Using Organizations: the Case of FEMA
Published on: Jun 11, 2006

Charles Perrow is professor emeritus of sociology at Yale University. His current interests are in managing highly interactive, tightly-coupled-systems (including hospitals, nuclear plants, chemical plants, power grids, aviation, the space program, and intelligent transportation systems). These interests grew out of his work on "normal accidents," with its emphasis upon organizational design and systems theory. An organizational theorist, he is the author of a number of award winning books in the field of sociology—including Complex Organizations: A Critical Essay (1972; 3rd ed., 1986), Normal Accidents: Living with High Risk Technologies (1984, revised, 1999), and The AIDS Disaster: The Failure of Organizations in New York and the Nation (1990).


Organizations are tools; their masters need not use them for their nominal ends. The focus of FEMA under President Clinton was natural disaster emergency relief, and preparedness. Under the Bush administration the focus was shifted to combating terrorism, and disaster relief capabilities decayed. That left us unprepared for Hurricanes Katrina and Rita. This unpreparedness led to the massive organizational failures we have been treated to by a shocked media.

Following Katrina, for days air conditioned trucks with no supplies drove aimlessly past “refugees” who were without water or food or protection from the sun. Reporters came and went, but food and water and medical supplies did not (Staff 2005b). The Red Cross was not allowed to deliver goods because it might discourage evacuation (American 2005). Evacuation by air was slowed to a crawl because FEMA said that post 9/11 security procedures required a (prolonged) search for more than 50 federal air marshals to ride the airplanes, and to find security screeners. At the gates, inadequate electric power for the detectors held things up until officials relented and allowed time consuming searches by hand of desperate and exhausted people (Block et al. 2005). Their only food, emergency rations in metal cans, was confiscated because it was thought that cans might contain explosives (Bradshaw and Slonsky 2005). Volunteer physicians watched helplessly; FEMA did not allow them to help because they had not been licensed in the state (Tierney 2005). Without functioning fax machines to send the required request forms, FEMA would not send help that local officials begged for. Perhaps a fifth of the New Orleans police force simply quit, exhausted and discouraged, under fire from looters, or themselves looting. A large National Guard force hid behind locked doors in the convention center, saying they were unprepared to help. A Navy ship idled off-shore, waiting for days to be called. Almost five days after Rita struck, at least one severely damaged Texas town remained without any outside help, out of power, water and food, with an alerted TV camera crew being the first to arrive. And so on.

Did these failures reflect what have been called “prosaic” organizational failures such as all organizations are likely to have in times of stress? Were the organizations simply overcome by an unprecedented challenge? Or had the resources for meeting natural disasters decayed or been diverted towards terrorist disasters? If it was the latter, decay and diversion, we will have to explain why relief organizations other than FEMA also failed.

The failures involved government agencies and the military at all levels, not just FEMA. But FEMA was the organization most responsible for disaster response. What happened to it? It seemed to have performed reasonably well the previous year when four hurricanes struck Florida (though there were charges of gross mismanagement in the dispersal of funds). A review of its history is not encouraging, and will offer some possible explanations for its failures in 2005.

FEMA’s rocky history

It got off to a modest but fairly good start when it was founded by President Jimmy Carter in 1979, one of his last attempts to restructure the federal government. But the bungled Iranian hostage crisis drove him from office, and put Ronald Reagan in. When it was first formed by Carter the agency had two goals. The main one was disaster relief, prevention, and mitigation. A secondary one was coping with a nuclear attack and, vaguely, national security, something normally in the hands of other agencies. Under Reagan the first goal was neglected and starved of resources, while the secondary ones flourished. FEMA set up a “Civil Security Division” with a training center for over 1,000 civilian police to handle riots and political disturbances (not disaster relief). A file was gathered on U.S. left-wing activists, and internment camps were planned. One national training exercise envisioned incarcerating 100,000 “national security threats” (Churchill and Wall 2002; Reynolds 1990; Ward et al. 2000). A top secret National Security Directive (NSDD 26) that Reagan issued in 1982 effectively linked FEMA with the military and the National Security Council (NSC). In FEMA, a small division, the National Preparedness Directorate (NPD), was charged with developing a classified computer and telecommunications network to insure the continuity of the government in the event of a nuclear attack. The network was developed by the National Security Council and subsumed within the broader Department of Defense (DOD) national defense information network. Though originated by FEMA, and drawing upon more and more of FEMA’s budget, FEMA’s disaster relief personnel could not have access to the network. It was “top secret”; only the DOD and the NSC could access it. Congress could not examine the activities or budget of the Civil Defense part of FEMA (Ward et al. 2000). As a result, “FEMA developed one of the most advanced network systems for disaster response in the world, yet none of it was available for use in dealing with civilian natural disasters or emergency management” (1023; see also Churchill and Wall 2002; Reynolds 1990)

The FBI was jealous and alarmed, and so was the Justice Department. The head of FEMA, Louis Giuffrida, was forced to resign when the Justice brought suit in 1985 over cronyism in the agency’s contract awards and a lavish bachelor pad for Giuffrida in Manhattan using FEMA funds. His collaborators, Lt. Col. Oliver North and the equally controversial General Richard Secord, had already left the agency. But the organization continued to ignore natural disasters, and when disasters came, the personnel were poorly trained and funded, and quite possibly inept. Hurricane Hugo in 1989 prompted U.S. Senator “Fritz” Hollings to declare that FEMA was “the sorriest bunch of bureaucratic jackasses I’ve ever known” (1024). The next year when disasters hit California, Representative Norman Y. Mineta of California declared that FEMA “could screw up a two-car parade.” When Hurricane Andrew hit in 1992, the primitive communications system of the agency forced it to buy Radio Shack walkie-talkies in last minute preparations, while the state-of-the-art one FEMA had paid for remained unavailable. President Bush had to call in federal troops and move the FEMA director aside. If this sounds familiar to those who watched the Katrina disaster, recall that the agency had been hijacked by those preoccupied with nuclear defense and domestic radicals. Its failure helped William Clinton push the first President Bush aside.

It recovered remarkably well under the leadership of James Lee Witt, an experienced disaster manager appointed by President Clinton in 1993, and performed as well as we might expect any agency to perform. It not only handled emergency relief well, but set up far-seeing programs to minimize damage from future disasters by, for example, buying up vulnerable land to prevent the establishment of settlements—the “mitigation” program. Employees performed well and shared the goals of the organizational masters. It had a minimum of political appointments.

FEMA under a Bush

FEMA swerved abruptly to the right again under President G. W. Bush, emphasizing privatization of disaster response and counter-terrorism rather than natural disasters. FEMA’s Project Impact was a model mitigation program created by the Clinton administration: it moved people out of dangerous areas and retrofitted structures (Elliston 2004). For example, when the Nisqually earthquake struck the Puget Sound area in 2001, homes that had been retrofitted for earthquakes and schools with FEMA funds were protected from high-impact structural hazards. The day of that quake was also the day that the new president, G. W. Bush, chose to announce that Project Impact would be discontinued (Holdeman 2005). Funds for mitigation were cut in half, and those for Louisiana were rejected. Disaster management was being privatized, with the person who was to be promoted to head the agency, Michael Brown, saying at a conference in 2001, “The general idea—that the business of government is not to provide services, but to make sure that they are provided—seems self-evident to me” (Elliston 2004). The administration tried to cut federal contribution for large-scale natural disaster expenditures from 75 percent to 50 percent, but Congress balked.

Worse still, when a Department of Homeland Security (DHS) was forced upon President Bush by Senator Joseph Lieberman and other Democrats, FEMA lost the cabinet status President Clinton had given it and was folded into the new department. The Government Accountability Office (GAO), Congresspeople, the Brookings Institution and others warned that this could hobble the agency’s natural disaster programs, and it did. Top personnel left (some to the companies enriched by the privatization of emergency relief and preparedness); a union survey of 84 union personnel found 80 percent saying it was a “poorer agency,” 60 percent said they would leave if they could get the same salary in another agency; and the GAO rated its morale as one of the lowest of any government agency (Ellison, 2004). While funds for the agency have actually increased somewhat in the last two years, those for disasters have shrunk while expenditures for counterterrorism have soared. FEMA has lost control of the federal preparedness grants to local and state governments. Those are distributed by a separate office and as a result, three out of every four grants are now spent on counterterrorism. (Much of the money spent on counterterrorism goes to corporations and private businesses; natural disaster money is more likely to be spent on training first responders, hardly a corporate feeding place.) This has been a major blow to states such as Louisiana that are prone to weather disasters.

FEMA, it is charged, not only shifted from natural disasters to counterterrorism, but to political favoritism, another example of using organizations, and it had consequences. Representative Bennie Thompson of Mississippi, hard hit by Katrina, said that during the Bush administration, “FEMA went back to being treated like a political resting place for favors that were owed,” and called for the resignation of FEMA head Michael Brown. Brown was brought into the agency in 2001 by his college roommate, Joe M. Allbaugh, who had run Mr. Bush’s first presidential campaign. Even his small claim to have disaster experience turned out to be fabricated. He said on a Thursday evening TV appearance, three days after Katrina struck, that he had just learned of the plight of thousands stranded at the convention center in New Orleans without food or water. They had been there since Monday, but that Thursday Mr. Brown told an incredulous TV interviewer, Paula Zahn, “Paula, the federal government did not even know about the convention center people until today” (Lipton and Shane 2005).

It also did not know where the ice was. Ninety-one thousand tons of ice cubes, intended to cool food, medicine, and victims in over 100-degree heat, were hauled across the nation, even to Maine, by 4,000 trucks, costing taxpayers over $100 million. Most of it was never delivered. In an age of sophisticated tracking (Federal Express, DHL, Wal-Mart, etc.) FEMA’s system broke down. Asked about the vital ice, Mr. Brown invoked privatization, and told a House panel, “I don't think that’s a federal government responsibility to provide ice to keep my hamburger meat in my freezer or refrigerator fresh” (Shane and Lipton 2005). The ice was not needed for his refrigerator but to keep drugs and medicine fresh, to treat people with heat exhaustion, and to keep the sick, old and frail cool.

Some explanations of recent failures

FEMA was not the only organization to fail so massively, but it, and its parent organization, the Department of Homeland Security under Michael Chertoff, was certainly a key one. Can we attribute this to the evisceration of FEMA under the Bush administration? Did its enfeeblement also enfeeble the response of the National Guards, the military when it was called in, and local and state agencies? At the present writing (October, 2005) it is not clear and much more research is needed to understand the response to Katrina and Rita. For we have three observations, and the lessons from them remain to be investigated: the response to four hurricanes in Florida in the previous year; the response to Katrina; and the response to Rita.

It is possible that FEMA was not deteriorating, but just overwhelmed by Katrina, and recovered somewhat under Rita. The response to Rita has been declared much better by some news stories (Hsu and Hendrix 2005; Block et al. 2005), and almost as bad by others (Staff 2005b). Rita should have been easier. It was less destructive; citizens were more likely to evacuate early based on the experience with Katrina; major cities were not hit; top FEMA officials would be unlikely to again be unable to alert the President; and state guards and the military were already mobilized. Here are four possible interpretations of the varying responses to the Florida hurricanes, Katrina, and Rita.

  1. FEMA’s natural disaster potential deteriorated steadily through 2005, as it was used for other purposes, but this was not noticed in 2004. The 2004 hurricanes were not as serious as those in 2005, and we did not get as many news stories about failures in 2004. (A close investigation of the Florida responses would be needed to judge the importance of this explanation.)

  2. The response to the Florida hurricanes was good, despite the deterioration, because Florida was a politically key state for the administration. Louisiana was not, and Texas was already in Republican hands. Therefore FEMA officials paid more attention to Florida. (FEMA approved payments in excess of $31 million to Florida residents who were unaffected by the 2004 hurricanes, for example (Leopold 2005; Staff 2005a). Research has shown that presidents designate areas as eligible for disaster relief, and give out much greater assistance, when these areas are politically important for them. Political scientists have found that nearly half of all disaster relief is motivated politically rather than by need (Garrett and Sobel 2002). The fact that President Bush had yet to establish a plan for housing evacuees, or a commission to oversee the rebuilding of New Orleans and other coastal cities in three states a month and a half after the hurricane suggests a lack of political incentive.

  3. Katrina and Rita (“KatRita”) were so much more powerful and damaging that even a well-performing FEMA would have been overwhelmed. This explanation does not assume deterioration on the part of the agency’s ability to deal with natural disasters. It assumes a tipping point, and when disasters are involved, the tipping point may bring about a sudden, rather than gradual, decline. Once it is challenged beyond its capabilities, the failures can be sudden and widespread even if the organization is not weak.

    This explanation is persuasive. But the problem is that the failures of FEMA in KatRita at all levels seem so enormous and widespread that it is hard to imagine that common sense and obvious responses would evaporate so widely. Disaster agencies have to be flexible and innovative, even if the challenge is overwhelming. This one appeared to revert frequently to rote training and inappropriate rules.

  4. The final explanation offered is that undoing an agency that had been performing well takes time, but FEMA’s existing deterioration was speeded up greatly when faced with an unprecedented task. While the previous interpretation has obvious merit, this explanation holds that substantial undermining of the agency had already taken place, and KatRita exposed these more fully than the Florida hurricanes of 2004 could have.

The examples given at the beginning of this article, and there are many more, seem to go well beyond “prosaic” failures, and even an “overwhelming” explanation, as in alternative 3. They do not involve panic, enormous overload, unfamiliar tasks or settings, which would accompany failures in unprecedented events. They involved going by the rules. Rather than being flexible and innovative, even when the challenge is overwhelming, these personnel appeared to revert to rote training, insistence upon following inappropriate rules, and an unusual fear of acting without official permission. This could be the result of the agencies’ downgrading of the importance of responding to natural disasters, replacing or losing skilled personnel, and diverting funds to the commercially and politically more attractive alternative of buying equipment such as chemical detection devices, bio-hazard suits, and perimeter surveillance devices, and paying for industrial and port upgrades that have little to do with terrorist threats.

Organizational dynamics could also be at work. I suggest that as the top ranks of the agency lost experienced personnel with high morale and commitment, and were replaced by political appointments, the next level would gradually lose confidence in their superiors, and their morale would slacken. I know of no statistics regarding FEMA, but nationally the Bush administration has increased the number of political appointees for government agencies by 15 percent since 2000 (Writers 2005a). (In President Clinton’s second term, the percentage of political appointments declined.) FEMA has always had many political appointees; most agencies do. But if they increased by 15 percent it would have an impact.

In time, the low morale of upper managers who were not political appointments would spread to lower management, and then to employees in general. In an organization with low morale it may be that sticking to the rules to protect your career is better than breaking them even if the rules are inappropriate. This defensive posture might spread to allied agencies, such as the Transportation Security Administration (TSA), which is already less concerned with safe transit than with terrorists’ potential to use transportation as a weapon. A hypothetical situation could prompt these questions: Is the TSA official in charge of the security of a local airport very likely to tell his employees to stop doing their principal job and just let the evacuees through? Not if he knows that FEMA officials are not sending water and food to the airport because airport staff cannot send the proper requisitions because the faxes are out. The message may be that in perilous times it is best to go by the book. (While not unreasonable, this possibility is not substantiated by research, as far as I know). This is a different explanation than “they panicked,” or “the storm was so large and the task so unprecedented.”

A further consideration is that the reorganization of FEMA into the Department of Homeland Security imposed a top-down, command-and-control model on an agency that most experts say should maximize the power of those at the bottom, Maximizing the ability of the lowest level to extemporize and innovate will minimize the bureaucratic responses that so characterized FEMA. A frequent criticism of FEMA was that the centralized DHS model, and the removal of authority for preparedness to other parts of DHS, would inhibit its responsiveness to unique events (Glenn 2005).

We are left with at least two interpretations. One is that FEMA was not hurt by incorporation into the Department of Homeland Security. It performed well in 2004, faced an unprecedented task with Katrina and could be expected to fail, but recovered and performed reasonably well in Rita, which was less devastating than Katrina but more damaging than the Florida hurricanes.

A second is that FEMA was progressively deteriorating; the deterioration was not picked up by the press in 2004, but was evident when the Katrina challenge was greater, and did only marginally better with the lesser challenge of Rita and the advantages of very recent experience, more credible warnings, and mobilized relief forces.

Each disaster is unique, and routines, such as pre-positioning and ordering ice ahead of time, certainly help. These appear to have been inadequate in KatRita. More important, the ability to scramble, extemporize, and innovate seems to have degraded. (Privatization fans have a point that some of the most creative responses came from private business, but this may reflect the state of FEMA rather than a public/private comparison (Harris 2005). It is possible that this was the most important failing. If so, it may be attributed to the use of the organization for purposes other than those for which it was designed. It may have been used to reward political friends and loyalists, to further an image of being “tough on terror” for political image reasons, and to make expenditures that favored private enterprises and political constituencies rather than training and first responders.


1 This is a section of a manuscript in preparation, “Disasters Evermore? Reducing U.S. Vulnerabilities to Natural, Industrial, and Terrorist Disasters.” Thanks to Lee Clarke for many suggestions and corrections. For other sections of the manuscript see “The Department of Homeland Security, our second great disaster after 9/11,” forthcoming in Homeland Security Affairs, and “Disasters evermore? Reducing U.S. vulnerabilities,” forthcoming, Handbook of Disaster Research, E. Quarantelli and R. Dynes, eds., Springer, 2006.


American Red Cross. 2005. “Hurricane Katrina: Why is the Red Cross Not in New Orleans?”,1096,0_682_4524,00.html.

Block, Robert, et al. 2005. “Behind Poor Katrina Response, a Long Chain of Weak Links.” Wall Street Journal, September 6, p. 1.

Bradshaw, Larry, and Lorrie Beth Slonsky. 2005. “Hurricane Katrina – Our Experiences.” EMS Network News.

Churchill, Ward, and Jim Vander Wall. 2002. The COINTELPRO Papers: Documents from the FBI’s Secret War against Dissent in the U.S. Boston, MA: South End Press.

Elliston, Jon. 2004. “Disaster in the Making.” Independent Weekly. September 22.

Garrett, Thomas A., and Russell S. Sobel. 2002. “The Political Economy of FEMA Disaster Payments.” Federal Reserve Bank of St. Louis. St. Louis, MO.

Glenn, David. 2005. “Disaster Sociologists Study What Went Wrong in the Response to the Hurricanes, but Will Policy Makers Listen?” Chronicle of Higher Education (September 29).

Harris, Gardiner. 2005. “Storm and Crisis: Early Reaction.” New York Times. September 11, p. 31.

Holdeman, Eric. 2005. “Disasters Keep Coming but FEMA Phased Out.” New York Times. August 31.

Hsu, Spencer S., and Steve Hendrix. 2005. “Hurricanes Katrina and Rita Were Like Night and Day.” Washington Post. September 25.

Leopold, Jason. 2005. “FEMA Chief Brown Paid Millions in False Claims to Help Bush Win Florida Votes.” The Free Press. September 15.

Lipton, Eric, and Scott Shane. 2005. “Leader of Federal Effort Feels the Heat.” New York Times. September 3.

Reynolds, Diana. 1990. “FEMA and the NSC: the Rise of the National Security State.” Covert Action Information Bulletin, p. 33.

Shane, Scott, and Eric Lipton. 2005. “Stumbling Storm-aid Efforts Put Tons of Ice on Trips to Nowhere.” New York Times. October 2.

Staff. 2005a. “FEMA: a Legacy of Waste.” South Florida Sun-Sentinel. September 18, p. 19.,0,7651043.storygallery?coll=sfla-home-headlines.

——. 2005b. “‘Hope is Fading’ at New Orleans Convention Center.” NBC, MSNBC.

Tierney, John. 2005. “Going (Down) by the Book.” New York Times.

Ward, Robert, Gary L. Wamsley, Aaron D. Schroeder, and David B. Robins. 2000. “Network Organizational Development in the Public Sector: a Case Study of the Federal Emergency Management Administration (FEMA).” Journal of the American Society for Information Science 51: 1018–32.

Staff. 2005a. “Bush Cronyism Weakens Government Agencies.” Bloomberg. September 30.

——. 2005b. “Katrina Redux? Beaumont Paper Finds Federal Storm Failures in Texas.” Editor & Publisher. September 25.