Economic applications in disaster research, mitigation, and planning



Download 97.41 Kb.
Page1/3
Date14.05.2016
Size97.41 Kb.
  1   2   3
Economic applications in disaster research, mitigation, and planning
Terry L. Clower, Ph.D.

Associate Director

Center for Economic Development and Research

University of North Texas

P.O. Box 310469

Denton, Texas 76203-0469

tclower@unt.edu.

Abstract

This chapter examines the contributions of the economics discipline to disaster research, mitigation, and planning. Economics offers modeling techniques for assessing the impacts of disasters, theories of development for understanding the choices that individuals and firms make in selecting residential and business locations, approaches for risk and vulnerability assessment in insurance and disaster planning, and policy insights in each of these areas that are affected by the political economy. The chapter gives particular attention to common and emerging techniques for assessing the indirect economic impacts of disaster events offering an assessment of the strengths and weaknesses of each analytic approach.





Introduction

Economics as a specific discipline, its many sub- and closely related disciplines, and research techniques pervade the systematic study of disasters and their human, social, and monetary impacts. The goal of this chapter is to provide the non-economist a look into how this discipline shapes scholarly and public understanding of disaster impacts, the roles that economic information can play in the realpolitik of disaster management and response, and reviews methods of analysis in assessing the impact of disasters.

To accomplish this goal, specific data analysis techniques used to estimate the economic impacts of disasters are presented along with a description of how this information is used to address issues of resource allocation and disaster avoidance. Discussion is presented on issues relating to disaster insurance and the contribution of economics to risk analysis. Finally, the chapter offers suggestions for expanded or new research approaches that economists should undertake to further contribute to the discipline of disaster management. But first, a historical perspective of the role of economics in disaster research.
A Brief Time in History

Assessing the economic impacts of disasters is a very recent systematic field of study. Disasters have been, and continue to be, human tragedies. History books tell us that more than 2,000 died in the Johnstown, Pennsylvania flood in 1889, the eruption of Krakatoa in 1883 – described as the first catastrophe of the communications age (USGS, 2005) – and the resulting tsunami killed more than 30,000, the Galveston hurricane of 1900 killed more than 6,000 of the island’s residents, and of course the 1,503 lives lost in icy North Atlantic waters on that ‘night to remember’ in 1912. These numbers represent horrific human tolls and each also represents economic losses in the tens or hundreds of millions of dollars. But, it is the loss of life that catches our attention.1

In addition, the provision of public monies to help those affected by disasters has been a comparatively recent occurrence. Historically, government policy and/or public sentiment simply did not support monetary aid to disaster victims. Barnett (1999) cites an example from 1887 where President Grover Cleveland in response to an emergency request for $10,000 in aid to Texas drought victims noted that there is no constitutional basis for public funds to be used to offset individual suffering as a result of a disaster. Barnett also observes that even though public policy had changed by 1915 with the advent of federal disaster relief grants and loans, it was many years before the public at large found the receipt of these grants and loans socially acceptable.

Concentrating on the loss of life to describe the magnitude of disasters, combined with public attitudes about the costs of disasters being borne by individuals, there was little demand for comprehensive economic assessments of disasters. One of the few early assessments of the economic impacts of a disaster was published in 1920 estimating the impacts of the Halifax ship explosion of December 1917 (Scanlon, 1988). Little else appears in the academic literature for more than 40 years, but during those years, public policy and public attitudes about disaster relief changed. With these changes came demand for information about the size (impact) of disasters from an economic perspective. If there were to be programs to provide aid to victims of disasters, then the impacts must be quantified.

The development of warning systems broadcast over radio networks and later television gave vital information that has saved innumerable lives. In addition, investments in infrastructure, enhanced construction techniques required by modern building codes, and other physical capital have with one notable exception resulted in fewer deaths due to disasters. For example, following the 1900 hurricane, Galveston Island and almost every structure on the island were raised several feet. Hurricanes have hit Galveston since 1900 but never with anything near the human losses of the 1900 event. As this chapter is being completed, recovery is underway for hurricane Katrina. In the largest disaster to hit a US city since the San Francisco fire, New Orleans, a city of 450,000, was inundated by flood waters after sections of the Mississippi River levee system failed due to storm-related flooding. (The City of New Orleans sits several feet below sea-level and has been a high-risk area for flooding since its founding in the late 18th century.) Inefficient and ineffective government response is being blamed for many of the city’s low-income population not being evacuated. Whether through inability to evacuate, or unwillingness by individuals to evacuate, over 200,000 people were still in the city when the levees broke. Still, less than one-half of one percent of the population perished.

Even with record numbers of people moving into relatively hazardous areas, such as the Florida coast or mud-slide prone hills in central and southern California, until Katrina we have seldom seen more than a few deaths in the US related to natural disasters since the early parts of the 20th century. To justify on-going public aid to victims and expenditures for disaster preparedness and management, efforts turned to estimating the economic impacts of disasters.


Political Economy of Disasters

Prior to the 20th century political economy was the proper name for the discipline of economics. In today’s context it means the convergence of politics and economics. In the previous section changing public policy in the US is illustrated by comparing President Cleveland’s strict interpretation of the constitution with the later advent of federally funded grants and loans to aid victims of disasters. The economic considerations were, in many respects, the same, but our policy (political) approach had changed. Economic analysis is at the heart, but is far from the whole, of the realpolitik2 of disasters. From the time of the nation’s founding through 1950, the US government enacted 128 pieces of legislation providing relief, mostly in the form of in-kind donations, for victims of disasters (Barnett, 1999). By the 1950s, the US had gone through a fundamental shift in the expected role of government. From the New Deal policies of the 1930s through the G.I. Bill providing for a college education to veterans of World War II, liberal ideas of government responsibility to the nation’s citizens was in its ascendancy. The Disaster Relief Act of 1950 and the Small Business Administration Act of 1953 both offered standing programs for disaster relief (Barnett) requiring economic analyses to support budget projections.

Programs of the Great Society of the 1960s and afterwards also included elements of disaster relief and mitigation in housing and introduced formal civil rights considerations in disaster management and planning. In 1953, the federal government provided just 1% of total disaster relief spending. By the mid-1970s that percentage had risen to more than 70% (Barnett, 1999).

Political considerations also influenced the distribution of private relief money. Prior to Hurricane Camille in 1969, the American Red Cross distributed disaster assistance based on economic need. After being heavily criticized in the press and in some political circles, the Red Cross standardized their rules for funds eligibility and removed economic need as a criterion. This can be seen as a reflection of the growing size and political influence of the middle class in the US after World War II. As observed in surveys conducted by Leitko et al. (1980), middle class victims of disasters view relief as “a corrective to a naturally induced injustice” (page 735) and tend to demand larger amounts of relief regardless of their own resources. This liberal approach to the distribution of disaster relief has survived the increasingly conservative nature of other public assistance in the US since the early 1980s. Leitko, et. al. observed that the public does not see disaster relief as welfare. How else can one politically account for general acceptance at the national level for potentially providing grants and low interest loans to wealthy families whose homes in gated Florida communities are damaged by hurricane events?

The good news about the surprisingly liberal attitudes of the US electorate towards disaster relief and mitigation is that we have had a steady, if not sufficient, stream of economic resources for disaster mitigation, preparation, and management. Of course, this has also been influenced by the realpolitik of 9-11 and there will certainly be a shift in federal government spending policy due to failures and perceived failures that led to the New Orleans/Katrina disaster – at least in the short run. The bad news is that federal intervention is increasingly distorting economic decisions at the local level. Kunreuther (1998) notes local governments are not seeking own-source solutions to disaster response needs, such as private sector insurance, because of perceived certainty of federal resource availability. These distortions are also apparent in residential real estate markets.

One of the clear reasons the costs of disasters have escalated rapidly in recent years is a function of the level of development in high risk areas. For example, in 2003, 153 million people, 53 percent of the US total population, resided in coastal counties – an area that comprises just 17 percent of the mainland US land mass (Crossett, et al, 2004). The coastal population has increased by 33 million in 23 years representing a rapid increase in population density and significant development intrusions into barrier islands and marsh lands that offer natural protection from storm events. Moreover, this population growth does not reflect the growth in the number of second and vacation homes, hotels, and resorts that increasingly fill the coastal landscape. The political reality is that local and state governments are willing to trade the potential for more expensive disaster events, which is offset by federal assistance, for tax base growth.

There are three other ways that political economy approaches can help explain the level and distribution of disaster mitigation and planning funding. The first is the political dimension of who qualifies for post-disaster assistance. The release of federal grants and loans for disaster relief is based on the declaration by the President that a specified region, most often a county, is a “disaster area.” The general public largely thinks this designation is about damage to buildings, homes, and infrastructure along the lines of the Fujita Scale of tornadic damage. However, it is the impact the disaster event has on local government that forms the basis for a disaster declaration. In theory, local government or state government are supposed to provide disaster assistance. If demand for assistance and services exceeds local capacities or if local government revenues are substantially threatened, then Federal resources are engaged. If a hotel is damaged by a tornado, as in the case of Fort Worth, Texas, in 2000 (see McEntire, 2002 for a description), local government experiences losses in revenue from sales taxes, hotel occupancy taxes, and property taxes and thus local government’s ability to provide services and recovery aid is diminished.3 This interesting quirk of US disaster policy is keenly felt by victims of certain types of disasters. Tornadoes can cause widespread damage qualifying the area for Federal disaster assistance. However, if the tornado destroys only houses located along one block, it is doubtful that the revenue of local government would be severely impacted and those victims will not qualify for federal assistance – even though their individual loss is as great as any individual in a much larger disaster.

In practice, presidential disaster declarations can be overt acts of political largess or electioneering. Sylves (1996) noted that in the winter of 1995, President Clinton waived qualification rules repeatedly in making federal funds available to residents and businesses in California as a result of two flood events. The fact that California had a Democratic governor at the time and holds the largest number of electoral votes in presidential races is assumed to have played a role in Mr. Clinton’s decision. In the spring of 1996, widespread flooding causing substantial damage occurred across Pennsylvania, yet only six counties were declared eligible for federal disaster assistance. Governor Tom Ridge publicly threatened consequences in the fall elections for federal officials “playing games with Pennsylvania.” Very quickly, 58 of 67 Pennsylvania counties received federal disaster area status (Platt, 1999). Platt describes the political influence in federal assistance as “disaster gerrymandering.”

Public policy also affects private insurance approaches to economic mitigation of disaster impacts. The insurance industry remains one of the most heavily regulated industries in the US with many states having oversight bodies approving rates based on allowable underwriting profitability.4 The problem is that the event horizons for disasters are often long, meaning that the insurer’s premiums should account for building risk event reserves over several years. However, accumulating risk event reserves can appear as profits in the short run and are thus the targets of regulators looking to deliver politically popular rate decisions. Moreover, accounting rules and taxing policies on retained earnings hurt insurers’ ability to build risk reserves (Andersen, 2004). Together, these policy factors result in wide fluctuations in disaster insurance availability5 and premiums with the lowest availability/highest rates following disaster events. These higher rates discourage private sector adoption of own-source risk mitigation increasing the dependence on federal level solutions (Klein, 1998). In addition, closely timed disaster events, such as this year’s early season hurricane in Florida just 11 months after the last major storm event, place further strains on insurance provider resources that are reflected in subsequent premiums.

The final political economy dimension to disaster research covered here is the potential for overt political considerations in the distribution of disaster relief. Though they specifically studied an Australian case, Butler and Doessel (1980) claim that politics can influence disaster relief in a federalist system of governance. Sverny and Marcal (2002), Scanlon (1988), and McEntire and Dawson (forthcoming) also discuss the politics of disaster relief and preparedness. Some disaster mitigation projects could be considered little more than pork-barrel politics. Moreover, as suggested earlier, there is more than a little of the political economy of wealth redistribution in some disaster policies in the US.

Several techniques and approaches will be presented in the remainder of this chapter for estimating the economic impacts of disasters and disaster planning and management. However, the application of the findings of these analyses remains an exercise in political economy.
Measuring Disaster Losses

Economists are rarely called on to estimate the direct physical damage caused by disasters. This is a job for engineers, architects, construction specialists, and others. These damages include property damage to buildings and infrastructure, debris removal, and the cost of emergency protective services (McEntire and Cope, 2004). It is the losses associated with employment income and indirect losses that occupy the efforts of economists in the field of disaster research. Though there is some disagreement among scholars as to exactly what counts as indirect costs, they include the loss of business activity due to reduced activities at damaged firms, loss of income in secondary and tertiary employment, and business disruptions not directly attributable to damage. For example, if a manufacturing firm is damaged sufficiently to disrupt production, then they will not require trucking services to deliver raw materials or pick up finished goods, which may impact the employment of drivers. Rose (2004) illustrates indirect effects with the example of a utility plant being damaged resulting in utility customers (businesses) not being able to operate. Cochrane (2004) uses the comparatively simple definitions that direct damage is property damage plus lost income, and indirect damage is anything else. Rose, along with other researchers cited in his study, find that direct and indirect business interruption losses can be as large as physical losses. Of course, the degree of impact of a disaster depends in large part on the scale of the analysis.


Macroeconomic Analyses

Macroeconomic analysis considers economic events and activities at a national or at least state scale. Dacy and Kunreuther (1969) held that the total national cost of a disaster is the replacement value of the property damaged, regardless of the presence of a relief program. Even when other costs are included, it is a matter of simple division to see that disaster impacts rarely have a meaningful impact on a national economy. Whatever the damage, the divisor is very large. As noted by Mileti (1999), capital markets are simply too large to be disturbed beyond a short period of time by natural disasters.6 The notable exception would be sustained droughts in countries with an agrarian-based economy (Albala-Bertrand, cited in Horwich, 2000). Nobel Laureate Gary Becker (2005) has noted that even the pandemic flu of 1918-1919 had no major effect on the world economy. To illustrate how this can be, Horwich (2000) offers an example based on the Kobe earthquake.

The Great Hanshin earthquake struck Kobe, Japan on January 17, 1995. In the earthquake and subsequent fires, more than 100,000 businesses were destroyed, 300,000 individuals became homeless, and 6,500 people were killed with total damages estimated at $114 billion (Horwich, 2000). The damage estimate represented about 2.5% of Japanese gross domestic product (GDP) in 1995. Yet within 15 months manufacturing was operating at 98% of the pre-earthquake trend, all department stores and 78% of small shops had reopened within 18 months, and trade at the port was operating close to pre-earthquake levels within 1 year (Horwich; Landers, 2001). That is a remarkable recovery based on GDP impacts. Similarly, Hurricane Katrina destroyed a sizable proportion of the economic capacity of Louisiana and Mississippi, but these states combine to represent less than 2% of US GDP. The most recent data from the US Department of Labor estimates that Katrina took 230,000 jobs from directly affected areas, but that total national employment for the month of September declined by only 35,000 – little more that a statistical blip on the economic map (Balls and Swann, 2005). Horwich suggests it is more telling to consider the impact of a disaster on economic potential as opposed to economic activity.

Economic potential can be measured by the level of capital stock including unused capacity in the economy. For example, other Japanese ports took on much of the trade activity while Kobe was under repair. In addition, Horwich (2000) suggests that human capital is the dominant economic resource and that, horrible as the losses were, 99.8% of the population in the earthquake impact zone survived. Horwich includes the economic value of life at $2 million per person, plus the $114 billion damage to capital stock, to estimate the capitalized value of the Hanshin earthquake on Japan at $127 billion ($114 billion +(6500*$2 million)). Horwich calculates Japan’s total resource value by capitalizing GDP (about $5 trillion in 1995) at a real interest rate of 3% for a total of $167 trillion ($5 trillion/0.03), which includes the value of a highly skilled workforce. Using this approach, the Great Hanshin earthquake had a total impact of 0.08% of the economic potential of the Japanese economy. Much of the economic activity lost due to the physical damage was regained in the form of rebuilding and repair. While Horwich makes some heroic assumptions in these calculations, they offer a clear indication of the resilience of the economy of large industrialized nations. However, even smaller nations, in terms of economic output, appear to possess economic resilience to disaster events.

One week after the Sumatra tsunami of 2004, the Indonesian and Malaysian stock markets had gained value from the pre-disaster level, the Thai stock market declined only slightly, and the Sri Lankan markets were off a few percent (Becker, 2005). Tavares (2004) using an ordinary least squares regression analysis calculates that natural disasters lower US GDP by 0.052% per year.7 Of course, the same may not hold true for smaller nations with more specialized economies.

In addition to the previously mentioned agrarian-based economies, Auffret (2003) finds that natural disasters are an important determinant of economic volatility in Caribbean economies, which is attributed, in part, to consumption shocks due to underdeveloped or ineffective risk management mechanisms. Of course, tourism-based economies are subject to market responses to disaster events – or predictions of disaster events – over which they have little control.

The other factor that minimizes the impacts of most disasters is their short duration. Waters recede, storms pass, and eventually droughts break. But for some types of disaster, the threat of an event can have a long-term effect on macroeconomic performance – specifically the threat of terrorism. Tavares (2004) estimates that the continuous threat of terrorist attacks reduces gross domestic product in Israel by 4%. The Basque region of Spain, which has seen decades of separatist terrorist activities, looses about 10% of its potential economic activity due to the threat of terrorism. Terrorism impacts national economies in 3 ways: 1) increased risk decreases business insurability meaning that risk is not spread across a greater number of economic actors, 2) trade costs are increased leading to lower levels of international transactions, and 3) increased public and private spending for security and defense decreases capital available for investment (Tavares). Hobijn (2002) estimates that increased security costs incurred after 9-11 has reduced US economic activity by 0.66%.

One area of national level impacts that has received press coverage, but little academic analysis to date, is the impact of disasters on the US energy industry. In 2004, hurricanes in the Gulf of Mexico substantially damaged that region’s oil and gas production and transmission capacity. Winds and high waves toppled or dislodged the moorings for offshore rigs, and hurricane-spawned underwater mudslides destroyed sections of transmission pipelines. This damage resulted in lower domestic energy supplies that increased the market price for oil and gas and was reflected in the cost of gasoline, diesel, and fuel oil that rippled throughout the US economy.8 Damage sustained by refineries located in the New Orleans region along with off-shore oil production losses as a result of Katrina and subsequent flooding is currently blamed for adding as much as 40 cents to the price of a gallon of gasoline at the pump. These impacts, though temporary, should be formally assessed.

The resilience a given economy has to disaster events is, of course, largely dependent on national resources committed to mitigation, planning, and response. Horwich (2000) reports comments by noted disaster researcher Fred Cuny stating that if the earthquake that hit San Salvador had instead hit San Francisco, it would have rattled the china, not killed 1,500 people. As national income rises, disaster costs tend to rise, but relative costs as well as the number of lives lost decrease (Dacy and Kunreuther, 1969; Freeman et al., 2003).

However, aggregated analyses at the macroeconomic level miss the intensity of regional and local impacts that create comparative winners and losers when disaster strikes. In addition, macro level analysis often fails to identify and address disaster impacts and vulnerability across populations at differing income levels. As an overall economy gains wealth, it is often the case that low income populations are forced to reside in lower-cost / higher-risk areas compounded by their inability to afford insurance (Barnett, 1999; Scanlon, 1988; Vatsa, 2004). The stark, often horrific, images of the low-income victims of Hurricane Katrina and their disproportionate death rate and loss of most all worldly goods has brought into focus how disaster events can disproportionately affect the poorest segments of our population. Even when the national or regional economy recovers in terms of production and employment, specific localities, groups, and individuals may still be paying the price of disasters.

It is said that all politics are local. Given the earlier assertion that politics intertwines disaster economics and policies, it is reasonable to assume that the politics of disaster are often local. This is one reason why the preponderance of studies examining the economic impacts of disasters are conducted at the local or regional level.




Share with your friends:
  1   2   3




The database is protected by copyright ©essaydocs.org 2020
send message

    Main page