the Bay Area, Central Coast, Central Valley, High Sierra & Desert Region, Northern California, and Southern California. In addition, our interview population included health officers from urban, suburban, and rural counties, as well as from some of the largest and smallest counties in California. This qualitative approach is critical to understanding informal capacity: looking at the directives promulgated by public health officers or at county-level outcomes do not tell us the process through which health officers actually accomplished their objectives.

These dimensions of informal capacity may not be surprising to anyone who has experienced local government, yet they are strikingly absent from the scholarly literature on state capacity. Further, these components of informal capacity do not simply enhance formal capacity. Instead, they can either create capacity where the state has few resources, or they can thwart bureaucratic objectives where formal capacity is strong. These dimensions of capacity are, in other words, just as important, if not more so, than formal capacity measures, particularly when the state wants to achieve an outcome that relies on behavior and compliance rather than relying on the force of law and direct enforcement. This is why we do not focus on COVID-specific outcomes (i.e., death rates, hospitalizations) but instead on the assessments and experiences of those public health officers at the center of California's policy response and on whether or not county citizens largely complied with health ordinances as they evolved over time.

Within-Government Cooperation and the Realization of Public Health Capacity

One repeated refrain that came from health officers from across California, from counties big and small, rural and urban, and from the Bay Area, to the Central Valley, to Southern California was that the only effective COVID-19 response was a whole government response. An effective COVID-19 response required the mobilization and coordination of multiple departments, collaboration between public health and county counsel (the county's top lawyer), and a significant increase in the interactions and interface between public health officers and Boards of Supervisors. So despite health officers being granted wide latitude to issue orders and make policies to protect the public health, particularly during a public health emergency, collaboration across county government was a necessity. The accounts of public health officers make clear that while the unilateral power of public health officers could not, in theory, be circumscribed by politicians (although the health officer can be removed by the Board of Supervisors), such a scenario only existed as theory.

When public health officers were empowered to act in a manner at odds with the Board of Supervisors, or to act without or ahead of the preferences and directives of the Board, that power often emanated not from formal statutes, but from collaboration and reliance on the County Counsel (the county's top lawyer) and the county's Chief Administrative Officer (CAO). Beyond the political autonomy that such collaboration between the health officer, counsel, and CAO could generate, within county and across agency collaboration was critical for both developing and implementing a county's COVID-19 policy response. While such collaboration was critical to the COVID-19 policy response, California counties varied widely in the extent to which health officers felt they had relationships across agencies within their counties.

Administrative Partnerships, Policy Learning, and Political Autonomy

At the beginning of the pandemic, many health officers were only vaguely aware, or even unaware, of the full latitude granted their office. One public health officer, for example, noted the good fortune of having a family member who was an attorney and who could provide guidance as to the legal boundaries of the public health officer's duties. This level of uncertainty about the bounds of public health power at the beginning of the pandemic was both a product of high levels of turnover among health officers prior to and during the pandemic, as well as a product of the unprecedented policy response that was required by the pandemic. Even the longest-serving, most highly expert public health officers described their pandemic experience as “building the plane while flying.”

Perhaps the most important factor and relationship influencing a county's COVID-19 response was the relationship between the public health officer and the Board of Supervisors. Public health officers described a range of experiences with Boards of Supervisors, including some who enjoyed strong relationships, describing a sense of support and appreciation from the Board, while others had antagonistic and conflict-ridden relationships, with supervisors openly undermining public health officers and orders. Still some public health officers experienced the full roller coaster of support and antagonism over the course of the pandemic. Interestingly, in almost all cases, even those characterized by an overriding sense of support from the Board, public health officers were at their most effective both in developing and implementing public health policies when the county counsel (the county's top lawyer) and/or the county's chief administrative officer provided political and policy cover from the Board. As one Bay Area public health officer described it, the county executive acted as “heatshield” from the Board of Supervisors, while the approach of the county counsel's office was very much tell us what you need to do, “And we'll help figure it out. How to do it legally.” It was not, as the public health officer detailed, a relationship in which the county counsel was there to “tell you all the things that you can't do.” This same public health officer described never having to consult the Board of Supervisors on public health policy. The message from the county counsel on the development and implementation of public health policy:

It's not their job. That is your job. That's the health officer's job…they can fire you if they don't like what you're doing. But they are not the public health expert. And they do not make public health decisions. The health officer makes public health decisions.

This was a uniquely empowering relationship between a public health officer and a county counsel, but versions of this type of political protection and empowerment of the public health officer by county counsel were present in those instances where public health officers were most effective in policy development and implementation. The full realization of the public health officers' statutorily defined policy capacity was, therefore, only possible in the presence of partnerships and collaborations that were not formally defined or structured. While some may dismiss this type of dynamic as contingent, as being dependent on the unique personalities that happen to occupy a particular government position at a particular time, these types of collaborative relationship can also be built and fostered by past policies and the long-running collaborative policy partnerships that characterize some counties. Policy collaboration on lead paint abatement, on opioids, on tobacco policy can all build a legal-policy relationship between a county's legal team and public health team. As such, long tenures for both the public health officer and county counsel, as well as an active and existing public health law policy agenda, are important to generating this type of effective collaborative partnership.

Where autonomy was absent, public health officers saw the curtailment of their most basic policymaking and public health tools. In one Northern California county, for example, immediate political blowback ensued after the decision to close schools in response to the first possible cases within the school district. In these first critical months of the pandemic, the Board of Supervisors in this particular Northern California county barred the public health officer from communicating with the superintendent of schools, while also pressuring the public health officer to refrain from communicating to the public about the increasing risk of COVID-19 in the Spring of 2020. In the face of political pressure and the curtailment of their duties, this particular public health officer also did not have the support of the county's chief executive, who the public health officer described as unsupportive of the need to declare a public emergency. Indeed, even possible avenues of support were stymied, with the county executive discouraging communication and collaboration between the public health officer and county counsel. In what is an indirect indication of the importance of such partnerships for effective COVID-19 policy, those actors who sought to minimize the COVID-19 response took aim at impeding the exact collaboration between a public health officer and county counsel that produced more robust responses in other counties.

Direct relationships with the Boards of Supervisors were not always confrontational and did not always result in compromised public health orders and interventions, but in conditions in which reliance on those direct relationships were necessary for effective public health policymaking, public health policy was more turbulent. As one public health officer from Southern California remarked, “it's critical for the health officer … to have a close relationship with the Board of Supervisors,” adding that this was especially important and effective when the relationship could exist, at least in part, behind closed doors, away from the more heated public scrutiny that characterized much of COVID policy debate. Effective communication and partnership between public health officers and supervisors was often most possible when the engagement could take place outside of the formal meetings of the Board of Supervisors, which were more prone to high levels of political pressure and confrontation. Public health officers referenced engaging with supervisors in “non-Brown Act meetings,” which refers to requirements established by the Ralph M. Brown Act that any meeting that involves a majority of the members of a legislative body be public. In meetings that were not “Brown-acted,” public health officers felt it was more possible to provide granular updates, provide overviews of the COVID response, updates on the emergency operation center, and get stakeholder input from the board in a less politicized environment. Smaller meetings of this type did not replace open, public meetings, but offered another venue for additional information sharing and partnership. Providing a balance between openness to the public and protections from political scrutiny in the policy development phase is difficult to achieve but important to promote within government partnerships and policy learning.

The importance of the Board and how varying levels of political interference and opposition directly influenced COVID policy response was often most clear when elections resulted in ideological shifts among the Board members. Another unique opportunity for assessing political influences on public health policy arose in instances in which a single public health officer served as the public officer for multiple counties simultaneously, which happened on several occasions during the pandemic. What the same public health officer was able to achieve and how they described the policy environment in the county with a more supportive board of supervisors, as opposed to a county whose board members were ideologically opposed to public health interventions, provides a fairly clear view of the limitations of a public health officer's formal sources of policy autonomy, and the importance of generating alternative mechanisms for policymaking autonomy and authority.

As a joint health officer of two counties described, one Board turned over in 2020 to be much more opposed to COVID restrictions. This health officer had support from one county, which was more cooperative and had mobilized more government workers and public health staff towards the COVID effort, while the other county was less willing to assemble the teams necessary for a full response. One county ended up dispatching workers, particularly nurses, to the other, all the while facing a political backlash for implementing any COVID policies there.

Building Capacity and Partnerships through Prior Disaster Response

For some counties and their health officers, an important alternative source of policy capacity and political authority came from the broad experience of responding to the climate change-related natural disasters and emergencies that have characterized California's recent history. This experience in responding to disasters such as wildfires, flooding, and mudslides also generated the emergency response expertise that was necessary in the initial year of the pandemic. Many health officers spoke of tapping into preexisting emergency procedures, which allowed them to operate by a playbook defining timing, roles, and responsibilities for disaster management. In one county that had suffered many natural disasters in recent years, including wildfires and mudslides, the health officer had worked closely with public health and emergency response. The public health officer had written quarantine orders with the county counsel, and learned which departments had excess medical and health capacity that could be diverted for the COVID response.

In areas where the health officer had little experience with prior pandemics, there were few ways to suddenly and quickly create relationships de novo. One health officer who joined the county a few months into the pandemic had participated in exercises in emergency planning, but had no way to really understand the many manuals and plans related to those exercises. Instead, the health officer described defaulting to the Incident Management Team structure. The incident management structure is the traditional model used from FEMA and emergency services that details the mission, the operations, and the logistics, as well as goals each day. This public health officer argued that “even if it's in writing somewhere, it's so different when you finally put it into action”: preexisting relationships and structures make it possible for the health officer to accomplish their tasks. Natural disasters also helped create regional cooperation—after wildfires, for example, Contra Costa County had worked with the North Bay to set up shelter operations and mutual aid, or worked with fire partners, public safety police, and Red Cross to manage wildfire effects within the county.

One public health officer from a rural county had experience with aspects of emergency management including Incident Command Systems, Medical Health Operational Area Coordinators, and Disaster Medical Health Specialists, and emergency operations manuals, and explained that public health is often the support agency, rather than the lead agency, in emergencies. Where counties had other robust social supports, health officers could tap into them. In one rural county, the health officer described leaning on the “super agency” of Health and Human Services (Adult Protective Services, Child Protective Services, Calworks (unemployment)) in order to facilitate writing and implementing policies. Another health officer in a rural (or, as they described it, “frontier”) county had an extensive medical and epidemiological background. In their time as the health officer of this county, they had led the emergency response to wildfires. When COVID began, this public health officer easily managed to put together a whole government response that required coordination across almost all agencies of the county. However, given anti-government sentiment among the county's residents (and even some of its employees), the public health officer also brought in an outside emergency management consultant to assist with efforts to coordinate a response and repurpose individuals across agencies.

Communications and Relations Across Agencies

One of the important aspects of managing the COVID outbreak was communicating with the public, but many California health departments did not have robust communications infrastructure. They therefore needed to rely on press conferences with other members of the government, sometimes alongside elected officials. One health officer described using daily press conferences to include guests such as school superintendents or behavioral wellness experts, or the sheriff, to facilitate the public's awareness beyond COVID-specific information. Some counties had public information officers but no communications infrastructure, and needed to work within existing institutions like schools to get information to families and communities.

Some counties used environmental health to help with enforcement—like inspectors at restaurants—but “public health generally [works] by recommendations … all those orders still rely on the majority of the population to voluntarily comply.” Health officers described expanding communications infrastructure within the health department, adding more people to conduct conference calls with local hospitals and doctors, with the faith community, with elder care advocates and nursing homes. This helped close a “a real gap of information” among stakeholders, especially about infection control practices and PPE—“we created a toolkit and a rotating team that went around to nursing homes and assess their PPE practices or infection control practices and provided training.”

One county went so far as to create a TV show that brought in community leaders discuss public health; other health officers set specific, recurring times they would appear on local television to provide updates and guidance. Because not all residents had access to television, particularly in rural counties, health officers also went out into the community. They utilized behavioral health organizations conducting outreach to native or migrant populations, and emailed public health briefs if people registered online. In one county that lacked postal delivery, a representative would deliver information at the post office, where people typically picked up mail in the afternoons.

Health officers also stressed the costs of a lack of within-government coordination. One noted that such cooperation may quickly arise in disasters, but the key to achieving consistent outcomes across other goals—like increasing health equity or access to health care—also depends on these ongoing relationships. Health officers have suggested that the state should help formalize coordination within and across government through training and onboarding of health officers, or assigning mentors within the CCLHO. These relationships help overcome the red tape that otherwise characterize bureaucracy, facilitating contracts and memoranda of understanding necessary in delivering policies.

Across-Government Cooperation

“[H]aving those collective minds to work together to design a response—that was huge.” The collective minds described here by one Bay Area public health officer were those brought together under the institutional structure and auspices of the ABAHO. Regional collaboration was repeatedly identified by public health officers as crucial to policy learning, policy development, and policy implementation during the pandemic. This was true not only for long-standing, institutionalized organizations like ABAHO, but for regional, cross-county collaborations between health officers that were far less institutionalized or even non-existent at the beginning of the pandemic.

Across the state of California, health officers are united by the CCLHO. The CCLHO brings together health officers twice a year at statewide conferences, and provides other opportunities for information-sharing. A lobbying arm, the Health Officer Association of California (HOAC), represents the interests of health officers in Sacramento. However, our interviews revealed that county-level cooperation (or antagonism) between counties was far more important than the statewide organization. There was also intergovernmental conflict between the state (the CDPH), which took over large parts of COVID planning (though not implementation or enforcement) in the summer of 2020, and the counties, which were then responsible for carrying out the state's mandates and explaining the state's reasoning to local citizens.

A foremost example of successful regional cooperation was the six counties of the Bay Area. The ABAHO was founded in the 1980s to coordinate a response to the HIV/AIDS epidemic. Its creation was sparked by the indifference of the federal government and the scope of the public health emergency in the Bay Area. ⁴⁰ The institutionalized collaboration continued, with ABAHO working together on initiatives including flu preparedness and exercises around the H1N1 outbreak. Their drills led to a joint policy that provided an off-the-shelf template for a coordinated response in early 2020, as county officials began monitoring the virus in Wuhan. In late February, six health officers of Bay Area counties began conference calls about the upcoming St. Patrick's Day festivities; ⁴¹ they toyed with the idea of restricting the number of people who could gather at one time. The weekend of March 14–15, the public health officers of ABAHO enlisted the help of their county counsels to write a shelter-in-place agreement using the exact same wording that would govern all of the ABAHO jurisdictions.

Where networks existed prior to the COVID pandemic, health officers had an immediate community of professionals and peers. One leading Bay Area public health officer immediately identified regional collaboration within ABAHO as a critical part of the response in early 2020. Relationship with non-government partners was a key source of early modeling (of viral spread), but when it came to the question of putting that modeling in context and translating the modeling into policy, those decisions were made in conversation with regional public health officer colleagues. The ability to game out scenarios with regional partners was important to the Bay Area's ability to move early—a move that was important for the Bay Area, California, and the nation. In addition to the early collective problem solving and communication, regional public health officers wrote health orders together, shared data, provided each other with political cover, relied on one another for support, and promoted policy alignment so that citizens traveling across counties (particularly for work) could expect consistency from one place to another. Where networks did not exist, or where they were even defined by competition or hostility, health officers were hampered in their ability to develop and implement policies.

While there were no regional groups that were as similarly long-existing or deeply institutionalized as ABAHO, there were other similarly important regional collaboratives, some that existed prior to the pandemic, while other regional collaboratives were formed only months after the beginning of the pandemic. One rural county health officer, for example, described a regional group that started a few months into the pandemic. Once it began, the counties worked together on policy development. One motivation for the regional collaboration even in the smallest, more remote counties was the fact that residents often worked and lived in different counties, creating a real need for sharing data and creating some level of uniformity in policy development that would help reduce confusion and ease compliance. So even in regions where no regional body existed prior to the pandemic, among the smaller counties of Alpine, El Dorado, Amador, for example, informal collaboration in early 2020 soon led to the formalization and regular meetings between county public health officers. As one rural public health officer described their approach when cases first started to climb in their area, “as we needed more information, as we needed more conversation with other health officers and other health departments … we developed both those individual contacts as well as the regional contacts so that we could have communication and bilateral exchange of information in that area. And part of the problem was developing consistency, of course.” It was in this context that this region began more formalized collaboration and bi-weekly calls. In these counties bordering Nevada, regional collaboration also spilled across borders, with the regional group in California joining up with the Quad County Healthcare Coalition in Nevada.

In another example of a regional association that was started during the pandemic, one set of counties, all of which included, in some part, a large national park, the health officers used their preexisting relationship from governing the park to create a public health consortium across agencies and counties. As described by one of the participating health officers, each county maintained its own Emergency Operations Center and its own incident response, but the formal association “allow[ed] us to coordinate policy and procedure across four counties in the park.” This regional association, like most others, established weekly meetings, set up a planning group, but also established a Multi-Agency Coordination Group, which established coordination not just between public health officers, but a broader government collaboration including county CAOs, board chairs, as well as leadership from the national park. It was, as the public health officer described it, an important mechanism for sharing “big policy issues in our region.”

Interestingly, the membership of this national park-related regional association overlapped with another regional association that pre-dated the pandemic. The San Joaquin Valley Public Health Consortium, while pre-dating the pandemic, gained new importance and expanded infrastructure during the pandemic. In the earliest, almost pre-pandemic days or early 2020, the Consortium, like ABAHO, held weekly calls to share data, information, and early ideas for public health interventions. Between the Consortium and the national park-related association, the data sharing allowed the public health officers to “make data-driven decisions for the region based on regional-level data.” In the early days of the pandemic the counties were able to make this Consortium more formalized with funding from the Blue Cross Foundation, which allowed them to hire an executive director and later hired a half-dozen staff and established standing committees—one of which focuses on epidemiology. The organization lives on after the end of the COVID-19 public health emergency, with a focus on health equity and rural care, and with formalized partnerships with the remaining parts of the national park-related association. This health officer emphasized the difficulty of coordinating information across the federal government (which oversees the park), the counties, and the state; the consortium does not press for the development of identical regional policies, but does generate much more information-sharing.

Many health officers described a lack of regional cooperation as an impediment to policy implementation. It was difficult to convince residents of one county to comply with mandates when they differed in a neighboring county, and health officers described the frustration of working alone—for example, on setting up testing-and-tracing capacities, or COVID dashboards.

Finally, the discrepancy between state mandates and local autonomy was a constant issue over the course of the pandemic. The initial COVID response was highly localized, with the CDPH somewhat embattled; by August 2020, the state's health officer, Sonia Angell, had resigned. Many health officers spoke of the challenges they faced over the course of 2020–2021, when the state's policy was to issue mandates that counties could make stricter, but could not loosen. Health officers felt that they were tasked with explaining the state's policies to confused (and often angry) residents, with little help from the state: “we felt like we had to justify the decision of the state where we didn't even know the decision-making process.” Others described an erosion of trust between states and localities over time. This is not to say that the state never provided political cover to the counties or that health officers were not at times grateful for their partnership with the state, but California's COVID policy response demonstrates that the collaborations and partnerships that facilitated information sharing, policy development, and efforts to achieve policy clarity and uniformity came largely from these regional associations and collaborations.

Social Embeddedness

Policy implementation was often dependent on a health officer or health department's relationships with non-state actors. We asked about the way health officers established relationships with social partners, community organizations, and private actors in the process of devising and implementing policy.

Stakeholders and Community Outreach: Formalizing Relationships

While informal communication with the community was expected of health officers both before and during the pandemic, our interviews revealed the ways health officers sought to formalize relationships with “stakeholders,” broadly construed, to implement policies. This was, as one health officer described it, an “upside of the pandemic: that we built very strong, very good lasting working relationships with a lot of different people.” Formalizing these relationships happened in a variety of ways. Some health officers stood up weekly meetings that included government and business leaders, religious leaders, educators and members of school boards. In areas dependent on tourism, health officers reached out to the lodging and restaurant industries, which were, themselves, often starting their own industry associations. Informal capacity was, therefore, endogenous, with public health officers able to take direct action to build on social ties and preexisting social capital to create more robust partnerships and enhanced policy development and implementation abilities.

Stakeholder involvement was part of both policy development and policy implementation. Sometimes this led health officers to change their health orders—compromising on restaurant mask mandates and instead settling on risk tiers, for example, to let consumers know the status of workers (vaccinated, tested, etc.). These fora “broke down the fear of going [into our] community” and created a forum for grievances. Not only did these stakeholder groups create buy-in for policies, but they also created accountability in both directions: if stakeholders agreed to policies and then violated them, health officers felt less compunction bringing in sheriffs or landlords to enforce policies.

When vaccinations were being rolled out, many county health officers put together task forces or think tanks to determine vaccine priorities and clinical strategies. These included leaders across business, labor, faith, education, and elder care communities. Together, they created dashboards and policies that were then disseminated through community-based organizations and schools. These stakeholder groups were particularly effective in engaging hard-to-reach communities, including non-English speakers or migrants. According to one health officer, “prior to COVID, we talked a lot in the health department about working with the community. But we really went a lot farther in actually broadening the wide scope of groups and community organizations that we actually worked with.” Again, this approach demonstrates how public health officers could enhance their informal capacity by building or extending partnerships with community-based organizations, sometimes where none existed prior to the pandemic. These relationships, for example, went beyond “outreach”: they allowed for the local government to enter communities with which it had no prior interactions. Health officers spoke of building relationships where none existed, particularly in affluent areas where most residents use only private health care, or among marginalized populations often overlooked by government. While such newly-developed partnerships and the associated informal capacity they produced may demonstrate that the existence of a community or community-based organization willing to partner with public health departments is necessary for informal capacity, the success of developing partnerships and capacity where none existed previously, however, shows that even when civil society organizations are weaker, informal capacity can be built.

Community stakeholders were particularly important in counties that do not run public clinics. Health officers occasionally hired consultants to help identify and bring together stakeholders; these were then formalized into health equity working groups. As the immediate threat from COVID dissipated, health officers tapped into these working groups for causes as varied as diabetes prevention, child vaccinations, and insurance sign-ups. Some health officers used CARES Act funding to assist with pandemic policies that more broadly addressed the social determinants of health, such as social assistance, evictions, or meals for the elderly. Health officers also brought in these stakeholders to broader conversations about economic recovery and reopening.

When possible, health departments hired community health workers to implement policies. They contracted out to nonprofit organizations to recruit volunteers for testing and vaccinations, and created communications hubs in senior centers or local churches. Community-based organizations working with labor, agricultural, and tourism groups helped to craft professional media messages and put together weekly meetings. Alternatively, health officers tapped into preexisting campaigns: one county's multiyear program to create a healthier community already joined together major sectors across the county, which the health department grafted onto to conduct community surveys and to vet policy ideas. Health officers described relying on promotores (community health workers) to reach communities of color, immigrants and refugees, and other operational areas of the Health and Human Services department.

Unsurprisingly, health officers also relied on their relationships with hospitals, which predated the pandemic but became critical to the iterative nature of policy development during the pandemic. Health officers emphasized the need for collaboration with hospitals and federally qualified health centers, and described the “constant” contact (it was not uncommon to hear that health officers and hospital leaders or medical teams were on speed-dial, or called each other in the middle of the night). Health department and hospital coordination helped health officers understand when they might need to declare local health emergencies or deploy emergency services. Further, they could work together on data-sharing, county-level COVID dashboards, and policy issues. A lack of coordination, on the other hand, created information gaps, which exacerbated pandemic problems like shortages of hospital beds, ventilators, drugs, and vaccines.

Conclusion: The Limitations of Public Health Capacity

“Relationships, trust, and education: those are the strongest tools that we have.” Protecting the public health is a vast charge for local government health agencies. The pandemic revealed just how dependent the state is on society itself—on partners, stakeholders, communities, and other actors in the health sector—to carry out its goals. The importance of relationships and partnerships to the full actualization of public health capacity were not limited to those between public officials and non-state actors. The instances in which public health officers felt they were best able to formulate and implement their desired policy response were characterized by strong, yet informal partnerships between public health and other actors within their county government, as well as between public health officers of different counties. Through our in-depth interviews and analysis of health officers in California, we show that state capacity cannot be determined through formal institutional measures alone. The number of professional staff, the amount of funding, and the statutory powers of health officers are not enough to explain how the state successfully, or unsuccessfully, carries out its public health objectives.

We have identified three aspects of informal capacity, including within-government cooperation, across-government cooperation, and social embeddedness, that we believe are critical to understanding state capacity. This analysis builds on the concept of “shadow capacity” articulated by Rich (2023) by showing how non-state actors achieve state capacity where bureaucrats are constrained. It also contributes to the literature on local and municipal politics, in that the highly decentralized nature of public health means capacity is determined by local governing institutions and actors. Future research needs closer attention to the ways national policies are implemented via local actors and institutions.

However, it is also worth noting that public health—like many areas in which governments try to achieve amorphous, ambitious goals that rely on individual behavior and compliance—will always face the challenge of occupying an area of law that is notably weak on enforcement. Health officers noted the democratic tensions inherent in their policymaking, whereby they can issue strict mandates without public deliberation or veto, but lack the ability to enforce them. Sheriffs often refused to enforce mask mandates, and a few months into the pandemic, partisan politics often determined whether or not local compliance was likely.

Given these circumstances, health officers relied heavily on the informal capacity they had built with communities. As one noted, “I don't have a badge, let alone a gun … I can issue a piece of paper, and then I'll disappear.” Health officers relied on the trust built through face-to-face relationships: “the partnerships we made with business…[and] community-based organizations and our schools. We took so many different avenues to get the word out.”

Health officers worked through proxies who commanded local trust, such as clinic physicians and fire chiefs. They stood up local task forces, conducted outreach through community health workers, and brought in voices from the community to assist with the policy process. These efforts sustained the difficult years of the pandemic, and have also expanded beyond the pandemic to facilitate health efforts beyond COVID.

Our examination of county-level policy responses to COVID-19 in California has highlighted the importance of what we have described as “informal” capacity, including partnerships with non-state actors and collaborative relationships both within and across county government. While we describe this type of capacity as informal, what has also become clear in the case of California is that this type of public health capacity can be built and formalized. Some of what is now informal is already showing signs of formalization and expansion among some counties. The effort to concretize the organization of regional collaboratives and the expanding policy focus beyond COVID-19 is a prime example of this phenomenon. Our research also demonstrates how a deeper investment in local public health is necessary for expanding and actualizing public health capacity not just through more personnel or technical infrastructure, but through policies that invest in community-based organizations, thereby, creating the community capacity that often synergized state capacity during COVID-19. Such investments in public health policy will also serve to better establish the within government partnerships required for a whole government public health response. Indeed, if experience gained through the development and implementation of past policy is critical to the development of such within government partnerships, then more policy is necessary to build future collaborative capacity. If the informal aspects of public health capacity can be made more formal, it is, therefore, possible to move away from the appearance of contingency and towards a public health capacity that is more consistently and evenly dispersed across units of government. We hope that future research can help to better understand and elaborate on the dynamic nature of public health capacity, including the relationships that exist both at the boundary of formal state institutions and between state and society. Doing so will help the state achieve its health goals.