The federal government’s top-down disasterresponse system is fundamentally flawed. The fed-eral government usually has neither the incentivenor the information needed to effectively coordi-nate relief management. Thus, the best reforms tothe Federal Emergency Management Agencywould take control away from the federal govern-ment, not give it more.

Effective disaster relief efforts have to overcomethe problems of bureaucracy, coordination, andadverse incentives. Nonfederal relief suppliers—particularly those in the private sector—are able toovercome those problems. FEMA—a top-heavybureaucracy that cannot effectively allocate reliefresources and subjects its decisionmakers to all thewrong sorts of incentives—suffers an inherent andunique inability to solve those problems.

In addition, the power to control relief funds

encourages federal policymakers to help ensurereelection by spending that money on key politi-cal districts. States that are politically importantto the president in his reelection bid usually havea significantly higher rate of disaster declaration.States represented on the congressional over-sight committees for FEMA receive significantlymore money for disasters than do states not rep-resented on those committees.

The best reform Congress could undertakewould be to decentralize and depoliticize thetask of disaster relief management by taking thefederal government out of the disaster reliefprocess altogether. Short of that, Congressshould enact reforms that restrict the federalgovernment’s role to only those activities thatenhance the ability of the private sector to moreeffectively respond to disasters.

Flirting with DisasterThe Inherent Problems with FEMA

by Russell S. Sobel and Peter T. Leeson

_____________________________________________________________________________________________________

Russell S. Sobel is the James Clark Coffman Distinguished Chair in Entrepreneurial Studies and Peter T. Leeson isassistant professor of economics at West Virginia University.

Executive Summary

No. 573 July 19, 2006�������

PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1PA Masthead.indd 1 2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM2/9/06 2:08:34 PM

Introduction

Why did the federal government fail somiserably in its response to the devastationof Hurricane Katrina?

Congressional committees—particularlythe Senate Homeland and GovernmentalAffairs Committee, which issued a lengthyreport in May 2006 on the federal response—spent months holding hearings and debatingthat complex and multifaceted question.Through it all, committee members seemedto miss some fundamental truths about thenature of federal disaster relief efforts. As aresult, many of the proposals—particularlyones that suggest the federal governmentshould have even more power, money, andauthority to react to natural disasters—aremisguided. In fact, they would not solve theinherent problems with federal disaster pro-grams; they would compound them.

This study looks at the flaws built into thestructure of the main federal disaster responseagency, the Federal Emergency ManagementAgency. It examines the conclusions of theSenate committee’s report and provides analternative framework that Congress can useto determine which reforms of FEMA are best.

In short, there are three problems inherentin the delivery of federal disaster relief thatneed to be recognized: the problem of bureauc-racy, the problem of coordination, and theproblem of adverse incentives. Another impor-tant fact that must be realized is that naturaldisasters are prone to being politicized. Most ofthe current proposals to reform FEMA fail torecognize those problems. This paper exploreseach of the problems and applies the insightgained to some of the conclusions of theSenate report. It closes by providing usefulguidelines for assessing proposed FEMAreforms.

The Problem of Bureaucracy

The conventional wisdom is that the fail-ure of the federal government to quickly andeffectively respond to the devastation wrought

by Hurricane Katrina was at least partly aresult of an unprepared top-heavy bureaucra-cy.1 As there is to much conventional wisdom,there is a great deal of truth to that.

Disaster relief that is managed by the feder-al government necessarily becomes bureauc-ratized. FEMA was created to oversee andadminister disaster relief. It is in turn overseenby people in other government agencies andmembers of Congress. Following 9/11, forexample, FEMA was placed under the umbrel-la of the Department of Homeland Security,which added a new layer of bureaucracy. Withso many political decisionmakers involved inthe actions of FEMA, it is easy for relief effortsto be slowed or stalled and resources allocatedto less important uses. As Louisiana’s gover-nor Kathleen Blanco complained after Hurri-cane Katrina, “No one, it seems, even those atthe highest level, seems to be able to breakthrough the bureaucracy.”2

That is not necessarily the result of incom-petence or malice on the part of the peopleinvolved in federal disaster relief operations.Nor should it imply that congressional over-sight of disaster relief is more hassle than it isworth. As we discuss later, however, politicaldecisionmakers face incentives that oftenconflict with the public interest. At each levelof bureaucratic action, the key decisionmak-ers involved may face different incentivesthan do the people above or below them inthe chain of command. What is important torealize here is that the result of the multiplelayers of bureaucracy inherent to centralizeddecisionmaking is usually slow and delayedaction.

Thus, it should come as no surprise that thereal success stories in the Katrina relief effortcame from those who flouted the bureaucraticdecisionmaking process and took action with-out explicit approval by FEMA. The U.S. CoastGuard, for example, began its helicopter rescueefforts without waiting for any other govern-ment agency’s approval or coordination. As theSenate Committee on Homeland Security andGovernmental Affairs concluded in its finalreport on its investigation of the federalresponse to Katrina, the Coast Guard was so

2

It should come asno surprise thatthe real success

stories in theKatrina relief

effort came fromthose who flouted

the bureaucraticdecisionmaking

process and tookaction without

explicit approvalby FEMA.

effective because it operated “without signifi-cant bureaucratic hurdles.”3

Further examples abound. Without seek-ing federal permission, a Canadian search-and-rescue team from Vancouver arrived inNew Orleans days before any FEMA-coordi-nated units, leading to slightly inaccurate butamusing media accounts of how the RoyalCanadian Mounted Police beat the U.S. gov-ernment into New Orleans.4 An emergencymanagement team deployed by the state ofFlorida was able to assist local officials inMississippi with law enforcement, search-and-rescue operations, and the delivery offood and water—all the sorts of things thatFEMA is supposed to do—mostly because theteam circumvented FEMA’s bureaucraticapproval process.5

Nongovernmental organizations were metwith resistance by FEMA, too. The Red Cross“begged to be allowed to go [into New Orleans]”to distribute essential relief supplies but was pre-vented by government officials from doing so.6

FEMA confiscated critical emergency supplies,shipped by the hospital’s out-of-state privateowner to assist the hospital’s 137 remainingpatients, while the supplies were in transit toMethodist Hospital in New Orleans. “Thosesupplies were in fact taken from us by FEMA,and we were unable to get them to the hospital,”one hospital representative remarked. To avoidFEMA’s confiscatory actions, the owner sent asecond shipment to Lafayette (130 miles fromNew Orleans) and had a private helicopter fly itdirectly to the rooftop of the hospital in NewOrleans.7

One of the best examples of the problemsbureaucracy creates in a disaster relief situationmight be called “the tale of two sheriffs.” SheriffWarren Evans of Wayne County, Michigan, andSheriff Dennis Randle of Carroll County,Indiana, were both eager to assist the hurricanevictims, and both had the resources necessaryto do so. Sheriff Evans ignored both FEMA andhis governor’s instructions to wait for FEMAapproval and went to New Orleans with ninetruckloads of supplies and 33 deputies to help.8

Sheriff Randle, on the other hand, followedprocedure and was buried under mounds of

FEMA paperwork. He never made it to NewOrleans.9

The shortcomings of the FEMA bureauc-racy are outlined extensively in the Senatereport, and most of that report’s recommen-dations are geared toward making federaldisaster response operations more efficientand streamlined. Those recommendationsare mostly useful. However, remedying theproblems of bureaucracy is only part of thesolution. Even if one assumes away bureau-cratic impediments, federal disaster relief willcontinue to be plagued by two problems thatare practically impossible for federal agenciesto overcome: the problem of coordinationand the problem of adverse incentives.

The Problem ofCoordination

Economic analysis is, at root, the study ofhow to allocate scarce goods among compet-ing uses. The failures of FEMA can beexplained by just that sort of analysis. In thewake of Hurricane Katrina, there were unusedprivate- and public-sector relief providers andfirst responders on the “supply side” of therelief market, many of which were waiting forpermission from FEMA to bring supplies tothe affected areas. On the “demand side” werethe victims of the storm—people in genuineneed who simply weren’t getting relief sup-plies. The puzzle FEMA faced after HurricaneKatrina was how to best use the scant infor-mation it was getting on the need for reliefand the overall situation on the ground tomake decisions about where to direct therelief effort.

More than 60 years ago, Nobel economistFriedrich Hayek discussed the problem of theeffective use of information in markets:

The peculiar character of the problem ofa rational economic order is determinedprecisely by the fact that the knowledgeof the circumstances of which we mustmake use never exists in concentrated orintegrated form, but solely as the dis-

3

Even if oneassumes awaybureaucraticimpediments,federal disasterrelief will continue to beplagued by problems that are practicallyimpossible forfederal agenciesto overcome.

persed bits of incomplete and frequentlycontradictory knowledge which all theseparate individuals possess. The eco-nomic problem of society is thus notmerely a problem of how to allocate“given” resources—if “given” is taken tomean given to a single mind which delib-erately solves the problem set by these“data.” It is rather a problem of how tosecure the best use of resources knownto any of the members of society, forends whose relative importance onlythose individuals know. Or, to put itbriefly, it is a problem of the utilizationof knowledge not given to anyone in itstotality.10

How should that problem be solved? Hayekconcludes:

If we can agree that the economic prob-lem of society is mainly one of rapidadaptation to changes in the particularcircumstances of time and place, itwould seem to follow that the ultimatedecisions must be left to the peoplewho are familiar with these circum-stances, who know directly of the rele-vant changes and of the resourcesimmediately available to meet them.We cannot expect that this problemwill be solved by first communicatingall this knowledge to a central boardwhich, after integrating all knowledge,issues its orders. We must solve it bysome form of decentralization.11

Can a bureaucracy use dispersed informa-tion to coordinate demands of disaster victimswith available supplies and scarce resources?As Hayek points out, that type of coordinationsimply cannot be achieved by channelingdemands and supplies through a centralizedagency. Individuals with local knowledge andthe ability to act on it must be allowed to makethe decisions. Decentralized markets forgoods accomplish this, and in doing so solvewhat economists call the “coordination prob-lem.”

In the wake of a natural disaster, the mostpressing issue is to determine what relief sup-plies are needed, who needs them, and whohas the means to meet those needs. Some dis-aster victims need water, others need shelter,and still others need food. It is likely that alldisaster victims will need basic necessities,but the extent to which different individualsneed those things will vary. Some individualswill be in greater need than others. In addi-tion, there will be many instances of specificareas needing specific help, such as rooftoprescues or massive bus evacuations. No oneperson or agency could ever have access to allof this knowledge or information.

That is certainly the case with FEMA.Examples of FEMA’s misallocation of relieflabor and supplies are legion. For instance,FEMA moved a medical team of 30 peoplecapable of treating hundreds of hurricane vic-tims from Alabama to Mississippi, and then toTexas. For 11 days, medical team members saytheir relief activities were reduced to treatingone small cut. And then FEMA moved themagain—everywhere but where they were neededand could accomplish the most, which was inNew Orleans. As one frustrated medical teammember lamented, “We joined the team tohelp people who need it and we are not help-ing anybody.”12

In addition, the bureaucratic tendencies ofFEMA only compound the inherent coordina-tion problems by diverting critical resources tosuperfluous uses or causing them to sit idleand unused. For example, a private-sectorcompany offered to donate a mobile commu-nications unit that could have provided muchneeded help to relief workers and victims.Instead, it sat in Germany on a chartered pri-vate plane ready to leave for nine days.13

Despite repeated attempts by the company tocontact FEMA and get the required permis-sions to come to New Orleans, the companygot no response and eventually gave up tryingto bring its resources to bear on the reliefeffort.

A similar situation prevailed in the case of1,000 firefighters who volunteered to help therelief effort and believed that their skills

4

Coordinationsimply cannot be achieved by

channelingdemands

and suppliesthrough a central-

ized agency.Individuals withlocal knowledge

and the ability toact on it must beallowed to make

the decisions.

would be put to use actually helping hurri-cane victims. Instead, they were sent to a hotelin Atlanta, forced to take days of sexualharassment courses, and eventually deployedby FEMA only to hand out fliers with FEMA’sphone number on them. As one firefighterastutely observed: “It’s a misallocation ofresources. Completely.”14

FEMA officials are supposed to follow theprinciple of letting those closest to the situationdetermine how best to meet the needs of disas-ter victims. Yet the federal agency often deniedlocal officials the latitude to make crucial deci-sions and allocate relief resources on the basisof the information they had. The LouisianaDepartment of Wildlife and Fisheries requesteda few hundred large rubber rafts from FEMA touse in search-and-rescue missions. But FEMAofficials declined the request because theythought the rubber rafts would not be strongenough to maneuver in debris-filled water.

The Louisiana officials, who knew the sit-uation and the area better than the FEMAofficials, disagreed with the decision whenquestioned by Senate investigators. Theybelieved that “the raft would have been valu-able—particularly in the early days—either formaneuvering in very shallow water neardoors and windows or in saving additionaltrips to collection sites by collecting rescueesin rubber rafts that could be towed behindregular boats.”15

Perhaps the most stunning example ofhow a centralized federal bureaucracy is inher-ently ill-equipped to coordinate the directionof relief resources is what has become knownas the “odyssey of the ice.” FEMA ordered 182million pounds of ice to be delivered to strand-ed families and aid workers. Yet some of the iceended up in Portland, Maine, more than 1,500miles away from the disaster area. The cost ofshipping and storing the 200-plus truckloadsof the Portland-bound ice was $275,000.

As the Senate report noted:

On September 16, NBC News reportedthat it had found trucks full of ice inlocations such as Maryland, Missouri,Georgia, and Tennessee. Some of the

trucks had been driving and/or sittingidle with their full loads for two weeks.One truck driver reported that he hadbegun his trip in Oshkosh, Wisconsin,traveled to Louisiana, then was sent toGeorgia, but was rerouted to SouthCarolina, before being sent to Cumber-land, Maryland. NBC News later report-ed that the truck was then sent to Iowa,where the ice was put into cold storage.The driver reported that this cost tax-payers at least an extra $9,000.16

A truckload of ice even ended up at theReid Park Zoo in Tucson, Arizona. The driverof the ice truck got so many conflicting com-mands from government relief officials thathe ended up traveling through 22 states with-out ever delivering a single bag of ice to a hur-ricane victim. Instead, he ended up donatingit to the Tucson zoo to be enjoyed by thepolar bears.17

If a private firm had misallocated itsresources the way FEMA did, it would havesuffered losses. That’s because the profit andloss mechanism of decentralized marketstells suppliers whether or not they are satisfy-ing the needs of demanders. Suppliers whodo so successfully earn profits and those whodo not suffer losses. Profits and losses com-municate to suppliers whether or not theiractivities are desirable to demanders andwhether or not output should be increased.In the context of relief management, the logicof profits and losses gives private providers ofdisaster relief essentials—such as water, food,and shelter—valuable information aboutwhether or not they are effectively fulfillingthe needs of disaster victims.

Taking a look at the track record of pri-vate-sector suppliers of disaster relief essen-tials in the wake of Katrina serves as an inter-esting counter-example to the centralizedgovernment-driven approach. Wal-Mart, forinstance, was able to quickly bring necessitieslike water to the Katrina victims who hadbeen hit the hardest. While FEMA was scram-bling to respond, Wal-Mart was providingthe items rescue workers and victims needed,

5

The track recordof private-sectorsuppliers of disaster reliefessentials in thewake of Katrinaserves as an interestingcounter-exampleto the centralizedgovernment-driven approach.

in the right quantities, at their everyday lowprices, and sometimes even for free. Wal-Mart was able to overcome the problem ofcoordination.

Wal-Mart has its own fleet of trucks and100 distribution centers across the country—a key part of the innovative supply-chainmodel that has made the company a leader inretail.18 This helped get necessary supplies—including chain saws, boots, sheets, clothes,water, and ice—to the affected areas.19

Wal-Mart also had an incentive to act fastsince failing to do so would result in forgoneprofits to the company. Sixty-six percent of theWal-Mart stores in the Gulf region that hadsustained damage were in operation againwithin 48 hours of Katrina’s making landfall.That number rose to more than 80 percentwithin six days.20 As one hurricane victim putit, Wal-Mart “was the only place we could findwater in those first days. . . . I still haven’t man-aged to get through to FEMA. It’s hard to say,but you get more justice at Wal-Mart.”21 Wal-Mart’s amazing capacity to bring the neededsupplies to the hard-hit areas had even itsstaunchest critics praising the company.22

Private-sector companies even respondedbetter than the federal government in one ofthe areas in which government should have themost expertise and, indeed, the most legitimaterole: maintaining law and order. Consider, forexample, the numerous private security agen-cies that protected the property of residentsand business owners. Those private firmsemerged to satisfy an unmet demand for pro-tection created by government’s failure to per-form this task in the wake of the disaster.Within 14 days of Katrina’s landfall the num-ber of private security firms with a presence inLouisiana climbed from 185 to 235.23 Severaltelecommunications companies hired privatesecurity firms to make sure their employeesand equipment—both resources that state andfederal rescue teams relied on to stay in touchwith one another—were transported safely.24

Why couldn’t FEMA, with the full powerand resources of the federal government at itsdisposal, be as effective as private companies?Federal officials usually have very little idea

about whether or not they should expandtheir activities, alter their activities, or dropthem altogether. They know the costs of theiractivities, but they have no information in theform of real-time feedback about the desir-ability of those actions. That makes allocatingresources and routing them to those whodesire and need them exceedingly difficult, ifnot impossible. Solving the problem of coor-dination requires that both the costs and thebenefits of activities be considered. But a cen-trally controlled relief operation provides lit-tle information about the latter to decision-makers. Thus, a decentralized system forresponding to disasters that kept centralbureaucratic intrusion to a bare minimumwould be preferable to the current system.

The Problem of AdverseIncentives

The incentives individuals face depend onthe institutional context in which they oper-ate. The incentives faced by individuals in thepolitical sphere are different from thosefaced by individuals in the marketplace. Asdiscussed above, in the market actors areguided by the profit and loss motive. To max-imize their return, they must satisfy the con-sumers of their goods or services.

Political actors, in contrast, face fundamen-tally different incentives.25 Understandingthose incentives might help to explain the slowand ineffective response of FEMA generally.

Economists distinguish between two typesof policy mistakes: “type-one” and “type-two”errors. Type-one errors are mistakes that resultfrom not being cautious enough. For example,if the Food and Drug Administration approvesa new drug without sufficient testing and thedrug makes millions of people seriously ill, theFDA has committed a type-one policy error.

Type-two errors, on the other hand, aremistakes that result from being too cautious.If the FDA has an overly burdensome testingrequirement for new drugs, potentially help-ful drugs are prevented from reaching, or atleast delayed in reaching, consumers who

6

The incentivesfaced by

individuals in thepolitical sphere

are different fromthose faced by

individuals in themarketplace.

could have greatly benefited from them. Abranch of economics called public choicetheory informs us that government agencieslike the FDA and FEMA are overly prone tocommit type-two errors.

The reason for that is more straightfor-ward than it might seem. Both type-one andtype-two errors can result in injuries or harmto the public. However, the visibility and pub-lic backlash are likely larger for type-oneerrors. If the FDA releases a drug that resultsin harm to the public, the FDA will receiveheavy criticism for its decision, and the harmdone will be directly blamed on the FDA.Being too cautious, however, also causesharm because people die or suffer needlesslywhile a drug is delayed in the approvalprocess. But in this second case the harm isnot as easily associated with the FDA. It’sharder to pinpoint exactly who died as aresult of FDA inaction. Because the FDAfaces heavier backlash from an instance of,say, 100 deaths clearly caused by the releaseof an unsafe drug than it does from 100deaths that might have been caused by delay-ing the introduction of a new drug, the FDAhas an inherent bias to be too cautious in itsdecisionmaking.26 Likewise, if a disaster isdeclared and FEMA jumps the gun by gettinginvolved immediately, it may commit a type-one error. Because type-one errors are overtmistakes, they are highly visible and aretherefore accompanied by a higher likelihoodof admonishments from citizens, the press,and, possibly, other government agencies.

Suppose, for example, that FEMA allowsrescue workers to enter a disaster zone andthose workers get hurt. FEMA could be blamedfor letting them in prematurely. Thus, bureau-cratic hesitancy has always been an operationalassumption of FEMA. Indeed, as the Senatereport points out, “FEMA has a longstandingpolicy of not putting its emergency respondersin the path of a storm so that they will not be inneed of rescue themselves.”27

Type-two errors, in contrast, are less visi-ble and thus less likely to result in admonish-ment. Or, to put it another way, if an actionresults in admonishment, it is likely to be less

severe than in the case of a type-one error. IfFEMA waits too long to enter a disaster zone,it may be blamed for acting too slowly as itwas in the case of Katrina. But that blame islikely to be less than what FEMA mightreceive if it entered a disaster zone immedi-ately, before a plan was worked out, and con-sequently bungled its relief effort in a moreovert fashion. FEMA, like the FDA, has anincentive to delay action even if more disastervictims are harmed by its not entering thanwould be harmed if it entered prematurely.Victims lost before FEMA enters because itdelays action are less obviously linked toFEMA’s lack of action.

FEMA’s extreme cautiousness in takingaction helps to explain its slow response toKatrina. That slow response was certainly notbecause FEMA was unaware of the potentialfor such a disaster in New Orleans. Accordingto experts at the National Hurricane Center,the danger in New Orleans was known bymany people for years, which gave FEMAplenty of time to devise a plan and work outits execution. For many years local, state, andfederal government had been warned numer-ous times about this very scenario.28

Hurricane experts from the center hadeven run drills of a Katrina-like scenario theyear before in an exercise funded by FEMAitself. FEMA officials who participated in thepresentation of the final study scoffed at theresults, discounted them as impossible, anddragged their feet in acknowledging andpreparing for this eventuality. Even afterFEMA became aware of the certainty thatsuch a strong hurricane would strike NewOrleans, it chose not to pre-deploy theresources clearly identified in the study,which was presented to FEMA officials theyear before the storm hit.29

As one observer described it, the entirerelief process exhibited tremendous “govern-ment hesitancy.”30 Take the case of the leveebreakthrough in New Orleans the day thehurricane hit. Although government agen-cies were aware that the levee system had bro-ken by 6:00 p.m. Monday, officials waiteduntil the next day before sounding the alarm,

7

FEMA has anincentive to delayaction even ifmore disaster victims areharmed by its notentering thanwould be harmedif it entered prematurely.

at which point the city had been flooding fornearly 24 hours.31 Similarly, FEMA did notrequest military assistance for a full day afterKatrina ravaged New Orleans—and whenFEMA finally did request military assistance,it asked for two helicopters to perform fly-overs.32

After disaster struck, government waitedsome more. Walter Maestri, emergency man-agement director of Jefferson Parish, report-ed that federal help of any kind took nearly aweek to arrive. “For approximately six dayswe sat here waiting.”33 And all of this slug-gishness occurred despite the fact that FEMAdirector Michael Brown declared the daybefore Katrina made landfall: “FEMA is notgoing to hesitate at all in this storm. We arenot going to sit back and make this a bureau-cratic process. We’re gonna move fast, we’regonna move quick and we’re gonna do what-ever it takes to help these disaster victims.”34

Now consider the private sector’s response.The relief planning began even before the hur-ricane made landfall near New Orleans. As theWall Street Journal noted: “Home Depot’s ‘warroom’ had transferred high-demand items—generators, flashlights, batteries and lumber—to distribution areas surrounding the strikearea. Phone companies readied mobile celltowers and sent in generators and fuel.Insurers flew in special teams and set up hot-lines to process claims. This planning allowedthe firms to resume serving customers inrecord time. . . . [T]he Business Roundtablehad by August of [2005] arranged for each ofits 160 member companies to designate a dis-aster relief point man. These folks were inplace and ready to help before Katrina madelandfall.”35

Unlike government, for-profit disasterrelief suppliers have no reason to consistent-ly err on the side of making type-two errors.Waiting too long to enter a disaster zonemeans giving away profitable opportunitiesto competitors who get there first. Consideragain the case of Wal-Mart and other private-sector organizations that began preparingfor the hurricane the week before it hit bymoving supplies and trucks into position.

They got aid to the region faster than anylocal, state, or federal government.

The Political Dimensionof Disasters

Another inherent problem with the feder-al government’s disaster relief programs isthat they are often subject to political manip-ulation. When government is in charge ofallocating some share of disaster reliefresources, political actors seeking privateends, such as reelection, face an irresistibleincentive to cater to important geographicconstituencies that are not always those mostin need of assistance. In addition, govern-ment officials in charge of agencies such asFEMA will cater to those who determinetheir budgetary allocations rather than to thecitizens they are supposed to serve. Theincentive of political actors is to help them-selves by distributing money in ways thatbenefit them and their political careers.36

For FEMA assistance to flow, a disastermust first be declared by the president. Aftera disaster has been declared, the allocation ofmoney across geographic areas is at the dis-cretion of FEMA, which is overseen by con-gressional committees. The vast majority ofdisasters declared are for rain, snow, andother mundane weather events.

After examining all disasters from 1991 to1999, a comprehensive study by Garrett andSobel found that states politically importantto the president in his reelection bid have a sig-nificantly higher rate of disaster declaration.37

Recent data confirm the continuation of thispolitical manipulation. In 1996, when BillClinton was up for reelection, he set a recordby declaring the largest number of major dis-asters in history: 75. Unsurprisingly, the sec-ond-highest year for disasters in history was2004, George W. Bush’s reelection year, whenhe declared 68. Ninety percent of the increasein disasters declared between 2003 (a nonelec-tion year) and 2004 were in the 12 battle-ground states where the election was decidedby 5 percent or less.38

8

For every representative astate has on theHouse disasterrelief oversight

committee, itreceives about $30 million in

additional funding when a

disaster isdeclared.

The year with the largest number of disas-ters declared during George H. W. Bush’sadministration was also the year he was up forreelection, and this holds true for RonaldReagan as well. Other striking individualexamples abound, including a two-foot snow-storm in Ohio39 (a state that went for Bush),which netted that state disaster relief duringthe 2004 election year, while Wisconsin40 (astate that went for Kerry) was denied disasterrelief in 2005 in the aftermath of a major tor-nado. House Speaker Dennis Hastert evenbragged about his political influence being asignificant determinant of his state’s beingdeclared a disaster area due to a lack of rain,which hurt agricultural production.41

States represented on the congressionaloversight committees for FEMA—which havesignificant influence over FEMA’s budget—received significantly more money for disas-ters than did states not represented on thecommittees. As Princeton economics profes-sor Alan Krueger wrote in the New York Timesabout the Garrett-Sobel study: “This figuremay seem to overstate the role of politicsbecause representatives from states prone tobe hit by disasters probably seek out seats onFEMA oversight committees. But [the analy-sis] simultaneously adjusts for the amount ofRed Cross assistance and private insurancelosses from disasters each year. So, havingCongressional representation on an oversightcommittee appears to matter even when com-pared with disasters in other states that causeroughly the same amount of damage and suf-fering.”42

For every representative a state has on theHouse disaster relief oversight committee, itreceives about $30 million in additional fund-ing when a disaster is declared. All told, thestudy found that nearly half of all disaster reliefis motivated by politics rather than by need.

The Implications forDisaster Relief Policy

If centralized disaster relief management isinherently prone to failure, the policy implica-

tion of this analysis is straightforward:Disaster relief management should be decen-tralized, which means taking it out of the fed-eral government’s hands altogether. Disasterrelief, like all other activities that entail coordi-nating suppliers and demanders, requires theright kinds of incentives and information tobe effective. Although political processes gen-erate neither of those things, markets generateboth. The question thus becomes how to goabout decentralizing and depoliticizing disas-ter relief management.

The problem with many “privatization”reforms, for instance, is that they do not fullyde-politicize disaster relief management. Forinstance, proposals for disaster relief out-sourcing still leave a substantial decisionmak-ing role for government.43 Although possiblyan improvement over the status quo, this typeof reform only partially and very imperfectlycorrects just one part of the problem. Leavinggovernment at the helm of disaster relief man-agement keeps in place the incentive problemsof centralized disaster relief discussed previ-ously. As long as the federal government hasthe power to dispense disaster relief funds, itsincentive is to do so in a way that maximizespolitical ends instead of dispensing them tothose with genuine need.

Furthermore, adverse incentives may beintroduced by allowing private suppliers tovie for federal disaster relief contracts.Potential suppliers might be selected on thebasis of favoritism.

Disaster relief reforms that only partially“privatize” disaster relief are also likely tocontinue to suffer from the governmentwaste and fraud that have repeatedly plaguedFEMA.44 An investigation by the South FloridaSun-Sentinel, for example, found widespreadfraud in FEMA spending. Looking at only 20of the 313 disasters declared between 1999and 2004, that investigation found that 27percent of the $1.2 billion doled out byFEMA went to areas (or individuals) that suf-fered little or no damage. Examples include$31 million paid to Miami-Dade County res-idents who did not experience hurricane con-ditions and $168.5 million to Detroit resi-

9

Nearly half of alldisaster relief ismotivated by politics ratherthan by need.

dents for a rainstorm in 2000 that the mayorat the time couldn’t even recall.45

A much more effective and consequentlymore appealing form of disaster relief manage-ment reform involves taking government out ofdisaster relief altogether. Hurricane Katrinademonstrated that even in the face of govern-ment-erected barriers private relief efforts areamazingly effective. Totally depoliticizing disas-ter relief also completely eliminates the poten-tial for the political problems, manipulations,and obstacles to genuine aid that centralizeddisaster management necessarily entails.

Unfortunately, the benefits of getting gov-ernment out of disaster relief entirely are pre-cisely the reasons why that option is political-ly the least likely. Politicians and bureaucratswho benefit handsomely from the presence ofFEMA and the ability to declare disasters andcontrol the flow of disaster aid resources willnot let go of their power without a seriousfight.

In light of that political reality, an alterna-tive type of disaster relief reform must beforged. Although government is unlikely torelinquish all control over disaster relief man-agement, the abysmal failure of FEMA afterKatrina might make it politically possible toget government to surrender a large portionof its control and accept a seriously dimin-ished role in providing disaster relief. Oneattractive option in this vein would reducegovernment’s role exclusively to (1) openingchannels of trade so that private aid supplierscan reach those in need, by repairing trans-portation infrastructure for instance, and (2)protecting the property of suppliers and dis-aster victims, so that suppliers will be securewhen entering a disaster zone.

A cornerstone of any reform must also beeliminating FEMA’s ability to forcibly preventother relief suppliers from entering disasterzones. It’s possible to argue that more harmwas done by FEMA’s keeping other suppliersout than by its own bungled relief efforts.After Katrina struck, when items such as bot-tled water were needed most, government’sresponse was to erect barriers that kept privateand nonfederal relief workers and aid out of

the area. Private suppliers should be allowed tomake their own decisions about the risks ofentering disaster areas.

Regardless of what specific approach istaken to reforming disaster relief management,two things are clear. Reform is necessary andgovernment’s role in any proposed changemust get smaller, not larger. Unfortunately, theMay 2006 report of the Senate Committee onHomeland Security and Governmental Affairsrecommends exactly the opposite strategy. Thereport advocates abolishing FEMA, but insteadof reducing federal responsibilities for disasterrelief, it advocates replacing FEMA with an evenlarger, more centralized and bureaucratic disas-ter relief management agency: the NationalPreparedness and Response Authority.

The inherent problems of governmentdisaster relief that plague FEMA will only bemagnified if it is replaced with a more expan-sive government agency. If there have beencoordination and incentive problems underFEMA owing to its centralized organization,it is only reasonable to expect those problemsto be exacerbated by greater centralization.The proposal that an NPRA be substitutedfor FEMA does not constitute disaster reliefmanagement reform; it is just more of thesame in ever larger doses.

Notes1. The failure of the federal relief bureaucracy isdiscussed at length in Senate Committee onHomeland Security and Governmental Affairs,“Hurricane Katrina: A Nation Still Unprepared,”May 2006, http://hsgac.senate.gov/, hereafterreferred to as the Senate report.

2. Quoted in Agence France Press, “Bush TakesBlame for Government’s Storm Failures as DeathToll Jumps,” September 14, 2005.

3. Senate report, chap. 23, p. 12.

4. Sonia Joshi, “Canadian Search-and-RescueTeam Reached a Flooded New Orleans Suburb toHelp Save Trapped Residents Five Days before theU.S. Military,” India Daily, September 8, 2005,http://www.indiadaily.com/editorial/4532.asp.

5. Senate report, chap. 23, p. 12.

10

Reform is necessary and

government’s rolemust get smaller,

not larger.

6. Larry Eichel, “What Went Wrong,” Knight-Ridder special report, September 11, 2005.

7. Bob Herbert, “Sick and Abandoned,” New YorkTimes, September 15, 2005, p. 31.

8. Kathleen Parker, “Three Heroes OutwittedBureaucracy,” New Hampshire Union Leader, Septem-ber 14, 2005.

9. NBC News, “What Went Wrong in HurricaneCrisis,” interview transcript, Dateline NBC,September 9, 2005.

10. F. A. Hayek, “The Use of Knowledge in Society,”American Economic Review 35, no. 4 (September1945): 520.

11. Ibid., p. 524.

12. Quoted in Lisa Myers and the NBCInvestigative Unit, “Relief Chaos in Katrina’sWake,” msnbc.com, September 8, 2005.

13. Ibid.

14. Quoted in Lisa Rosetta, “Frustrated: FireCrews to Hand Out Fliers for FEMA,” Salt LakeCity Tribune, September 12, 2005, p. A1.

15. Senate report, chap. 21, p. 9.

16. Senate report, chap. 28, p. 5.

17. Ibid.

18. Senate report, chap. 23, p. 15.

19. John Tierney, “Let Wal-Mart Take Over Emer-gency Management,” New York Times, September21, 2005.

20. Jason Jackson, director of business continuity,Wal-Mart Stores, Inc., Testimony before the SenateCommittee on Homeland Security and Governmen-tal Affairs, 109th Cong., 1st sess., November 16, 2005.

21. Ibid.

22. Sean Higgins, “Wal-Mart Is Lauded for FastRelief Aid to Katrina Victims,” Investor’s BusinessDaily, September 9, 2005.

23. Jeremy Scahill, “Blackwater Down,” Nation,October 10, 2005.

24. Senate report, chap. 18, p. 4.

25. James Buchanan and Gordon Tullock, TheCalculus of Consent: Logical Foundations of Consti-tutional Democracy (Ann Arbor: University of

Michigan Press, 1962).

26. For evidence of this bias within the FDA, see DaleH. Gieringer, “The Safety and Efficacy of New DrugApproval,” Cato Journal 5, no. 1 (Spring–Summer1985): 177–201; and Sam Peltzman, “The Benefitsand Costs of New Drug Regulation,” in RegulatingNew Drugs, ed. Richard L. Landau (Chicago: Univer-sity of Chicago Press, 1973), pp. 114–211.

27. Senate report, chap. 12, p. 18.

28. Senate report, “Findings,” p. 4.

29. Ibid.

30. Eichel.

31. Ibid.

32. Ibid.

33. NBC News.

34. Ibid.

35. “Private FEMA,” editorial, Wall Street Journal,September 8, 2005.

36. Thomas A. Garrett and Russell S. Sobel, “ThePolitical Economy of FEMA Disaster Payments,”Economic Inquiry 41, no. 3 (July 2003): 496–509.

37. Ibid.

38. Alan Krueger, “At FEMA, Disasters and PoliticsGo Hand in Hand,” New York Times, September 15,2005, p. C2.

39. Ibid.

40. “FEMA Hits Wisconsin,” editorial, CapitalTimes, September 27, 2005, p. 8A.

41. Ibid.

42. Krueger.

43. Daniel Henninger, “Bureaucratic Failure,” edi-torial page, Wall Street Journal, online edition,September 2, 2005.

44. Sally Kestin, “Katrina’s Aftermath: FEMABattered by Waste, Fraud,” Los Angeles Times,September 18, 2005, p. 32; and G. Robert Hillman,“Oversight of FEMA Aid Draws Concern,” DallasMorning News, September 13, 2005.

45. Megan O’Matz et al., “Disaster AssistanceTangled Up with Politics—At a Cost to Taxpayers,”South Florida Sun-Sentinel, September 18, 2005.

11

OTHER STUDIES IN THE POLICY ANALYSIS SERIES

572. Vertical Integration and the Restructuring of the U.S. Electricity Industryby Robert J. Michaels (July 13, 2006)

571. Reappraising Nuclear Security Strategy by Rensselaer Lee (June 14, 2006)

570. The Federal Marriage Amendment: Unnecessary, Anti-Federalist, and Anti-Democratic by Dale Carpenter (June 1, 2006)

569. Health Savings Accounts: Do the Critics Have a Point? by Michael F. Cannon (May 30, 2006)

568. A Seismic Shift: How Canada’s Supreme Court Sparked a Patients’ Rights Revolution by Jacques Chaoulli (May 8, 2006)

567. Amateur-to-Amateur: The Rise of a New Creative Culture by F. Gregory Lastowka and Dan Hunter (April 26, 2006)

566. Two Normal Countries: Rethinking the U.S.-Japan Strategic Relationship by Christopher Preble (April 18, 2006)

565. Individual Mandates for Health Insurance: Slippery Slope to National Health Care by Michael Tanner (April 5, 2006)

564. Circumventing Competition: The Perverse Consequences of the Digital Millennium Copyright Act by Timothy B. Lee (March 21, 2006)

PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2PA Masthead.indd 2 2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM2/9/06 2:08:35 PM