GAO Issues Report on Economic Effects of Hurricanes Katrina, Sandy, Harvey, Irma

WASHINGTON, Sept. 11 -- The Government Accountability Office has issued a report (GAO-20-633R) entitled "Natural Disasters: Economic Effects of Hurricanes Katrina, Sandy, Harvey, and Irma".

The report was sent to Sen. Gary Peters, D-Michigan, ranking member of the Senate Homeland Security and Governmental Affairs Committee, Sen. Marco Rubio, R-Florida, chairman of the Senate Small Business and Entrepreneurship Committee, Rep. Peter DeFazio, D-Oregon, chairman of the House Transportation and Infrastructure Committee, and Rep. Sheila Jackson Lee, D-Texas, on Sep. 10, 2020. Here are excerpts of summaries associated with the report.

What GAO Found: "Hurricanes Katrina, Sandy, Harvey, and Irma (selected hurricanes) caused costly damages and challenges for some populations in affected communities. In these communities, the National Oceanic and Atmospheric Administration (NOAA) estimated the cost of damages to be approximately $170 billion for Katrina, $74 billion for Sandy, $131 billion for Harvey, and $52 billion for Irma. These estimates include the value of damages to residential, commercial, and government or municipal buildings; material assets within the buildings; business interruption; vehicles and boats; offshore energy platforms; public infrastructure; and agricultural assets. These hurricanes were also costly to the federal government. For example, in 2016, the Congressional Budget Office (CBO) estimated that federal spending exceeded $110 billion in response to Katrina and $53 billion in response to Sandy.

GAO analysis suggests that the selected hurricanes were associated with widely varying effects on overall economic activity and total employment in affected metropolitan areas and counties. Economic activity was lower than expected in the month of the hurricane or some of the three subsequent months in three of the affected metropolitan areas GAO analyzed. Within one year, average economic activity in these three metropolitan areas was similar to or greater than what it had been the year before the hurricane. Total employment was lower than expected in the month of the hurricane or some of the three subsequent months in 80 of the affected counties GAO analyzed. Total employment was higher than pre-hurricane employment on average in 47 of those counties within one year but remained below pre-hurricane employment on average in the other 33 counties for at least one year. Finally, state and local government officials said that the selected hurricanes had significant impacts on communities, local governments, households, and businesses with fewer resources and less expertise, and that challenges faced by households may have impacted local businesses.

Communities affected by selected hurricanes have been taking actions to improve resilience, but multiple factors can affect their decisions. Actions taken after selected hurricanes include elevating, acquiring, and rehabilitating homes; flood-proofing public buildings; repairing and upgrading critical infrastructure; constructing flood barriers; and updating building codes. A community's decision to take resilience actions can depend on the costs and benefits of those actions to the community. Multiple factors affect these costs and benefits, including the likelihood, severity, and location of future disasters, as well as the amount of federal assistance available after a disaster.

Finally, vulnerabilities remain in areas affected by selected hurricanes. For example, state and local government officials indicated that many older homes in these areas do not meet current building codes. In reports to the Federal Emergency Management Agency (FEMA), states indicate they anticipate that the scope of damages via exposure to weather hazards, such as hurricanes, will likely remain high and could expand across regions affected by the selected hurricanes. In addition, some local governments have projected that population will grow in the regions affected by selected hurricanes."

Why GAO Did This Study: "Between January 1980 and July 2020, the United States experienced 273 climate and weather disasters causing more than $1 billion in damages each, according to NOAA. The total cost of damages from these disasters exceeded $1.79 trillion, with hurricanes and tropical storms accounting for over 50 percent of these damages, according to NOAA. Across the regions affected by these hurricanes over the period from 2005 to 2015, CBO estimated that federal disaster assistance covered, on average, 62 percent of the damage costs. GAO has reported that the rising number of natural disasters and reliance on federal disaster assistance is a key source of federal fiscal exposure.

GAO was asked to review the costs of natural disasters and their effects on communities. This report examines (1) estimates of the costs of damages caused by hurricanes and hurricanes' effects on overall economic activity and employment in the areas they affected, and (2) actions subsequently taken in those areas to improve resilience to future natural disasters. GAO conducted case studies of Hurricanes Katrina, Sandy, Harvey, and Irma, selected for two reasons. First, they were declared a major disaster by the President under the Robert T. Stafford Disaster Relief and Emergency Assistance Act, which establishes key programs through which the federal government provides disaster assistance, primarily through FEMA. Second, they had sizable effects on the 50 U.S. states and the District of Columbia during the period from 2004 through 2018. GAO analyzed federal agency and other data on costs, economic activity, employment, and recovery and mitigation projects in selected areas affected by these hurricanes. GAO also visited selected recovery and mitigation project sites; interviewed experts and federal, state, and local government officials; and reviewed federal, state, and local government reports and academic studies."

* * *

September 10, 2020

To: The Honorable Gary C. Peters, Ranking Member, Committee on Homeland Security and Governmental Affairs, United States Senate

The Honorable Marco Rubio, Chairman, Committee on Small Business and Entrepreneurship, United States Senate

The Honorable Peter A. DeFazio, Chairman, Committee on Transportation and Infrastructure, House of Representatives

The Honorable Sheila Jackson Lee, House of Representatives

Natural Disasters: Economic Effects of Hurricanes Katrina, Sandy, Harvey, and Irma Between January 1980 and July 2020, the United States experienced 273 climate and weather disasters for which the National Oceanic and Atmospheric Administration (NOAA) estimated damages costing $1 billion or more each./1 NOAA estimated that the total cost of damages from these disasters exceeded $1.79 trillion and attributed over 50 percent of these costs to hurricanes and tropical storms. The Congressional Budget Office (CBO) estimated that federal disaster assistance covered, on average, 62 percent of the costs across the regions affected by these hurricanes for calendar years 2005 through 2015./2 GAO has reported that the rising number of natural disasters and reliance on federal disaster assistance is a key source of federal fiscal exposure./3

You asked us to review the costs of natural disasters and their effects on communities. This report examines (1) estimates of the costs of damages caused by hurricanes and hurricanes' effects on overall economic activity and employment in the areas they affected, and (2) actions subsequently taken in those areas to improve resilience to future natural disasters.

To address these issues, we identified hurricanes that were declared a major disaster by the President under the Robert T. Stafford Disaster Relief and Emergency Assistance Act (Stafford Act), which establishes key programs through which the federal government provides disaster assistance, primarily through the Department of Homeland Security's Federal Emergency Management Agency (FEMA)./4 We examined FEMA and NOAA data on hurricanes and selected as case studies the four hurricanes--Katrina (August 25-30, 2005), Sandy (October 30-31, 2012), Harvey (August 25-31, 2017), and Irma (September 6-12, 2017)--that caused the most damages in the 50 U.S. states and the District of Columbia from 2004 to 2018./5 To examine hurricanes' effects on overall economic activity and employment in the areas they affected, we first used FEMA data to identify counties (and parishes/6) eligible for its Individual Assistance (IA) and/or Public Assistance (PA) programs after each selected hurricane (affected counties). Then, we used Census Bureau data to identify metropolitan areas that overlapped with affected counties (affected metropolitan areas). One limitation of this approach is that the severity of a hurricane and the damages it caused may have varied across affected counties even though they were all eligible for assistance. Similarly, the severity of a hurricane and the damages it caused may also have varied across affected metropolitan areas, even though they all overlapped with affected counties.

To describe overall economic activity in affected metropolitan areas, we used monthly indices of economic activity for the period from February 1990 through December 2019 created by researchers at the Federal Reserve Bank of St. Louis and Saint Louis University to reflect labor market, housing market, and credit market activity, as well as overall income and output./7 These indices are only available for large metropolitan areas. For each selected hurricane, we therefore analyzed economic activity in each of the large metropolitan areas that was affected by the hurricane./8 We examined whether economic activity in the month of the hurricane and the subsequent three months was lower or higher than the average of economic activity over all months. We report that economic activity was lower or higher than expected in a given month when it was lower or higher than average economic activity and the difference is statistically significant at the five percent level. In metropolitan areas where economic activity was lower than expected in the month of the hurricane or in any of the subsequent three months, we also examined whether the average of economic activity over the first and second years after the hurricane was lower or higher than the average of economic activity over the year before the hurricane to describe how economic activity evolved after the initial shock. We calculated the average of economic activity over a year as the average of monthly economic activity for each of the 12 months of that year.

To describe employment in affected counties, we used monthly employment data from the Bureau of Labor Statistics' (BLS) Quarterly Census of Employment and Wages for the period from August 2000 through September 2019. For each affected county we analyzed, we compared total employment the month of the hurricane and the subsequent three months to the average of total employment over all months after accounting for long-term trends and seasonal variation./9 We report that total employment was lower or higher than expected in a given month when it was lower or higher than average total employment after accounting for long-term trends and seasonal variation and the difference is statistically significant at the five percent level. In counties where total employment was lower than expected in the month of the hurricane or in any of the subsequent three months, we also examined whether the average of total employment over the first and second years after the hurricane was lower or higher than the average of total employment over the year before the hurricane, in order to describe how total employment evolved after the initial shock. We calculated the average of total employment over a year as the average of monthly total employment for each of the 12 months of that year.

Finally, we compared the distribution of employment across economic sectors for all affected counties one year after each hurricane to the distribution before the hurricane. Economic sectors in the BLS data we analyzed include construction; manufacturing; natural resources; education and health; finance; leisure and hospitality; professional and business services; trade, transportation, and utilities; information; and other private services./10

Our analyses of economic activity and employment have limitations and our results should be interpreted with caution. The patterns we observed in economic activity and employment may have occurred even in the absence of the hurricanes, and we cannot isolate the effects of the hurricanes from the effects of other events that occurred at the same time. In addition, hurricanes may have had effects on economic activity or employment that are not captured in the data we used or were significant only in certain parts of the metropolitan areas and counties we analyzed. Finally, our results do not generalize to other locations, hurricanes, or time periods.

To examine the actions taken to improve resilience to future natural disasters in the areas affected by the selected hurricanes, we examined data from FEMA's Hazard Mitigation Grant Program (HMGP), FEMA's PA program, the U.S. Department of Housing and Urban Development's (HUD) Community Development Block Grant Disaster Recovery program (CDBG-DR), and the U.S. Army Corps of Engineers (USACE).

For both objectives, we interviewed federal, state, and local officials, and academic experts. We visited areas affected by each selected hurricane and met with federal, state, and local officials.

We assessed the reliability of the data we used by interviewing agency officials, reviewing relevant documentation, and electronically testing the data. We determined the data were sufficiently reliable for our purposes. Enclosure I provides a more detailed description of our scope and methodology.

We conducted this performance audit from January 2019 to September 2020 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Selected Hurricanes Caused Costly Damages and Challenges for Some Populations; Effects on Overall Economic Activity and Employment Varied Widely

In communities affected by the selected hurricanes, NOAA's damage estimates were $170 billion for Katrina, $74 billion for Sandy, $131 billion for Harvey, and $52 billion for Irma.11 These estimates include the value of damages to residential, commercial, and government or municipal buildings; material assets within the buildings; business interruption; vehicles and boats; offshore energy platforms; public infrastructure; and agricultural assets./12 The selected hurricanes were also costly to the federal government, primarily as a result of federal disaster assistance and subsidized flood insurance payments./13 In 2016, CBO estimated that federal spending exceeded $110 billion in response to Katrina and $53 billion in response to Sandy./14 In 2018, we reported that Congress and the President have also provided federal agencies with at least $120 billion in supplemental appropriations for activities related to natural disasters in 2017, including Hurricanes Harvey and Irma./15

In addition to the costly damages they caused, we found that the selected hurricanes were associated with widely varying effects on overall economic activity in the affected metropolitan areas we analyzed. Our analysis suggests that economic activity was lower than expected the month of the hurricane or in some of the first three months after the hurricane in New Orleans Metairie, Louisiana, after Hurricane Katrina, as well as in Miami-Fort Lauderdale-West Palm Beach, Florida, and Columbia, South Carolina, after Hurricane Irma./16 Within one year, average economic activity in these three metropolitan areas was similar to or greater than what it had been in the year before the hurricane./17 Our analysis suggests that economic activity was not lower than expected in the month of the hurricane or any of the subsequent three months in the remaining affected metropolitan areas we analyzed.

We found that the selected hurricanes were also associated with widely varying effects on total employment in affected counties. Our analysis suggests that in 80 affected counties, total employment was lower than expected the month of the hurricane or in some of the first three months after the hurricane (see table 1). In 47 of these 80 counties, total employment was at least as high as pre-hurricane employment on average within one year, but in the other 33 counties, total employment was lower on average than pre-hurricane employment for at least one year./18 Our analysis suggests that in the remaining affected counties, the selected hurricanes were not associated with total employment that was lower than expected the month of the hurricane or any of the subsequent three months.

Table 1: Number of Affected Counties with Low Total Employment after Selected Hurricanes

See table here: https://www.gao.gov/assets/710/709293.pdf

Finally, we found that the selected hurricanes were not associated with noticeable changes in the distribution of employment across economic sectors./19 Construction employment increased temporarily after each hurricane across all affected counties, but the year after each hurricane, the distribution of employment across economic sectors was similar to that the year before.

In addition to their effects on overall economic activity and total employment, the selected hurricanes created challenges for certain populations within the communities they affected.

State and local government officials told us that the selected hurricanes had significant impacts on communities, local governments, households, and businesses with fewer resources and less expertise, and that challenges faced by households may have impacted local businesses.

* Local officials in Louisiana told us that New Orleans' weak economy after Katrina was in part associated with the decline in the city's population and reduction in the local oil and gas industry, a major employer, before the storm.

* State officials in Texas noted that the size of a community and its overall capacity to provide public services are important factors influencing how quickly it can recover after a disaster. They told us that recovery is faster in larger communities, such as Houston, where local governments have staff who specialize and have experience in disaster management, and that recovery is slower in smaller communities where the local government may not have staff solely dedicated to or specializing in disaster management. They also told us that, because of the size of their tax bases, larger communities have more money to invest in emergency services with significant fixed or upfront costs.

* Local officials in Texas explained that a large part of a household's recovery is either entirely self-funded or partially funded by insurance. Accordingly, local officials in Texas and Louisiana told us that households with low income and wealth may struggle because they have fewer resources to draw on and are less likely to be insured. Local officials in Louisiana noted that seniors, in particular, struggled after Hurricane Katrina because they lacked insurance and the resources to relocate. State officials in Florida indicated that low income households tend to live in older homes and thus were more likely to be displaced after Hurricane Irma. Local officials in Florida and Texas informed us that low income households have more difficulty taking the steps required to obtain federal disaster assistance, such as producing documentation, finding transportation to places to fill out applications, and taking time off to attend meetings or interviews./20

* Local officials in Florida explained that smaller business that have less access to capital struggle to cover operating expenses, including salaries and inventories, while they wait for disaster assistance or the economy to recover. Similarly, local officials in Louisiana said that many small businesses without insurance, capital, or continuity plans prior to Katrina closed because of Katrina./21

* State officials in Florida and local officials in Texas told us that even businesses that rebuilt and reopened their establishments quickly may have had difficulty recovering if their employees were displaced after Hurricane Harvey and Hurricane Irma. Similarly, state officials in Florida indicated that some businesses reopened quickly but had difficulty recovering from Hurricane Irma because their customers had relocated to a different area after the storm.

Affected Communities Are Taking Actions to Improve Resilience but Multiple Factors Can Affect Decision-making and Vulnerabilities Remain

We found that communities affected by the selected hurricanes have been taking actions to improve resilience to future hurricanes and similar natural disasters (hereafter resilience actions). As we have previously reported, resilience actions encompass hazard mitigation-- actions taken to lessen the impact of future disasters./22 We found that affected communities in selected states have been using post-disaster federal financial assistance from FEMA and HUD to implement hazard mitigation projects, as well as recovery projects with a hazard mitigation component (see table 2). Affected communities have also implemented such projects in collaboration with USACE.

Table 2: Recovery and Mitigation Project Amounts in Selected States Associated with Selected Hurricanes

See table here: https://www.gao.gov/assets/710/709293.pdf

Affected communities have been using FEMA HMGP grants primarily to (1) elevate, relocate, flood-proof, retrofit, and add safe rooms to public and private structures in Louisiana after Hurricane Katrina; (2) improve the resilience of critical facilities and infrastructure in New York after Hurricane Sandy; (3) acquire private real property for flood plain management in Texas after Hurricane Harvey; and (4) elevate, relocate, flood-proof, retrofit, and add safe rooms to public and private structures, as well as improve the resilience of critical facilities and infrastructure, in Florida after Hurricane Irma (see table 3 for examples of specific projects).

Table 3: Largest FEMA Hazard Mitigation Grant Program Projects by Estimated Project Amount in Selected States Associated with Selected Hurricanes as of May 2020

See table here: https://www.gao.gov/assets/710/709293.pdf

Affected communities have been using FEMA PA grants primarily to improve resilience of public buildings in all four selected states after selected hurricanes, as well as to improve resilience of roads and bridges, public utilities, and recreational and other facilities in Florida after Hurricane Irma (see table 4 for examples of specific projects).

Table 4: Largest FEMA Public Assistance Program Mitigation Projects by Mitigation Amount in Selected States Associated with Selected Hurricanes as of February 2020

See table here: https://www.gao.gov/assets/710/709293.pdf

Recipients of HUD CDBG-DR grants have allocated most of the funds to (1) compensation and incentive payments to eligible homeowners in Louisiana, after Hurricane Katrina, under the Road Home Homeowner Program,/23 and (2) rehabilitation and reconstruction of residential structures in New York after Hurricane Sandy, Texas after Hurricane Harvey, and Florida after Hurricane Irma.

In collaboration with nonfederal community partners, USACE has rehabilitated and constructed flood barriers (e.g. levees, floodwalls, floodgates) and water diversion projects (e.g. retention basins, waterways enlargements) in Louisiana after Hurricane Katrina, and they are taking similar actions in New York after Hurricane Sandy, Texas after Hurricane Harvey, and Florida after Hurricane Irma. For example, in collaboration with the Coastal Protection and Restoration Authority Board of Louisiana, USACE spent billions of dollars replacing, rebuilding, and raising levees and floodwalls along Lake Pontchartrain bordering New Orleans after Hurricane Katrina.

Community officials we spoke with also reported taking resilience actions outside of federal programs. For instance, state officials in Louisiana and local officials in Texas told us that Louisiana and Harris County, Texas, approved new building code regulations after Hurricanes Katrina and Harvey, respectively. In addition, local officials in Florida constructed a storm water park to collect and store water that could otherwise flood nearby homes during storms.

A community's decision to take a resilience action can depend on the costs and benefits of that action to the community. If the community decides to take the action, then the community incurs the cost of taking the action and forgoes other uses of the funds. Under certain circumstances, federal financial assistance is available to a community to help fund a resilience action and thereby lower its costs to the community.

The community's primary benefit from a resilience action is improved resilience to future natural disasters. For example, a 2019 report by the National Institute of Building Sciences suggests that the benefits of several types of resilience actions may exceed their construction and maintenance costs by protecting lives and property and preventing other losses./24 However, the benefits from a resilience action are uncertain and depend on the likelihood, severity, and location of future disasters. The benefit also depends on the extent to which the community bears the cost of damages from a future disaster. If the community expects that federal assistance will be available post-disaster to cover damages that the resilience action could have prevented, then the expected benefit of that action could be reduced. The community's tolerance for incurring damages can factor into this assessment. Indeed, the resilience action may prevent costly damages from occurring, whereas post-disaster federal assistance does not.

When evaluating the costs and benefits of an action, the standard criterion to decide whether the action can be justified on economic principles is net present value--the discounted monetized value of expected net benefits (i.e., benefits minus costs). Net present value is computed by assigning monetary values to benefits and costs, discounting future benefits and costs using an appropriate discount rate, and subtracting the sum total of discounted costs from the sum total of discounted benefits. An action with a positive net present value is generally preferred, and the sensitivity of the net present value to important sources of uncertainty should be considered. GAO has outlined in its Disaster Resilience Framework how the federal government can contribute information and integrated analysis that enhance a community's understanding of the costs and benefits associated with resilience actions./25

Even as communities are taking resilience actions, state and local officials we spoke with indicated that vulnerabilities remain. Local officials in Louisiana and Texas told us that numerous older homes in their communities remain vulnerable and do not meet current building codes. Texas officials also reported that flood losses covered by insurance are growing in areas outside FEMA's identified base flood areas./26 For example, Harris County Flood Control District in Texas reported that of the 154,170 homes flooded in Harris County during Hurricane Harvey, 68 percent were located outside the FEMA base flood areas./27 Further, in reports to FEMA, states indicate they anticipate that the scope of damages via exposure to weather hazards, such as hurricanes, will likely remain high and could expand across regions affected by the selected hurricanes, and some local governments have projected that population will grow in the regions affected by the selected hurricanes./28 GAO has reported that population growth in hazard prone regions is increasing the nation's vulnerability to losses from natural hazards./29 To help address this vulnerability, GAO has identified key principles for the federal government to facilitate and promote resilience to natural disasters in its Disaster Resilience Framework./30 Agency Comments

We provided drafts of this report to the Department of Commerce, the Department of Defense, the Department of Homeland Security, and the Department of Housing and Urban Development for their review and comment. The National Oceanic and Atmospheric Administration within the Department of Commerce provided technical comments, which we incorporated as appropriate.

The other agencies told us that they had no comments on the draft report.

See footnotes here: https://www.gao.gov/assets/710/709293.pdf

* * *

The text of the GAO report is available at https://www.gao.gov/products/GAO-20-633R

TARGETED NEWS SERVICE (founded 2004) features non-partisan 'edited journalism' news briefs and information for news organizations, public policy groups and individuals; as well as 'gathered' public policy information, including news releases, reports, speeches. For more information contact MYRON STRUCK, editor, [email protected], Springfield, Virginia; 703/304-1897; https://targetednews.com