Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Natural Hazard Science. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 27 February 2021

Policy Process Theory and Natural Hazardsfree

  • Thomas A. BirklandThomas A. BirklandDepartment of Public Administration, North Carolina State University


Natural disasters pose important problems for societies and governments. Governments are charged with making policies to protect public safety. Large disasters, then, can reveal problems in government policies designed to protect the public from the effects of such disasters. Large disasters can serve as focusing events, a term used to describe large, sudden, rare, and harmful events that gain a lot of attention from the public and from policy makers. Such disasters highlight problems and, as the public policy literature suggests, open windows of opportunity for policy change. However, as a review of United States disaster policy from 1950 through 2015 shows, change in disaster policy is often, but not always, driven by major disasters that act as focusing events. But the accumulation of experience from such disasters can lead to learning, which can be useful if later, even more damaging and attention-grabbing events arise.

Because of the threat of massive losses of life and property damage, natural disasters are important problems throughout the world. Natural disasters cause hundreds (in some cases, thousands) of deaths and billions of dollars in losses to property and economic productivity every year. While the death toll in developed countries as a result of natural disasters has generally declined, disasters in developing countries continue to exact a higher cost in human lives. Repeated exposure to natural hazards means that some nations or regions find it difficult to recover from and mitigate the effects of disasters before the next disaster strikes.

A major role of government, regardless of the form (federal or unitary, democratic or non-democratic, presidential or parliamentary), is the protection of its citizens and residents from harm. Natural disasters are one of the risks that governments are charged with addressing. Other risks include crime, economic privation, accidents, and disease. Natural disasters, because of their scope and suddenness, have a particular influence on setting agendas and inducing policy change.

Relatively small events, such as localized flooding, can test local and regional governments’ capacity to effectively prepare for and mitigate such events, and to respond and promote recovery. Large natural disasters, such as Hurricanes Katrina and Sandy, the South Asia Tsunami of 2004, the Haitian earthquake of 2010, or the Japan earthquake and tsunami of 2011, gain considerable attention from the media (and thus, the public) and from policy makers. Large-scale disasters tend to capture media and public attention around the world, particularly in places that are exposed to similar hazards. Indeed, the range of effects, from the localized effects of small-scale events, to the catastrophic effects of the best known storms, tsunamis, floods, and hurricanes, suggests that the term disaster may be inadequate to describe an event’s effects. In this essay, the term disaster is used to describe any event that leads to a disruption in the typical lives of people in a particular area, whether at the neighborhood, regional, or national level. But the term catastrophe can also be employed for the largest of these disasters, where disaster is measured by its scope—the amount of damage or the number of victims—its scale—that is, its geographic extent—and its effects (Quarantelli, 2005). By this standard, Hurricane Katrina was a catastrophe that crippled the city of New Orleans and staggered Louisiana and Mississippi; by contrast, Hurricane Sandy was not catastrophic in New York and New Jersey broadly—but in several towns and neighborhoods, it was catastrophic in that it entirely remade daily lives and routines.

Because such disasters focus attention on the effects of natural disasters, students of natural disaster have adopted the term focusing event to describe such events and the sudden attention paid to these problems. As a shorthand term to describe important events, the concept is useful. But many disaster scholars have adopted this term, which originates in the public policy literature, without considering its roots in policy theory (in particular, John Kingdon’s [2011] discussion of the idea in terms of the “streams” metaphor of the policy process), without a great deal of knowledge of what a focusing event is in policy making terms, and why not every event has the potential to alter our thinking about disaster policies. Conversely, while it is often said that disaster policy is more reactive than it is proactive, not all changes to policies related to natural disasters are made in direct response to such events. Rather, other things also happen in the political realm that influence change in disaster policies.

In this article, I explain how natural disasters influence the policy process. In particular, I will describe how these focusing events can have important and sometimes unexpected effects on policy making. I will turn to a discussion of the influence of events on disaster policy in the United States. The literature on natural disaster policy in the United States has consistently said that natural disaster policy is reactive to recent events, rather than proactive in anticipating potential future effects and mitigating their potential harms before they occur. This is also true in Europe, where events can also serve as drivers of policy change (Nohrstedt, 2009). There are good reasons why our policy making institutions tend to react to events and make policy based on recent events—that is, why policy makers learn from disaster experience. However, there are times when disaster policy is made without any apparent learning, which means that focusing events may not be influential.

Overview: Frameworks of the Policy Process

Public policy making has been of considerable interest to social scientists for years, because public policy is the expression of what governments view as problems, and what they plan to do—or not do—about public problems. The study of public policy is the understanding of how policy problems come to the attention of policy makers, how policy makers create and evaluate alternative solutions to problems, and how policies are implemented to have their desired effect.

For many years, students of the public policy process in democratic societies have been working to create models or frameworks or approaches to the study of policy making. The earliest efforts to model some sort of policy process rested on input-output models of the political system, most closely associated with John Easton (1965). Using various different terminology, scholars posted some sort of policy cycle that consisted of a series of stages, from problem identification and agenda setting, through alternative policy selection, adoption, implementation, and feedback to the start of the process (deLeon, 1999; Nakamura, 1987; Sabatier, 1991).

Since at least the late 1980s, these scholars have known that this is a far too simple model of the policy process. The stages model implies an orderly series of steps toward enactment of legislation, akin to the sort of “how a bill becomes a law” discussions common in civics textbooks. The problem is, of course, that not all policies follow this step-by-step process. Some policies skip steps in the process, and sometimes the process halts after agenda setting, and no change results. And feedback happens from and to all these stages of the policy process. Thus, while what Paul Sabatier (1991) calls the “stages heuristic” is a handy way of categorizing which aspects of policy making a scholar may be studying, it is not a proper model of the policy process. Still, as deLeon (1999) notes, categorizing elements of the policy process has been a useful way for scholars to focus on particular aspects of policy making.

One such important element of the policy process is agenda setting. An agenda is a collection of problems; understandings of causes, symbols, solutions, and other elements of public problems that come to the attention of members of the public and their governmental officials. Agendas exist at all levels of government. Every community and every government entity has a collection of issues that are available for discussion. All these issues can be categorized based on the extent to which an institution is prepared to make a decision to enact and implement or to reject particular policies, or, at least, to allocate more resources or administrative attention to a problem (Gainsborough, 2009). We can order issues from those that are furthest away from actual governmental action, such as any idea about a problem and its solution. These ideas are contained in the “systemic agenda,” which contains any idea that could possibly be considered by participants in the policy process in a particular system.

Agenda setting is the process by which groups compete to move issues from the systemic agenda to the institutional agenda. The systemic agenda “consists of all issues that are commonly perceived by members of the political community as meriting public attention and as involving matters within the legitimate jurisdiction of existing governmental authority” (Cobb & Elder, 1983, p. 86). If a problem or idea is successfully elevated from the systemic agenda, it moves to the institutional agenda, which is “that list of items explicitly up for the active and serious consideration of authoritative decision makers.” If an issue moves “up” on the institutional agenda, to the point where, say, a committee must vote on a bill, the issue reaches the decision agenda, at which point a decision making body must do something, either by enacting some sort of policy change, or choosing not to do so.

Group competition to set the agenda is fierce because no society, political system, official actor, unofficial actor, or individual person has the capacity to address all possible alternatives to all possible problems that arise at any one time (Baumgartner & Jones, 2009; Cobb & Elder, 1983; Walker, 1977). Even when an issue gains attention, groups must fight to ensure that their depiction of the issue remains in the forefront and that their preferred approaches to the problem are those that are most actively considered. At the same time, groups fight to keep issues off the agenda. This competition is why agenda setting is interesting to political scientists: “the definition of the alternatives is the supreme instrument of power” (Schattschneider, 1975, p. 66), where alternatives can mean issues, events, problems, and solutions.

Group competition in agenda setting is both about raising an issue on the agenda and about the competition over problem definitions and solutions. This story telling is important because stories of problem definition compete with other problem definitions, and strongly signal pre-existing preferences for particular policies (Birkland & Lawrence, 2009; Hilgartner & Bosk, 1988). The agenda setting process is, therefore, a system of sifting issues, problems, ideas, and implicit solutions, because so often in policy debate, the definition of a problem presupposes the solution to the problem. For example, after the 1999 Columbine High School shootings, the issue of school violence quickly rose to national prominence, to a much greater extent than it had after other incidents of school violence. Competition for attention then was about depictions of school violence as a result of, among other things, lax parenting, easy access to guns, or the influence of popular culture (TV, movies, video games) on high school students (Birkland & Lawrence, 2009; Lawrence & Birkland, 2004). This competition over why the Columbine shootings happened and what could be done to prevent future school shootings was quite fierce, more so than the competition between school violence and the other issues vying for attention at the time.

Interest groups, or, more often, groups of groups called advocacy coalitions (Jenkins-Smith, Nohrstedt, Weible, & Sabatier, 2014) seek to gain attention in different venues (national, state, or local governments, the legislative, executive, or judicial branches, and so on), to elevate these issues on the agenda (see also Pralle, 2006). They also seek to keep the issues of other groups off the agenda; indeed, agenda denial may be a more important strategy in policy debate than promoting ideas on the agenda (Cobb & Ross, 1997).

Kingdon’s multiple streams approach is an attempt to be a model of the policy process in political institutions. In Kingdon’s model, inspired by Cohen, March, and Olsen’s (1972) work on organizational decision making, three conceptual streams describe agenda-setting and alternative selection. These are the problem stream, which contains ideas about various problems in society to which public policy might be applied; the politics stream, containing the ebb and flow of electoral politics, public opinion, and the like; and the policy stream, which contains a set of ideas about how problems could be addressed. In all three streams, problems and solutions, and the means by which to implement solutions to solve problems, are not self-evident, but are subject to considerable debate and conflict. As noted, competition is fierce over which problem definition and policy alternative will dominate the debate and lead to change.

A window of opportunity opens when the streams come together at a moment in time where problems are matched with solutions, usually through the efforts of policy entrepreneurs who remain consistent actors in the policy process. Policy entrepreneurs include elected officials, government officials, interest group leaders, and the like who harness knowledge of substantive policy and knowledge of the policy-making process to promote ideas for policy change. Policy entrepreneurs may be aided by changes in politics that make a window of opportunity arise. Many of these ideas have existed for a long time, but a change in the streams may push the issue up the agenda and will open a window of opportunity for change.

Kingdon argues that agenda setting is driven by two broad phenomena: changes in indicators of underlying problems (such as data about unemployment, disease, teen pregnancy, or crime); and focusing events, or sudden shocks to policy systems that rapidly increase attention to a suddenly revealed problem. In the streams approach, a focusing event opens windows of opportunity because events provide an urgent, symbol-rich example of claimed policy failure (Birkland, 1997, 2006; May, 1992). They therefore change the nature of the politics surrounding an issue. However, a focusing event by itself is often an insufficient driver of agenda. Change comes, according to Kingdon, with “a powerful symbol that catches on, or the personal experience of a policy maker” (2011, pp. 94–95). For example, while Katrina was a very large hurricane, it took on even greater symbolic importance when, for example, images of an elderly lady using an American flag as a blanket, or people stranded on rooftops, were widely propagated in the news as both symbols of problems in their own right, and as symbols of policy failure, because, normatively, people should not need to seek refuge on rooftops or use a flag as a blanket.

Kingdon’s conception of disasters as focusing events involves events that “simply bowl over everything standing in the way of prominence on the agenda” (Kingdon, 2011, p. 96) is a useful starting point, because such events usually involve disasters that concentrate their obvious damage and harms in one place, thereby providing opportunities for attention. Kingdon’s discussion of the idea of a focusing event is very influential to those of us who study disasters and crises, but few scholars have tested the idea across a long period of time in a policy domain that addresses natural hazards broadly, or specific natural hazard domains, such as earthquakes and hurricanes. Most disaster policy research adopts this “bowling over” notion, and assumes that the event so dominates the agenda that talk of policy change—that is, learning the “lessons” of the recent event—is the normatively desirable result, because the event itself often reveals policy failures from which we normatively expect to learn (Birkland, 2006; drawing heavily on May, 1992).

Before concluding this discussion, it is important to understand that Kingdon’s streams approach is not the only major approach to the study of public policy. The discussion that follows also draws on fundamentals of the Advocacy Coalition Framework (ACF), associated most closely with Paul Sabatier and other scholars who have advanced the ACF as a way of explaining group competition, learning, and policy change (Jenkins-Smith et al., 2014). Another very influential framework is the Punctuated Equilibrium Theory (PET), in which Frank Baumgartner and Bryan Jones have described how policy making undergoes long periods of stasis, punctuated by sudden policy changes. Both the ACF and PET take sudden, attention-grabbing events into account as drivers of policy change. In the ACF, events can be exogenous shocks that create new coalitions to press for policy change; in PET, such events can change the policy image of existing policy from positive to negative, causing pressure to move away from status-quo policies. This is, of course, a highly simplified depiction of these approaches to policy studies. Their influence on how we think about natural disasters in the policy process will be illustrated in the balance of this article.

Focusing Events in the Policy Process

Most attention to natural disasters in policy studies can be located in the literature on agenda setting. This is sensible, as natural disasters are powerful agenda setters. John Kingdon understood this in terms of focusing events. Cobb and Elder called events such as the Santa Barbara oil spill of 1969 a particular sort of issue initiator they called a circumstantial reactor that changed the political climate (Cobb & Elder, 1983, p. 83). Throughout their application of the idea of punctuated equilibrium to the policy process, Baumgartner and Jones (2009) note that events are important in altering the agenda, creating new or newly important information, and creating opportunities for change.

This change opportunity is important, because, in Baumgartner and Jones’ punctuated equilibrium theory (PET), long periods of policy stasis are punctuated by sudden, often-unexpected shocks to the system. These shocks are usually only predictable in the sense that a major event, such as a natural disaster, will happen at some uncertain time in the future. Some hazards, such as floods or hurricanes, may be more likely at certain times of the year, such as during monsoon or hurricane seasons. But others, such as earthquakes, are wholly unpredictable. And while emergency managers may be prepared for typical natural disasters, particularly large ones can create substantial shocks to policy making, creating a greater likelihood of policy change. This is because, as Baumgartner and Jones note, sudden events are often negative events from the perspective of the most powerful actors in a policy domain, because these powerful actors are often most interested in maintaining the policy status quo.

With this negative attention, then, come calls for policy change, often made on behalf of the less powerful members of a community; this is consistent with E. E. Schattschneider’s notions of the “scope and bias of the pressure system,” in which those who favor the status quo seek to avoid broader public debate and conflict (Schattschneider, 1975, p. 20). But Baumgartner and Jones also note that “simply because a disadvantaged interest proposes a new interpretation of events and attempts to attract the attention of allies in other areas of the political system does not mean that those challengers will be successful” (2009, p. 8). After all, while the Exxon Valdez (1989) and Deepwater Horizon (2010) oil spills were the biggest of their type in U.S. history, the legislative and regulatory reaction to these events was neither as punitive to the oil industry as the industry had feared—or the environmental movements had hoped.

Because we cannot know in advance whether an event will be a focusing event, Birkland defined a potential focusing event as an event that is:

sudden, relatively rare, can be reasonably defined as harmful or revealing the possibility of potentially greater future harms, inflicts harms or suggests potential harms that are or could be concentrated on a definable geographical area or community of interest, and that is known to policy makers and the public virtually simultaneously.

(Birkland, 1997, p. 22)

A focusing event can serve as prima facie evidence of policy failure (May, 1992) that needs to be addressed to prevent a recurrence of the negative outcomes of the event. The public and policy makers need to understand the event and its causes. This process involves telling what Deborah Stone calls causal stories about why the event was as bad as it was (Stone, 1989; see also Stone, 2012). A key feature of the politics of such events turns on the arguments the advocates make in a policy domain, about whether an event was unintentionally or purposefully caused. Where events are the result of neglect or carelessness, we find greater debate over the nature and solutions for the event and the problem it revealed than we have traditionally seen with “acts of God.”

The reason events are important is that they attract attention to problems, which, in turn, increases the likelihood that organized interests, including some that are influential and powerful, could enter policy debates and advocate policy change (Baumgartner & Jones, 2009; Schattschneider, 1975). Most of the attention triggered by a focusing event is negative attention, which further motivates political debate, and potentially moves issues closer the decision agenda.

The foregoing discussion should not be taken to mean that policy change is a typical or inevitable result of a focusing event. The nature of event-driven policy change is very complex and is comprised of many actors in many different types of policy subsystems, depending on the event itself and the policy actors an event stimulates (Nohrstedt & Weible, 2010). Even if policy change does result, the change may not be the “best” policy change, as judged by experts in the field. For example, continued legislation to broaden availability of disaster relief funding may be helpful to the survivors of an accident in the near term, but such funding, or the accompanying sort of responses, such as larger flood walls or levees, may not substantially reduce a community’s vulnerability to disaster and may increase the likelihood of a future, more damaging event.

Focusing Events and Disaster Legislation in the United States, 1950 to 2015

In simplest terms, I argue that big disasters become focusing events, and that such large disasters generate policy change and, perhaps, learning. If this is the case, we should be able to link particular events to particular significant legislative enactments at the federal level. To assess this idea, I list major federal disaster policy enactments in Table 1.

Table 1. Significant disaster legislation in the United States, 1950–2014 (adapted from May & Williams, 1986)



Related Event



Disaster Relief Act of 1950, PL 81-875

Formalized existing practice allowing for funding to repair local public facilities.


Federal Flood Insurance Act, PL 84-1016

Flood insurance program that never started because the House rejected funding for it.


Flood Control Act of 1965, PL 89-298 Title II

Hurricane Betsy (1965)

More fully engaged the Army Corps of Engineers to work with the Orleans Parish Levee Board to build levees and related works. The Corps assumed responsibility for building this flood protection system and for a system in south Louisiana. The design storm for this system was not as strong as was recommended by NOAA. Local levee boards retained maintenance responsibility. The act authorized other flood control projects nationally.


Disaster Relief Act of 1966, PL 89-769

Alaska earthquake of 1964; Midwest tornado outbreak of Palm Sunday 1965, Hurricane Betsy of 1965

Amended 1950 act to allow rural communities to participate; aid for damaged higher education facilities; repair of public facilities under construction. Created office of emergency planning (OEP). Broadened eligibility for aid to rural communities, public universities, and colleges, SBA loans to private universities and colleges, and empowers feds to participate more in wildland firefighting. Expansive program that reflected current New Deal priorities (Morris, 2014).


National Flood Insurance Act of 1968, PL 90-448.

Hurricane Betsy, and reaction to continued ad hoc relief policy after the 1964 Alaska earthquake (Knowles & Kunreuther, 2014)

First flood insurance program, enacted as Title VIII of the Housing and Development Act.


Disaster Relief Act of 1969, PL 91-79

Hurricane Camille (1969)

Debris removal, food aid, unemployment benefits, loan programs revised; duration limited to fifteen months.


Disaster Assistance Act of 1970, PL 91-606

Hurricane Camille (1969)

Continues most provisions of the 1969 law, plus grants for temporary housing or relocation, funding for legal services.


Flood Disaster Protection Act of 1973, PL 93-234.

Hurricane Agnes (1972)

Expanded coverage; imposed sanctions on communities in flood zones that failed to participate in flood insurance. Made purchase of insurance mandatory in “special flood hazard areas.”


Disaster Relief Amendments of 1974, PL 93-288

Hurricane Agnes (1972)

Defined “major disasters” and “emergencies,” broadened categories of allowable expenditures. Served as template for most policy until the Stafford Act. In 1977, this act was reauthorized through 1980 (PL 95-51). It was reauthorized in 1980 (PL 96-568). Solidified disaster relief as something approaching an entitlement program (Morris, 2014).


National Earthquake Hazards Reduction Act, PL 95-124

1964 Alaska and 1971 Sylmar (California) earthquakes.

Bill enacted to address concerns raised by Alaska and San Fernando earthquakes, among other events. Included provisions to support research on prediction and mitigation.


Robert T. Stafford Disaster Relief and Emergency Assistance Act, PL 100-707

No precipitating event. This enactment was a reaction to Reagan administration attempts to change cost sharing formulas with states.

Amends Disaster Relief Amendments of 1974. Clarifies presidential disaster declarations, increased emphasis on mitigation. Removed federal emphasis on disaster recovery.


Stafford Act Amendments, 103-181

Midwest floods of 1993

Amended the Stafford Act to emphasize mitigation; enacted in the aftermath of the 1993 Midwest floods.


Disaster Mitigation Act, PL 106-390

Hurricane Fran, other floods

Encourages state and local hazard mitigation, requires enhanced state and local mitigation planning.


Homeland Security Act, PL 107-296

September 11 terrorist attacks

Made FEMA a part of the new Department of Homeland Security as, officially, the Emergency Preparedness and Response Directorate. Moved FEMA further from a direct line to the president. Terrorism became the most important hazard in the eyes of federal leaders at the time.


Flood Insurance Reform Act of 2004, PL 108-264

Indicator driven—realization that flood insurance was not risk based or actuarially sound

Provisions to encourage owners of repetitively flooded properties to accept buyouts or lose eligibility for flood insurance.


National Windstorm Impact Reduction Act of 2004, Title II of PL 108-360

Result of efforts of the wind hazard and storm surge research community to address these hazards in a way similar to earthquakes under the NEHRP.

Creates a National Windstorm Impact Reduction Program patterned after the Earthquake Program. Part of the Earthquake program reauthorization.


Post-Katrina Emergency Management Reform Act of 2006

Hurricane Katrina

Reestablished FEMA as a formal entity in DHS; designated FEMA Administrator as lead for national disaster response; reestablished direct report of FEMA administrator. To the president in disasters.


Biggert–Waters Flood Insurance Reform Act of 2012

Indicator-driven: huge deficits in the flood insurance program. These deficits, however, were generated by Hurricane Katrina: “The NFIP paid out more claims in 2005, however, than it had paid out over the entire life of the program to that point” (citing Hayes & Neal, 2011; Kousky & Kunreuther, 2013, p. 3)

Required flood insurance to be priced using actuarial data, thereby removing subsidies.


Disaster Relief Appropriations Act of 2013 (PL 113-2)

Hurricane Sandy

Funds several federal agencies for recovery operations after Hurricane Sandy. Funding for direct use by federal agencies, and for the provision of aid through federal programs to states and local governments.


Homeowner Flood Insurance Affordability Act of 2014

Indicator driven: large increases in insurance rates led to political controversy

Delayed implementation of actuarial rates for four years.

Events are very important in the natural disaster domain in the United States. This history illustrates what Morris (2014) termed “a call-and- response pattern of major disasters producing increasing levels of assistance from Congress after World War II” (p. 407). Indeed, this increasing tendency to create broader groups of interests eligible for disaster relief, and the increasing generosity of such relief, reflects them as a “ratcheting up” of federal policy, yielding broader and deeper federal involvement in what has traditionally been a local responsibility (May & Williams, 1986). Indeed, all federal disaster policy is marked, first, by the reiteration that disaster preparedness, response, recovery, and mitigation is primarily a local and state function, with the federal government playing a support role; and, second, by the continued growth in federal involvement in planning for disasters, supporting response, and encouraging better response and recovery.

This table shows several important enactments and related events. However, this history starts with the Disaster Relief Act of 1950, which was less the result of any single event than of the accumulation of experience with ad-hoc federal responses to disasters that date to the earliest days of the republic. The 1950 act was an attempt to create a system for the orderly provision of aid to states upon a request from the governor of a stricken state. The law established the doctrine that prevails in the United States: “to supplement the efforts and available resources of States, local governments, and disaster relief organizations in alleviating the damage, loss, hardship, or suffering caused” by a disaster (now codified at 42 U.S.C. 5122). Primary responsibility is to rest with the state and local governments. This doctrine is reflected in the nature of federal disaster relief organization—no federal emergency management organization has ever been very large, and FEMA has less than 3,000 people, most of whom work with state and local governments.

The 1950 act was also an attempt to routinize the process for seeking and granting federal disaster assistance to state and local government. Congress has, since this act, sought to create a system of predictability that will allow what Lindsay and Murray (2011) call “normal disasters” that do not unduly tax federal relief funds. But while the doctrine and the sense of a disaster declaration process remain the key features of federal law, many disasters triggered changes to the law to broaden the scope of federal assistance. Enactments in 1966, 1969, 1970, and 1974 were linked to experiences with particular disasters.

The 1974 Disaster Relief Act was a second landmark body of legislation. This established a basic framework that continues to this day. This enactment rested on the doctrine that the federal government supports state and local efforts, but it also greatly broadened the range of aid that could be offered to state and local governments and, for the first time, provided a system of direct aid to individuals harmed in the disaster.

The next major event, although not legislative in nature, was the establishment, in 1979, of the Federal Emergency Management Agency (FEMA) though Reorganization Plan No. 3 of 1978. The period between 1979 and the late 1980s was characterized by very few major disasters. During this period, FEMA was charged with a much broader range of civil defense and national security tasks, and its director for most of the Reagan Administration, Louis Giuffrida, was an enthusiastic proponent of FEMA as a national security agency. This raised considerable ire among other executive branch officials and Congress, and FEMA’s authority was constrained by the enactment of the Robert T. Stafford Disaster Relief and Emergency Assistance Act of 1988. With the Stafford Act, as implemented through a series of executive orders, FEMA finally had something like an enabling statute. The Stafford Act’s genesis, however, is not found in any one disaster event. Sylves notes that in “April 1986, FEMA proposed changing the process of declaring disasters, the criteria for eligibility for federal assistance, and the nonfederal responsibility for major disasters. The proposed regulations would have decreased the federal share of disaster costs to 50 percent from 75% … Due to strong opposition in Congress, FEMA subsequently withdrew the proposed rules” (Sylves, 2015, p. 70). The Stafford Act, therefore, placed FEMA at the center of federal natural disaster policy, rebuffed the Reagan administration’s efforts to reduce federal cost sharing, and introduced a significant new hazard mitigation program while, curiously, removing disaster recovery as an important federal function. Direct experience with flooding in the Midwest in 1992 led to the enactment of the Stafford Act Amendments of 1993, which created a much more robust hazard mitigation program, while Hurricane Floyd in 1999, and experience with the implementation of the 1993 act led to the enactment of the Hazard Mitigation Act of 2000. During the same period, FEMA established a mitigation directorate in 1994, and created a public-private partnership program called Project Impact that was designed to increase community involvement in hazard mitigation. This program—and a great deal of funding and effort for mitigation—was slashed upon the inauguration of the Bush Administration in 2001.

Of course, 2001 was also when terrorists attacked the World Trade Center and the Pentagon. These spectacular events yielded profound changes to national security and emergency management policy, and the emergency management structure in the United States returned, in many ways, to a civil defense posture. While federal officials continued, after September 11, to insist that preparedness for terrorist attacks was part of an all hazards approach, there were troubling signs that federal emergency management would be less effective. While FEMA retained its name, it was buried deep in the Department of Homeland Security (DHS) bureaucracy, and its administrator’s access to the president was constrained. While the agency performed reasonably well during four significant hurricanes that struck Florida in 2004, this was likely more the result of solid preparedness in Florida, a state prone to hurricanes, than federal action. In any event, the stage was set for Hurricane Katrina.

Hurricane Katrina is striking because it revealed how much FEMA’s capacity and commitment had atrophied since it was buried in the DHS. While FEMA in any form would have been severely challenged by the catastrophic storm that Katrina became, its response in Louisiana and Mississippi was broadly considered inept, and the criticisms leveled at FEMA sounded very similar to those leveled against it during its equally inept response to Hurricane Andrew in Florida in 1992.

Katrina was clearly a focusing event, and it led directly to the enactment of the Post-Katrina Emergency Reform Act of 2006 (PKEMRA), which restored FEMA’s administrative autonomy to some extent. Major preparedness functions housed in other agencies were returned to or moved to FEMA, and the FEMA administrator’s title was changed to “director,” with a reporting line directly to the Secretary of DHS. And the storm and the poor response led to a shift in federal planning away from the National Response Plan to a new National Response Framework.

This is clearly a brief summary. What this history reveals is that disaster policy is, more than most policy domains, clearly driven by disasters. But not all such policy is disaster driven, meaning that disaster policy shares similar characteristics with most other public policies—change driven by proponents of policy who can, at times, use events to gain greater attention, but who, at other times, must rely on other arguments or accumulated experience from past disasters.

Variation in the Focal Power of Disasters

Most of the literature on disasters as focusing events does not adopt the potential focusing event definition, and merely claims that key historic events were focusing events, such as in the Disaster Time Line (Rubin, Renda-Tanali, & Cumming, 2009). For example, Farley et al. (2007) state that “Catastrophes have a long history of serving as focusing events that open policy windows, often resulting in profound societal change,” and cite the Johnstown Flood of 1889, the Mississippi River floods of 1927, and the September 11 attacks. They do not cite any events that had little or no focal event; in short, they, like most scholars who adopt the focusing event concept, do something social scientists call “selecting on the dependent variable.” This means that they and other scholars often study major events and extrapolate from that idea that all disasters, by and large, are focusing events. This is problematic; some events gain more attention than others, some events have different policy effects than others (sometimes, little or no effect, or the “wrong” effect from the perspective of the disaster management community). Literature grounded in the policy process literature, such as Gerber (2007) adopts my definition of potential focusing events and adopts key elements of the “event-related policy change” model of the disaster policy process. Such models can accommodate the idea that some policies change as a result of something other than major focusing events, and that, conversely, some policy change is not attributable to a single event or to events themselves, but that they are the result of other factors.

How are disasters potential focusing events? Recalling my definition of potential focusing events, disasters gain massive public and elite attention—at least for a short while—because of the obvious signs of the harm done by the event. News media, in particular, produce many stories about disasters, including stories of survivors and what they will do next, of the heroic or futile efforts of “first responders” to help those in need, the success or failure of government preparedness and response, and, usually contrary to what actually happens, the news media carry stories of actual or feared looting or civil disorder (Tierney, Bevc, & Kuligowski, 2006), particularly when covering disaster stories in less developed nations. Indeed, one of the challenges for disaster professionals is confronting the media’s propagation of myths about disasters (Clarke, 2002).

But in dealing with the political environment as it is, not as we would want it to be, we need to understand that all this news coverage, and the related public discourse from politicians and other leaders, regardless of its accuracy, becomes part of the “primordial soup” of ideas that these events add to and stir up. This is to say that there is not an automatic or smooth translation between a disaster, the stories of fault and blame told after the event (typically about response, not about causes or preparedness), and the policy changes that result.

In understanding how focusing events influence policy, we have to understand that disaster events do not merely “impact” politics; rather, disasters are fully embedded in politics. The responses to such events will use policies that have already been developed, in political institutions, and the events will reveal, for the most part, the shortcomings in the management of responses to these events.

What makes disasters rather different is the nature of the political and policy-oriented discourse around disasters. In most policy domains, when something bad happens, we would expect that pro-change policy entrepreneurs would seek to use the event to promote change against the wishes of more powerful interests. We often conceive of mobilization of pressure for policy change to work against the most powerful economic and policy elites, which tend to dominate policy making. This is consistent with what we call a “conflict model” of policy making and discourse around events, such as oil spills (see also Gunter, 2005; Molotch & Lester, 1975). I reviewed politics of this sort in my discussion of the politics of oil spills and nuclear power plant accidents (Birkland, 1997, chs 4 and 5).

However, there are very few groups that mobilize around the problem of natural hazard losses after a disaster. My research into the nature of policy discourse in congressional committees charged with managing disasters found that, after earthquakes and hurricanes there is considerable effort on the part of local officials to press the federal government for more generous and more rapidly delivered disaster relief. But between events, while there is very little discussion of topics other than the delivery of disaster relief after hurricanes, in the earthquake domain, there is considerable inter-event discussion of the scientific and technical aspects of earthquake hazards and their mitigation (Birkland, 1998). In this way, natural disaster policy can be thought of as “policies without publics” (May, 1990), where such policies yield internal mobilization (Cobb & Elder, 1983) efforts among experts to promote policy changes that the public generally do not press for, because such policy interventions are often too technical, and because the sorts of policies being proposed are public goods around which interest group mobilization is unlikely to promote the change that policy elites or experts believe will actually improve the performance of, in this case, disaster policy.

These “policies without publics” make it less likely for there to be many pro-status quo and pro-change groups, aligned in what Sabatier and his followers have called “advocacy coalitions” (Jenkins-Smith et al., 2014; Sabatier, 1988; Sabatier & Jenkins-Smith, 1999) in the disaster policy domain. Typically, most policy domains will contain two to four competing advocacy coalitions that support or oppose policy change. My research has found that there are few such advocacy coalitions—in earthquake policy, a coalition of scientists and other professional advocates for better earthquake mitigation policies—while no such coalition appears to exist in the hurricane domain, and, when a focusing event mobilizes broader participation, both earthquake and hurricane policy debate focuses on disaster relief. But in the absence of a recent event, participants in making improved earthquake mitigation policy are much more active than are advocates for inter-event improvements to hurricane mitigation, to the extent they are organized and exist at all.

The professions that deal most closely with disasters—engineering, meteorology, emergency management, and the like—may mobilize because they believe their ideas are technically, professionally, and ethically correct. I found that scientific and technical expertise has dominated the discussion of earthquakes to a far greater extent than occurs in discussions about hurricanes, even as the latter hazard may well need more attention from scientists and technical experts. However, the “easier” solutions to earthquakes may be more likely found among ideas developed and promoted by scientists and engineers, such as through changes in building practices. The “obvious” solutions to mitigating hurricane hazards—strict coastal development regulations, or active retreat from the riskiest areas—are usually the most controversial, and are more subject to debates based on political values and public desires than on weighing technical alternatives.

Consider Charles Perrow’s (2008) pessimistic finding about the power of Hurricane Katrina to yield meaningful change:

Hurricane Katrina in 2005 has not proved to be a wake-up call for addressing hurricane vulnerabilities. New building goes on in all the southeastern coastal areas, and despite the recommendations of experts, flooded areas of New Orleans are being resettled … Rather than wait for another Katrina to depopulate vulnerable area, we should forbid further building in vulnerable areas and relocate the most vulnerable populations.

(Perrow, 2008, p. 734)

One might argue that Perrow’s call to abandon the most physically vulnerable areas is the most obvious solution. Yet, it was also extremely controversial, because it failed to take into account the long-term assurances many in New Orleans had been given that their property was safe from storms, and it failed to take into account the communities that had developed, over time, in areas that, it turns out, were particularly vulnerable to disasters, but where appropriate risk information was neither used nor communicated to people most vulnerable to such a storm (Cooper & Block, 2006). Such a storm had been anticipated and tested under the Hurricane Pam scenario, a simulation exercise that sought to identify areas of weakness in preparedness and plans for response, but little effective action followed this exercise (Mycoff, 2007, p. 16; Scavo, Kearney, & Kilroy, 2007, p. 98), so whatever lessons could have been derived from that were not learned before Katrina.

Even if we had better scientific information about the impending Hurricane Katrina, the question of what do to ensure that city’s safety from wind, rain, and storm surge transcends simple scientific solutions, and enters the realm of trans-scientific problems that involve conflicts of preferences or values that are not easily understood in terms of science (Majone, 1989, p. 3). In this light, the politics of focusing events generally, and disasters in particular, are remarkably complex.

Disasters, Focusing Events, and Policy Learning

Disasters, like all focusing events, do not happen within a social, political, or economic vacuum. Looking at one individual “event” does not tell us much about what learning processes might have occurred after a disaster. Essentially, to comprehend policy change, research cannot analyze the attributes of a single event at the event level, but instead it is necessary to look at the domain level and consider multiple events. It is necessary to study focusing events as drivers of agenda change and policy change over several decades, and that learning is an iterative and accumulative process.

The public and, to some extent, theorists agree that events should yield policy change, and when they do not, the failure to “learn the lessons” of an event is prima facie evidence of dysfunctional governance. And, indeed, in many events, one instance does not yield positive change. But if we stopped our analysis here, rather than taking the long-term view, we would see a particular event as “failing” to yield change and learning. The longer view allows us to consider how domains allocate both immediate attention and change or the accumulation of experience and learning. In simplest terms, the first event of one kind—say, a mass casualty terrorist incident—may be largely met with bafflement, but government, the public, and the news media may react to subsequent events using ideas and frames developed in earlier events. In this way, events can yield the accumulation of knowledge about the substance of policies and about effective arguments about policy change. Peter May (1992) calls the former type of learning policy learning, and the latter type political learning. Within policy learning, May argues that we can see evidence of “instrumental policy learning,” which involves learning about the effectiveness and nature of the policy instruments intended to yield desired outcomes, or “social policy learning,” which involves learning about the more fundamental causes of such learning. Birkland (2006, 2009) argues that these two types of learning are akin to what Argyris (1976) calls “single loop” and “double loop” learning.

I have argued that events can have an immediate influence on policy, or they may not have a detectable influence on policy, but may contribute to experience about the problems revealed by events. This process is shown in Figure 1.

Figure 1. A model of event-related policy learning (Birkland, 2006, Figure 1.2).

We may also see evidence of May’s notion of superstitious learning, in which policy makers adopt policy changes in response to events or other stimuli based on an unsubstantiated belief that the new policies somehow respond to the problems raised by the recent event (Birkland, 2006). The September 11 attacks are an example of this sort of superstitious learning, in which all manner of policy changes were advanced in support of the “war on terrorism,” thereby creating a real but incoherent policy regime around the diffuse notion of homeland security (May, Jochim, & Sapotichne, 2011).

As Figure 1 suggests, events that do rise on the agenda can also generate experience about events, which can be tapped later when more fundamental policy change is desired. Disasters appear to behave like the sort of focusing events that can trigger learning. A prima facie case can be made that there is strong evidence of instrumental learning after disasters, either in the near-term or later, as experience accumulates. For example, experience with the law’s unclear standards for the sorts of events that could be considered major disasters led to clarification in the 1988 Stafford Act. The 1993 Midwest floods powerfully highlighted the shortcomings of existing flood policy and the need to devote more effort to hazard mitigation, and a series of policy instruments—such as buyouts and planning mandates—were developed to promote this goal. Such a goal was also institutionalized in FEMA in the mitigation directorate.

On the other hand, we have considerable evidence of superstitious learning and political learning in this domain. Proponents for broader and more generous aid were able to use events as a reason for significant policy change, by appealing to policy makers’—and the public’s—sympathy for disaster victims. This political learning made advocates more effective in expanding the scope of federal disaster policy. After September 11, the putative lessons of that event included the need, never clearly articulated, to create a Department of Homeland Security (itself a questionable move) and to bury FEMA under several layers of bureaucracy. Diminishing FEMA’s visibility and role, and the decision to, once again, have it led by inexperienced political appointees, suggested considerable unlearning of the lessons that led to the appointment of professional leadership at FEMA in the early 1990s.

This illustrates the extent to which spillovers from one policy domain can influence another domain, particularly when the ascendant domain gains so much attention. FEMA was caught in the post-September 11 spillover, as were a wide range of other federal, state, and local functions, ranging from local law enforcement, to aviation security, to food safety and security. In a way, emergency management moved from being a relatively well-bounded policy regime to being a part of a less coherent policy regime that ran into trouble when the “old style” of disaster—Hurricane Katrina—revealed the new regime’s shortcomings.


Natural disasters often work as focusing events that gain a lot of attention and drive policy change. Not every disaster drives policy change, and not every policy change is driven by a recent and obvious disaster. But there is little doubt that disasters matter in the making of disaster policy in the United States. They gain considerable attention, mobilize professionals and some lay people to press for improved policy, and the experiences gained from these events inform policy makers as they make new policy. Perhaps most important, major disasters are signals that if we do not prepare for, respond to, recover from, and mitigate the harms of future disasters, we are doomed to repeat the damage and suffering that these disasters—and, more properly, our decisions about hazards—can do to people and communities. Focusing events can, when carefully considered and understood, teach important lessons that, if put into practice, can reduce future suffering.

Suggested Readings

  • Birkland, T. A. (1997) After disaster. Washington, DC: Georgetown University Press.
  • Birkland, T. A. (2006) Lessons of disaster. Washington, DC: Georgetown University Press.
  • Birkland, T. A. (2015) An introduction to the policy process (4th ed.). New York: Routledge.
  • Godschalk, D., Beatley, T., Berke, P., & Brower, D. (2000). Natural hazard mitigation: Recasting disaster policy and planning. Washington, DC: Island Press.
  • May, P. J. (1992). Policy learning and failure. Journal of Public Policy, 12(4), 331–354.
  • Platt, R. (1999). Disasters and democracy. Washington, DC: Island Press.
  • Sylves, R. T. (2015). Disaster policy and politics: Emergency management and Homeland Security (2d ed.). Washington, DC: CQ Press.
  • Waugh W. L., Jr. (2007). Local emergency management in the post-9/11 world. In: W. L. Waugh, Jr. & K. J. Tierney (Eds.), Emergency Management: Principles and Practice for Local Government (pp. 3–21). Washington, DC: ICMA Press.


  • Argyris, C. (1976). Single-loop and double-loop models in research on decision making. Administrative Science Quarterly, 21(3), 363–375.
  • Baumgartner, F. R., & Jones, B. D. (2009). Agendas and instability in American politics (2d ed). Chicago: University of Chicago Press.
  • Birkland, T. A. (1997). After disaster: Agenda setting, public policy, and focusing events. Washington, DC: Georgetown University Press.
  • Birkland, T. A. (1998). Focusing events, mobilization, and agenda setting. Journal of Public Policy, 18(1), 53–74.
  • Birkland, T. A. (2006). Lessons of disaster. Washington, DC: Georgetown University Press.
  • Birkland, T. A., & Lawrence, R. G. (2009). Media framing and policy change after Columbine. American Behavioral Scientist, 52(10), 1405–1425.
  • Clarke, L. (2002). Panic: Myth or reality. Contexts, 1(3), 21–26.
  • Cobb, R. W., & Elder, C. D. (1983). Participation in American politics: The dynamics of agenda-building. Baltimore: Johns Hopkins University Press.
  • Cobb, R. W., & Ross, M. H. (1997). Cultural strategies of agenda denial: avoidance, attack, and redefinition. Lawrence, KS: University Press of Kansas.
  • Cohen, M. D., March, J. G., & Olsen, J. P. (1972). A garbage can model of organizational choice. Administrative Science Quarterly, 17(1), 1–25.
  • Cooper, C., & Block, R. (2006). Disaster: Hurricane Katrina and the failure of Homeland Security. New York, NY: Times Books.
  • deLeon, P. (1999). The stages approach to the policy process: What has it done? Where is it going? In P. A. Sabatier (Ed.), Theories of the Policy Process (pp. 19–32). Boulder, CO: Westview.
  • Easton, D. (1965). A systems analysis of political life. New York: Wiley.
  • Farley, J., Baker, D., Batker, D., Koliba, C., Matteson, R., Mills, R., et al. (2007). Opening the policy window for ecological economics: Katrina as a focusing event. Ecological Economics, 63(2–3), 344–354.
  • Gainsborough, J. F. (2009). Scandals, lawsuits, and politics: Child welfare policy in the U. S. States. State Politics & Policy Quarterly, 9(3), 325–355.
  • Gerber, B. J. (2007). Disaster management in the United States: Examining key political and policy challenges. Policy Studies Journal, 35(2), 227–238.
  • Gunter, V. J. (2005). News media and technological risks: The case of pesticides after Silent Spring. The Sociological Quarterly, 46(4), 671–698.
  • Hayes, T. L., & Neal, D. A. (2011). Actuarial rate review: In support of the recommended October 1, 2011, rate and rule changes. Washington, DC: Federal Emergency Management Agency.
  • Hilgartner, J., & Bosk, C. (1988). The rise and fall of social problems: A public arenas model. American Journal of Sociology, 94(1), 53–78.
  • Jenkins-Smith, H., Nohrstedt, D., Weible, C. M., & Sabatier, P. A. (2014). The advocacy coalition framework: Foundations, evolution, and ongoing research. In P. A. Sabatier & C. M. Weible (Eds.), Theories of the policy process (3d ed.). Boulder, CO: Westview Press.
  • Kingdon, J. W. (2011). Agendas, alternatives, and public policies (updated 2d ed). Boston: Longman.
  • Knowles, S. G., & Kunreuther, H. C. (2014). Troubled waters: The National Flood Insurance Program in historical perspective. Journal of Policy History, 26(03), 327–353.
  • Kousky, C., & Kunreuther, H. (2013). Addressing affordability in the National Flood Insurance Program. Working paper #2013-12. Philadelphia, PA: Risk Management and Decision Processes Center, University of Pennsylvania.
  • Lawrence, R. G., & Birkland, T. A. (2004). Guns, Hollywood, and school safety: Defining the school-shooting problem across public arenas. Social Science Quarterly, 85(5), 1193.
  • Lindsay, B., & Murray, J. (2011, April 12). Disaster relief funding and emergency supplemental appropriations. Congressional Research Service.
  • Majone, G. (1989). Evidence, argument, and persuasion in the policy process. New Haven, CT: Yale University Press.
  • May, P. J. (1990). Reconsidering policy design: Policies and publics. Journal of Public Policy, 11(2), 187–206.
  • May, P. J. (1992). Policy learning and failure. Journal of Public Policy, 12(4), 331–354.
  • May, P. J., Jochim, A. E., & Sapotichne, J. (2011). Constructing Homeland Security: An anemic policy regime. Policy Studies Journal, 39(2), 285–307.
  • May, P. J., & Williams, W. (1986). Disaster policy implementation: Managing programs under shared governance. New York: Plenum Press.
  • Molotch, H., & Lester, M. (1975). Accidental news: The great oil spill as local occurrence and national event. The American Journal of Sociology, 81(2), 235–260.
  • Morris, A. (2014). Hurricane Camille and the new politics of Federal Disaster Relief, 1965–1970. Journal of Policy History, 26(3), 406–426.
  • Mycoff, J. D. (2007). Congress and Katrina: A failure of oversight. State & Local Government Review, 39(1), 16–30.
  • Nakamura, R. T. (1987). The textbook policy process and implementation research. Policy Studies Journal, 7(1), 142–154.
  • Nohrstedt, D. (2009). External shocks and policy change: Three Mile Island and Swedish nuclear energy policy. Journal of European Public Policy, 12, 1041–1059.
  • Nohrstedt, D., & Weible, C. M. (2010). The logic of policy change after crisis: Proximity and subsystem interaction. Risk, Hazards & Crisis in Public Policy, 1, 1–32.
  • Perrow, C. (2008). Disasters evermore? Reducing our vulnerabilities to natural, industrial, and terrorist disasters. Social Research, 75(3), 733–752.
  • Pralle, S. B. (2006). Branching out, digging in: Environmental advocacy and agenda setting. Washington, DC: Georgetown University Press.
  • Rubin, C. B., Renda-Tanali, I., & Cumming. (2009). Disaster timeline: Major focusing events and U.S. outcomes, 1988–2008.
  • Sabatier, P. A. (1988). An advocacy coalition framework of policy change and the role of policy-oriented learning therein. Policy Sciences, 21, 129–168.
  • Sabatier, P. A. (1991). Political science and public policy. PS: Political Science and Politics, 24(2), 144–156.
  • Sabatier, P. A., & Jenkins-Smith, H. (1999). The advocacy coalition framework: An assessment. In Theories of the Policy Process. Boulder, CO: Westview.
  • Scavo, C., Kearney, R. C., & Kilroy, R. J. (2007). Challenges to Federalism: Homeland Security and disaster response. Publius: The Journal of Federalism, 38(1), 81–110.
  • Schattschneider, E. E. (1975). The semisovereign people. Hinsdale, IL: The Dryden Press.
  • Stone, D. A. (1989). Causal stories and the formation of policy agendas. Political Science Quarterly, 104(2), 281–300.
  • Stone, D. A. (2012). Policy paradox: The art of political decision making (3d ed.). New York: W. W. Norton.
  • Sylves, R. T. (2015). Disaster policy and politics: Emergency management and homeland security (2d ed.). Washington, DC: CQ Press.
  • Tierney, K., Bevc, C., & Kuligowski, E. (2006). Metaphors matter: Disaster myths, media frames, and their consequences in Hurricane Katrina. Annals of the American Academy of Political and Social Science, 604, 57–81.
  • Walker, J. L. (1977). Setting the agenda in the U.S. Senate: A theory of problem selection. British Journal of Political Science, 7, 423–445.