The Lasting Legacy of Hazard and Choice Perception in Flood Plain Management by Robert Kates

Introduction

The intensity and magnitude of flooding in many locations in the USA and around the world appears to be getting worse as a consequence of climate change, but major floods have happened throughout the history of the USA. For instance, costly historical flooding includes: the Mississippi River in 1874 and 1927; in Johnstown, Pennsylvania in 1889; and along the Ohio River in 1937. Federal legislation to assist with flooding came in the form of the House Committee on Flood Control in 1916 and the Flood Control Act 1917, but it was not until the 1930s that the Flood Control Act 1936 established a nationwide flood control program (Arnold, 1988). This act allowed the Army Corps of Engineers and other federal agencies to erect civil engineering projects—such as dikes, dams, and levees—to help combat flooding (Arnold, 1988).

This piece of legislation is where Hazard and Choice Perception in Flood Plain Management (Kates, 1962) begins. This University of Chicago Research Paper is, perhaps, Kates’ most seminal work and led him to a research career that examined people’s perceptions of natural hazards as well as how to help alleviate the unequal distribution of food supplies around the globe. As a student at the University of Chicago, Kates was mentored by Gilbert F. White, who has sometimes been called the “father of floodplain management” (Wescoat, 2006). Kates became acquainted with all that was involved with flood plain management and saw a niche in examining flood perceptions. This research work at Chicago culminated in his 1962 publication.

Kates’ 1962 publication

Chapter 1 (Geographic Research and Flood Damage Reduction) reviewed the 25 years of flood control authority following the enactment of the Flood Control Act 1936. Kates discussed problems with data collection and interpretation (especially damage estimates) based on the research that he examined. He outlined the following five correlations between flood control and flood damage.

Average flood damage had not decreased as a result of large investments in flood control works.

Kates then summarized what a flood damage reduction program might include. He discussed a theoretical choice of action (bearing the loss; emergency flood-fighting and evacuation; structural change and land elevation; changing land use; controlling the flooding; and flood insurance), what individual action may occur based on the choice of action, and the actions of various levels of government that encourage, reinforce, or mandate the individual action.

This chapter ends with Kates discussing the need for more accurate and usable flood hazard information, as well as a synthesis of human behavior to help explain why people occupy flood plains. Kates noted a shift in geographical studies from the physical factor of land and flooding, as White was doing, to examining changes in urban and rural flood plain occupancy, attitude studies, disaster studies, and decision-making studies.

In Chapter 2 (Decision-Making and Resources Management), Kates discusses decision theory and decision-making schema. He outlines some underlying factors that appear in most decision-making processes, including: an underlying view of the rationality of humans; the type of decision processes involved (conscious choice, habitual choice, and unconscious or trivial choice); the conditions of knowledge under which the choice is made (risk and uncertainty, and dispersion of knowledge); and the criteria used to guide such choices (prescriptive and descriptive criteria). Kates relates how each of these items relates to resource managers and their decisions.

The main focus of Kates’ research, the actual study, begins in Chapter 3 (The Study Site: Strategy, Sites, and Methods). Kates introduced LaFollette, Tennessee, the main town of his study, which had experienced a major flood in 1950. He provides information on the geographic, social and economic characteristics of the town, which is about 64 km (40 miles) north of Knoxville and lies in the flood plain of Big Creek. He outlines the flooding problems in the town, such as having a major reservoir nearby, the fact that Big Creek runs through the town, and how increased development (mostly commercial and industrial) has exacerbated this hazard.

To obtain useful information from each type of manager, Kates used a 16-page questionnaire designed for one of two types of flood plain manager (residential or commercial). Kates also made field observations of the types of land use units that had the potential to be damaged by flooding, noting any visible damage from prior flooding. The residential managers were home-owners or landlords with property on the flood plain. These managers indicated that the outskirts of the town were hillier and disadvantageous for older residents, so they endure the odor from Big Creek, dampness in their homes, and damage from flooding when it occurs. The commercial managers were business owners with businesses located on the flood plain. These people tended to be the elite of the town and provided a strong indication that they would like to relocate closer to the main highway south of the town to improve business, not to alleviate the issues associated with flooding. The chapter ends with a brief description of the five towns that Kates uses as comparisons with LaFollette: Aurora, IN; Darlington, WI; Desert Hot Springs, CA; Cerrito Creek, CA; and Watkins Glen, NY.

Kates’ interpretation of flood hazard information from all sites is discussed in Chapter 4 (Probability and Hazard Evaluation). He carries out a focused case study of LaFollette, noting that the majority of respondents saw the 1950 flood as one in a pattern and not a single event; some respondents felt that dredging Big Creek removed future flood hazards. People had different ideas about why the 1950 flood event occurred. Half of the residential managers were skeptical of planning efforts, whereas only about 5 per cent of commercial managers were skeptical. There was a strong association between interpretation classes (expectation of future floods and interpretations related to affirmative, negative, and uncertain expectations). Kates also states that there was a lack of association between future flood expectations and variables such as education and length of time on site. In terms of the comparison sites, most flood plain managers in Aurora and Darlington had a high certainty of future floods; El Cerrito managers had an intermediate certainty of future floods; and managers in Desert Hot Springs and Watkins Glen felt that future floods had a low probability.

The range of practical choices that people have for reducing or dealing with flood damage is discussed in Chapter 5 (The Perception and Adoption of Alternative Flood Damage Reduction Measures). The first choice discussed is the individual bearing the loss. Interestingly, the managers did not expect to bear the loss if they did not perceive the hazards and bearing the loss was more attractive if the community helped with relief activities. Another choice was to take emergency precautions, such as sand-bagging or temporarily evacuating people and materials from the flood’s path. Most commercial managers felt that there was at least one emergency action that they would take in the case of a flood, whereas less than a quarter of the residential managers identified an emergency action. Changes in structure or land elevation was the third choice, but could be difficult to implement because of cost, especially to residential managers. Included in this choice is flood-proofing buildings, elevating older structures, bricking-up basement windows and installing sump pumps. Another choice is flood control and abatement using engineering works or eliminating bottlenecks in the river. The last choice is insurance, which was not common in 1962. Insurance agents in LaFollette did not know the conditions under which flood insurance could be obtained, so did not offer the option to their clients. This is another choice that may be prohibitive for some people in terms of cost. In addition, having insurance reduced the likelihood of adopting other measures. Kates specified that managers have a different quantity and quality of perceptions on these range of choices, which are based heavily on experience and observations.

Some of these choices had been used in the comparison towns. Elaborate elevation and the removal of goods and furnishings to higher elevations, sump pumps, and back-up valves had been installed in Aurora. In Darlington, in addition to the elevation changes, a local warning system has been installed and some upstream work had been carried out to alleviate flooding. Retaining walls, sand-bagging, sump pumps, and annual inspections of emergency and structural measures had been implemented in El Cerrito. Desert Hot Spring and Watkins Glen had done very little, if anything, to reduce flood damage.

Kates ends this chapter with a discussion of some generalizations he noticed about the decision-making processes of these towns. These noted generalizations are: (1) those who actually expect to bear future losses try to do something about it; (2) elaborate adjustments to floods have evolved by repeated experiences; (3) full attention to floods comes at the moment of a flood and shortly after it in areas that do not believe there is a high certainty of a flood occurring; (4) in areas where perceived loss or losses have been experienced, there is a higher motivation to spend money and effort to reduce future damage.

In Kates’ final chapter (Information and Decision-Making in Flood Damage Reduction), he attempts to answer the age-old question “Why do they live there?” His answer is that there are many reasons why people continue to live on a flood plain. Kates summarizes decision-making by flood plain managers in terms of bounded rationality, choices of mechanisms, conditions of knowledge, and evaluation criteria. He rounds the chapter out with a discussion on how improved information may affect managers’ actions to reduce damage from flooding, which includes flood frequencies, experience, simplification of choice, the effect of the certainty–uncertainty scale on information, and improving the effectiveness of the information provided. In his own words (Kates 1962: 144; gendered usage original):

The paradox of rising damages with increased flood control with which this study began will probably exist for some time. It is not a major problem as problems go – the catastrophe of floods and the magnitude of flood control expenditures are dwarfed by the accepted realities of the nuclear age. Yes in the need for new approaches, insights, and actions, it is symptomatic of a variety of resource problems accompanying the increase of man’s numbers and the spread of his works.

Lasting significance of the 1962 publication

Kates’ work was revolutionary because he introduced ideas about how commercial and residential managers perceived flood hazards, how they think about flooding in their decision-making, and how perception factors into methods of dealing with flood damage and losses.

This work started a new awareness of the flood plain and its natural resources and functions, and much work has continued from it. Some of these important milestones include: a guidebook created by the Federal Interagency Floodplain Management Taskforce that aids communities in developing a local program for flood plain management and protection; the National Flood Insurance Reform Act 1994, which established a task force to further identify flood plain functions and provide suggestions on how to reduce flood losses while protecting these functions; and the establishment of buyout and relocation programs from structures in high-risk flood zones (Wright, 2000). A report from the Federal Interagency Floodplain Management Task Force (1992) noted that substantial progress had been made in institutional awareness of the flood risk, but that individual awareness was still lacking. This continued to put people in danger of flooding, although national flood protection standards and changes in construction practices had improved flood protection. Disaster assistance had also been increased since the publication of Kates’ work. This assistance has reinforced people’s expectations of government aid in times of flood disasters, which also harkens back to this publication.

The 1970s and 1980s were an era of flood insurance. The National Flood Insurance Program was enacted in 1968 (Wright 2000). Flood insurance was something that Kates mentioned as being a way that people could combat flood losses, but found that it was not offered to clients because agents were not familiar with the rules and criteria for obtaining it. Now flood insurance is available to anyone living in a National Flood Insurance Program community and is mandatory for anyone (residential or commercial) in high flood risk areas with a government-backed mortgage (FEMA, 2021).

Climate change is keeping Kates’ work relevant today. As a result of global climate change, flood plains in the USA are expected to grow about 45 per cent by the end of the century (Denchak, 2019), which means more people are at risk of flooding (coastal, flash, and urban floods). Understanding why people occupy these areas is an important step in helping them survive (physically, mentally, and financially). Many of the reasons that Kates noted as to why people live and work on flood plains in 1962 are similar to the reasons why people continue to inhabit them today. However, people are more aware of flooding, the assistance that is available, the modifications that can be made to existing structures, and new construction codes and materials because of his work and that of his contemporaries. As long as people live in flood zones, Kates’ work will always be relevant, important to revisit, and serves as a reminder of how far flood plain management and perceptions have come since the 1960s.