4 The Disaster after 9/11
The Department of Homeland Security and a New FEMA

If fema represents a promising but flawed attempt to involve the government in natural disasters, the 9/11 crisis presented a challenge to our government that went way beyond those that natural disasters had presented for decades. It also presents a challenge to the thesis of this book.¹ The World Trade towers were large targets that could not be reduced; depopulating New York City is not an option. But terrorists could choose to attack any number of other targets that we are now trying to protect, and these targets could be reduced in size. Regardless of that, this chapter will demonstrate the manifold failures of the agency to protect, mitigate, and even recover from terrorist attacks and other disasters. These failures make the case for vulnerability reduction all the more urgent.

The 9/11 crisis presented a challenge to our government that went way beyond any challenges that natural disasters had presented for decades. The terrorist attack suggested possible future ones that would involve more of society than a natural disaster might, and we had few institutional structures to cope with it. Something far beyond the recombination of disaster agencies that produced FEMA was needed. What we got was a very expensive and slight increase in protection from terrorist attacks, and a marked reduction in protection from industrial and especially natural disasters.

The first thing required was a change in the mental model that our top officials in the White House were using to address terrorist threats to the nation. Years of preoccupation with state-sponsored threats would be hard to set aside, buttressed as they were by many institutional arrangements in the rest of the nation. Second, we needed a new institutional capacity for dealing with the new terrorist threat. The question here was whether this capacity should reside primarily in the executive branch (White House) or in the legislative branch (Congress). Without a convincing change in its threat model, the White House, I will argue, was unable to mobilize support for allowing it to control the effort, and control passed to Congress. But such a huge project affected so many interests that in congressional hands, the project was bound to be unwieldy. Furthermore, the institutional framework chosen for protecting homeland security followed a cultural script that organizational designers most easily revert to, namely, centralized control, even though the problem would be more amenable to the empowerment of diverse, decentralized units with central coordination rather than central control. The many organizational difficulties we will examine that flowed from the central control format, coupled with distracting wars and lack of White House urgency, make the failure of the reorganization inevitable.

Finally, the same format was used to attempt a restructuring of the intelligence agencies and once again, powerful interests, this time in the Pentagon as well as Congress, appear to have thwarted this effort as well. Compared with the restructuring response to the 9/11 tragedy, the initial creation and shaky evolution of FEMA, for all its problems, looks like a piece of cake.

Terrorism, of the Al Qaeda variety, has special interest for those who would protect our nation by reducing its vulnerabilities. Unlike natural and industrial threats, terrorism will intentionally seek out and strike at concentrations of populations, concentrations of hazardous materials, and concentrated nodes in our critical infrastructure. I doubt that we will ever eliminate the threat of terrorism, so reducing those concentrations seems at least as important as catching terrorists (since there are replacement terrorists standing in line) or locking the many doors to our way of life.

TERRORIST THREATS COMPARED WITH NATURAL AND INDUSTRIAL DISASTERS

Deliberate disasters can be directed toward our key vulnerabilities. In contrast to nature and industrial accidents, terrorists are a strategic enemy, looking for our weakest and most consequential targets. There is evidence that Al Qaeda has considered targeting nuclear power plants and chemical plants, national monuments with concentrated populations about (the White House, prominent buildings), key government agencies such as the CIA headquarters, prominent infrastructure institutions such as financial institutions, and the heart of our communications system, the Internet.

The scope of a terrorist attack could exceed that of most natural disasters in several respects. It may not be confined to a region or a single floodplain or earthquake zone. An earthquake could destroy the Los Angeles airport, but airports elsewhere are not thereby threatened. A terrorist attack on the airport would prompt us to drastically curtail operations at all large airports. (A colleague, Jake Shapiro, imagines terrorists at three major airports going through the terminal with a small rag soaked in an explosive the detectors are set to detect, such as fertilizer, and brushing up against suitcases, briefcases, and coats.) Except for epidemics, natural disasters are limited to local areas; they are not nationwide, whereas terrorism potentially is. The scope is also greater because of the psychological impact; we fear, but generally understand, the industrial or natural disaster, but the terrorist attack is almost beyond our comprehension, increasing the level of tension.

The terrorist threat is not confined to a particular area of our social life, such as our defense policy, or the agencies that deal with unemployment, transportation, or energy. These are reasonably circumscribed with limited constituencies, so the Departments of Defense, Labor, Transportation, and Energy are agencies that are manageable. The threat involves many more agencies than a natural disaster might, entailing more jurisdictional disputes in Congress than FEMA ever did.

It is true that natural disasters involve many segments of life: transportation, health, buildings and homes, industry, local government, public order, and so on. But to all these, preventing terrorism adds the detection element—information demands that go far beyond the weather forecast—and the deception element, wherein the threat is disguised. We need much more surveillance with terrorism to detect plots, and because of deception, much more alertness, training, and preparation for a variety of scenarios. As elaborate as FEMA’s communication system was, as we saw in chapter 3, its system would be no match for the challenge of terrorism. Even our intelligence agencies, which are much better equipped for communication than our disaster response agencies, do not appear to be able to interpret and share information on terrorists, at least judging from their performance in connection with the 9/11 attack. Finally, the nation can emotionally come to terms with natural disasters as “acts of God,” mistaken as that is, and regard industrial disasters as inevitable but infrequent “errors.” But the willful, intended destruction of noncombatants may bring about a rage that interferes with rational response.

Truly, the terrorist threat goes beyond all others, natural and industrial, and we cannot expect our organizations or our political culture to be up to it. Since 9/11, we hope our efforts have reduced the danger of a terrorist attack to some extent, even if probably not by much. But the safety of the United States has little to do with locking doors or watching mosques. The two main theories of radical Islamic terrorists are that (1) they want foreign troops out of their lands (Pape 2005), and (2) they seek to topple corrupt, authoritarian Muslim regimes such as that of Saudi Arabia, which the United States continues to support. (Benjamin and Simon 2002) According to these analyses, our foreign policy, by putting us at risk, has far more to do with our domestic safety than anything the Department of Homeland Security can do.

But even if these theories are correct, we have reason to examine our DHS experience very closely. First, the department is concerned with all hazards, not just terrorism, and the five years since 9/11 have seen deaths and destruction from natural and industrial disasters that exceeds the toll of decades of terrorist activity—by domestic terrorists as well as foreign ones. And while a significant terrorist attack may never again occur, we may be certain that our future holds more floods, hurricanes, earthquakes, forest fires, and industrial disasters. Second, our DHS experience is a cautionary tale that exposes some basic fault lines in our political system that must be taken into account as we continue to shape our security institutions. Third, it illustrates the inevitable imperfections of complex organizations, indicating the limits of what we can accomplish with them. And finally, it once again points toward the advantages of reducing basic vulnerabilities rather than just locking doors and setting alarms.

THE THREAT MODEL

Why was the administration of George W. Bush so unprepared for a terrorist attack on our homeland? The answer to this question is one part of the explanation for the dismal performance of the Department of Homeland Security. (The other part of the explanation is that we expect too much of our organizations.) It is a truism that the military is always prepared to fight the last war. Given the difficulty of predicting which adversary will strike next and how they will strike, this may be the best the military can do. When we live for decades under the threat of foreign attack, all our major institutions, not just the military and diplomatic corps, become firmly configured to meet this threat. The political area develops scripts and slogans to mobilize defense; a candidate or party cannot be weak on defense. Parts of business and industry thrive on defense expenditures and gain more power than those parts that are hurt by defense expenditures. Research money goes to the organizations that can build bigger bombs or bigger bombers, not to those who develop better body armor, rifles, battlefield communications. The media builds appropriate images of heroes and villains; the judicial system and social institutions such as education are altered to reflect the worldview. (Early in the cold war, university Russian Studies programs were quickly staffed with faculty without PhDs, contrary to university policy; Black Studies programs had a hard time getting started when they came along because they recruited faculty without PhDs, which was considered contrary to university policy, conveniently now evoked.)

A cold war that ran for some four decades can build up an impressive institutional support system that is confident that the next engagement will be similar to those anticipated for the last four decades. Many ways of life are elaborately, and sometimes comfortably, built around this sensible hypothesis.

But it was beginning to erode in the 1990s when no obvious enemy nation was apparent after the collapse of the Soviet Union. Since institutions, by their nature, change slowly and reluctantly, there remained a physical, organizational, and ideological infrastructure that could be mobilized by any president who argued that we remained threatened by foreign states.

Even events that challenged the lingering worldview need not shake it. The United States was repeatedly attacked by terrorists with no apparent state support in the 1980s and 1990s, sometimes on its own homeland. By the mid-1980s, Al Qaeda, a terrorist organization not supported by a foreign state, was identified as a serious threat to our security at home. The Soviet Union had collapsed in 1989, and there was irrefutable evidence that the United States was the only superpower left. But this prompted only minor adjustment in our defense strategy (more rapid-deployment troops; some reduction in nuclear weapons). But our intelligence agencies and military assets still thought of nations as the enemy, even after the first Twin Towers attack in 1993 was laid at the door of terrorists that were not sent by a foreign state.

Regardless of our superpower status, the mindset and the associated institutional direction led the George W. Bush administration to place something as difficult and expensive as an expansion of a missile defense system on a high-priority list when entering office in 2000, while counterterrorism funds appropriated under President Clinton were actually cut. This was not a case of institutional lethargy or drift, or even maintaining a steady state. Compared with the two previous administrations, the new administration vigorously reinstated and refurbished the subsiding cold war ideology. It was a trajectory, built on a mindset that had powerful institutional traditions. These can be mobilized. (For evidence of the cold war ideology, see former CIA bin Laden unit head, Michael Scheuer’s Imperial Hubris. [Anonymous 2004])

Our vulnerabilities, in the administration’s eyes, were not crowded skyscrapers symbolizing our economic power, like the Twin Towers, or the heart of the imperialist beast, like the Pentagon (attacked in 9/11) and the White House (almost attacked). The administration saw our vulnerabilities as the thousands of miles of open skies through which a nuclear missile could fly from a ship at sea that perhaps North Korea could someday launch. This vision is also a way of not seeing.

FAILED WARNINGS

To understand how the Department of Homeland Security and the reorganization of the intelligence community might turn out to be enormous failures, we have to examine the low priority that the White House put on its “war on terror.” Before 9/11 we had been repeatedly attacked abroad (in coordinated attacks on two of our African embassies, our base in Saudi Arabia, and the U.S.S. Cole), and at home (in the first attack on the Twin Towers in New York City). All were identified as attacks by nonstate terrorists. But the Bush administration dismissed the warnings about non-state terrorists by the Clinton administration’s transition teams on coming into office and the repeated warnings from CIA head George Tenet and security experts such as Richard Clarke and Rand Beers. It never even convened its task force on counterterrorism before 9/11, despite repeated warnings that Al Qaeda could be planning to hijack airplanes and use them as missiles. (Greenstein 2003)

In contrast, the Clinton administration got word in 1998 of a planned attack in the United States that entailed hijacking aircraft and then demanding the release of an imprisoned Muslim cleric. It responded by placing all major airports on the east coast, especially those serving New York, on alert until the end of January. Steps included intensified passenger screening and intensified supervision of screeners. The next year, 1999, fearing a “millennium attack” (on the basis of much less evidence of terrorist activity than the Bush White House had in the months prior to 9/11), President Clinton shared his presidential daily briefings with up to twenty-five people. (President Bush limited it to six after 9/11.) Clinton activated resources abroad, foiling some attacks, and activated resources at home. Airlines and airports were kept on alert, the border guards were alerted, and one terrorist was apprehended who was linked to Al Qaeda. (Perrow 2005; Staff 2004, 127, 533, n. 2) At least some dots were connected as a result of attention from the Clinton White House. Similar actions in September 2001 might have foiled the attacks of 9/11, but the even more urgent warnings were dismissed by the Bush White House.

Just before 9/11, the FBI’s requests to add hundreds more counterintelligence agents were rejected (Hirsh and Isikoff 2002) and a program designed to provide equipment and training for first responders in the event of a terrorist attack was cut. (Van Natta and Johnson 2002) The report of the Commission on 9/11 (Staff 2004) notes that the White House tried to cut funding for counterterrorism grants authorized by Congress, terminated a highly classified program to monitor Al Qaeda suspects in the United States, and in general downplayed the threat. In the early days after 9/11, the White House cut another FBI request for counterterrorism funds by nearly two-thirds, its requests for computer networking and foreign-language intercepts by half, a cyber-security request by three-quarters, and a program for increased collaboration entirely. (Milbank 2006; 2005; for other examples see Perrow 2006a) The major achievements of the administration in the months after the attack were the invasion of Afghanistan to destroy bin Laden’s base and training ground and speeded-up preparations for an invasion of Iraq. Both were conventional operations, whereas 9/11 also required unconventional ones. Even the Afghanistan operation barely pursued bin Laden; instead it “outsourced” the job to unreliable Afghani forces. The domestic terrorism program was limited to a small office of fifty or so professionals in the White House to deal with homeland security, set up shortly after 9/11.

ORGANIZING OUR DEFENSE

The First Homeland Defense Option: White House Control

Political scientist Charles Wise usefully lays out three options available to the government to cope with the threat to homeland security. The threat was acknowledged years before 9/11, of course, and the three options had been considered in various governmental reviews and task forces. Wise labels the option first undertaken by President Bush in the month after the attack as “executive order coordination.” (Wise 2002a)

This option was to coordinate and stimulate homeland defense from the White House by establishing an Office of Homeland Security (OHS), and appointing an assistant to the president to run it (Tom Ridge, the former governor of Pennsylvania, who was widely liked but had no terrorism experience). While this might suggest a minimalist response, as compared with the new Department of Homeland Security that we eventually got, it was in many respects the most promising option, since it emphasized coordination rather than centralization. It has the advantages of rapid response and flexibility and would give the president more direct authority than he would have with a Department of Homeland Security, which would give much control to Congress. But while it was in place, from October 2001 to July 2002, the OHS had an uncertain status and achieved little. As noted, the White House was cutting terrorist-related funds a month after 9/11.

This option, a White House office, would require exceptional activism from the White House, whereas a department would not. The presidential Assistant for Homeland Security would have to be aggressive, rather than just likable, and have international experience rather than gubernatorial experience. She or he would have to be strongly supported by the president in struggles with the Office of Management and Budget to reallocate funds and in struggles with the all-important intelligence agencies, particularly the CIA and the FBI, who were not sharing information. For example, using the threat of personnel replacements and budget cuts, the assistant would have to demand concrete evidence that the FBI changed it priorities from catching and prosecuting criminals and especially the purveyors of illicit drugs (its top priority), to investigating terrorist activity and sharing information with the CIA.

It is true that there are laws on the books governing cabinet bureaus and congressional actions that prevent the president or the head of the Office of Homeland Security from running roughshod over them, but presidents have been able to do a great deal outside of them. (Sylves and Cumming 2004) The theme of the “imperial presidency” mobilizing political and economic resources is a strong one in our history, even without crises. In a thoughtful personal communication, political scientist Patrick Roberts points out that while a White House czar for domestic policy can work during a time of crisis, coordinating domestic policy from the White House is more difficult than coordinating foreign policy. Domestic agencies are so closely tied to corporate, professional, and electoral interests that a White House adviser rarely has enough control to coordinate their efforts. Foreign policy is somewhat more removed from electoral politics by its nature and because of congressional committee control, so it is more responsive to White House direction. President Franklin Roosevelt was able to control domestic policy during World War II through a White House aide; Presidents Reagan and George H. W. Bush both tried it but had limited success. (Roberts 2003)

Executive order coordination has the virtue of preserving the decentralized structure of the many agencies that would be involved while adding a new measure of direction that would promote coordination. Decentralization does not necessarily mean fragmentation. Fragmentation comes when decentralized units do not receive explicit policy direction and the general oversight that indicates the policy is being carried out, and do not share information. The overwhelming problem with our homeland security and defense was a failure to coordinate—to connect the dots, as we endlessly heard. Coordination can be achieved through centralized control in small and moderately sized and homogeneous agencies, but mammoth projects are almost impossible to coordinate through centralization. Mammoth agencies require a great deal of decentralization because of the diverse tasks and skills involved. Decentralized systems are coordinated not by giving central orders but by signaling the policy direction, making sure that information is shared, and monitoring outcomes rather than behavior.

What, concretely, would this mean in terms of executive action? I am sure that the FBI director, Robert Muller, who was appointed just prior to 9/11 and was by all accounts more motivated and effective than any of his recent predecessors, can testify that changing an organization’s culture is a tough job. But that is what administrators are paid for. Unfortunately, there was no great reallocation of personnel from drug interdiction and prosecution to, say, translating intercepts and documents, huge amounts of which remained untranslated more than three years later. As mentioned above, the White House was cutting these funds, so the FBI was getting signals that terrorism was not much of a priority. But suppose it was a high priority, and the receipt of new funds signaled that? The Assistant for Homeland Security could have ensured that the FBI’s priorities were adjusted accordingly. The assistant could also have demanded that personnel be rotated between the FBI and the CIA. Such rotation had been ordered during the Clinton administration, but no one followed up so it did not happen—a typical Clinton administration failure. Following up would be the job of the Office of Homeland Security, and would not have required a platoon of staffers, only the support of the president.

For example, the FBI’s top management would have had to demonstrate that they took seriously and investigated the complaints of FBI translator Sibel Edmonds. She wrote superiors that she was ordered to slow up her productivity in order to justify a bigger budget request for the translation department, and that favoritism had led to hiring a translator who had not only failed to translate messages in Arabic languages accurately but was married to a man who was on the terrorist watch list. (They both left the United States immediately after the 9/11 attack.) The FBI management dismissed her charges, and instead they promoted her supervisor and his two superiors and fired Edmonds. (Perrow 2005) Strong signals from the Assistant for Homeland Security would have made it clear that investigating such complaints, and there were many, would be rewarded, instead of promoting the individuals charged.

Or, to take another illustration of the type of effort required, the so-called firewall erected in 1995 that supposedly restricted information sharing between the FBI and the CIA. The report of the 9/11 Commission notes this was the result of the “misinterpretation” of judicial rulings and procedures. The FBI conveniently saw them as prohibiting information sharing, and the Clinton administration did not address the problem, even though they were warned about it in 1999. (Still, in two crises in 1998 and 1999, information was shared in the Clinton administration.) But the Bush administration received strong warnings in 2000 and 2001 that intelligence agencies were not sharing data, and that the intent of the 1995 procedures was ignored routinely. (Perrow 2005; Staff 2004, 79) These warnings came at a time of higher threat levels than the previous administration had experienced, but still had no effect on the Bush administration. (Moreover, the forty or so warnings in 2000 about Al Qaeda in the presidential daily briefings—accorded a much higher threat level than in the 1998 and 1999 crises—appeared to have no effect.) These warnings of misinterpretations could have been addressed by a homeland security assistant who was fully empowered by the president and his top staff. It was front-page news that the FBI was not changing its procedures. In fact, the FBI’s several attempts to prosecute suspected terrorists were thrown out of court because of faulty procedures. Redirecting the FBI was certainly not a “slam dunk,” but it was far from impossible. When given tasks that fall within its capabilities and tradition, it could perform very well. Investigating the first World Trade Center bombing in 1993 and prosecuting the case was an example. Investigating an actual crime that supplies physical evidence and prosecuting the case is what the agency is about. It is not very good at dealing with the shadowy threats of terrorists, long-term surveillance, terrorists’s national and international movements, and cooperating with other agencies. This clearly could be changed. FEMA has been redirected at least three times by various White Houses (though it is a much smaller and weaker agency than the FBI). (An enlightening analysis of the difficulty of changing the intelligence community is by Amy Zegart, though she downplays political interests and especially presidential intent; see Zegart 2007.)

A former White House staffer who had been transferred to the new department indicated that she had even less power in the DHS than when she was in the White House. “You had a platform at the White House. Whenever you called a meeting at the White House, the other agencies came,” she said. “Now we’re over at the department and the agencies didn’t come; they came up with all sorts of excuses.” (Glasser and Grunwald 2005) This neatly captures the difference between executive coordination and the other options.

In his dramatic memoir Against All Enemies (2004), Richard Clarke indicates that President Bush’s initial plan might have worked. It envisioned Tom Ridge heading up a White House Homeland Security staff of about fifty professionals to lead, coordinate, and conduct oversight on the many federal programs involved with security and disasters. As Assistant to the President for Homeland Security, Ridge expected to have real authority, but soon complained that everything had to be cleared with the White House Chief of Staff. (Presumably that meant that homeland security had to be politically vetted, making it less than the top priority of the president.) While this was still consistent with presidential control of the program, it revealed that the president did not intend to take the domestic security issue seriously enough to deal directly with Ridge.

This was not a good sign, but the Office of Homeland Security still might have worked had the president given it a high priority. Clarke observes: “I believe that adept White House coordination and leadership could get the many agencies all working on components of a consistent overall program.” It would be difficult, of course, but over the years his own organization, the National Security Council, had managed to be quite effective during the Clinton administration. He continues: “the alternative method, rewiring the organizational boxes, would make us less able to deal with domestic security and preparedness for years to come.” (Clarke 2004, 250) He appears to have been correct. It had taken years for the mergers that created the Energy Department and the Transportation Department to jell, and both were smaller. Ridge apparently agreed with Clarke that the last thing needed was the reorganization involved in creating a new department.

Executive order coordination was the option that President Bush maintained without much enthusiasm from October 2001 until July 2002. The lack of enthusiasm is important; it opened the door to other interests.

The Second Option: Power Sharing

In the second option, what Wise calls the “statutory coordinator” option, the Homeland Security agency is established by law, rather than by presidential order. Laws are passed by Congress, so this gives Congress more power than under the first option, and thus reduces that of the president. This was proposed by the Gilmore Commission, which was considering these matters well before 9/11, having been formed in 1999. (Its official government title was the Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction.) Despite commissions like this before 9/11, aside from a counterterrorism group in the White House, little was done in the way of government reorganization. The Gilmore Commission had examined all efforts in its second annual report in December 2000 and concluded: “The organization of the Federal government’s programs for combating terrorism is fragmented, uncoordinated, and politically unaccountable.” The problem, it said, was lack of central authority, a routine and easy finding that commissions are prone to make. “The lack of a national strategy is inextricably linked to the fact that no entity has the authority to direct all of the entities that may be engaged.” (Wise 2002a) It called for a national office for combating terrorism that would have legal—that is, statutory— authority and be located in the executive office of the president. This location gives the president considerable authority, but it is shared with Congress, which writes the laws governing it.

The commission found that oversight responsibilities for combating terrorism presently lay in the hands of at least eleven full committees in the Senate, plus numerous subcommittees, and fourteen full committees in the House, with all their subcommittees, eighty-eight in all, each carefully guarding their turf. (Ornstein 2003) It was the same problem that earlier commissions had identified for FEMA. The homeland security effort had to be centralized, they, and almost everyone else, argued. This would mean new laws, and Congress would pass them, giving it influence. But its influence would not be scattered over eighty-eight committees and subcommittees. The key to the Gilmore Commission plan was that it recommended a reorganization of Congress itself to centralize control over homeland security matters by having a powerful oversight committee in each congressional house. These could override the authority of the eighty-eight parochial and fragmented committees. This would mean that the committees overseeing, say, the Coast Guard or border control, could see their authority reduced. This recommendation never succeeded. Individual committee and subcommittee chairs—which is where the power lies—were unlikely to support this reduction in their power.

But we have to be careful here. I will make much of congressional resistance to giving up committee and subcommittee power. But these committees are closest to the public, and are expected to be responsive to their constituencies. They are responsible for the fact that our federal system is very decentralized (or fragmented, depending on whose ox is being gored). The committees overseeing, say, border control and immigration or the Coast Guard, would be able to cry out that the agencies they oversee are underfunded, or the congressional committee could publicize the failures of the FBI or the CIA to coordinate with the agency the committee oversaw (the agency itself could not) or investigate charges of torture as members of a military affairs committee, or charges of waste in Pentagon contracts, and so on. The agency being examined is unlikely to bring such matters to the attention of the committee or the White House, nor, if it believes it is underfunded, does it have a spokesperson. There is a legitimate congressional advocacy and representative function, as well as funding pork-barrel projects. If a supercommittee was established, the charges of failures to coordinate, of torture, or of underfunding might not surface; any actions a regular committee might want to take could easily be overridden by the supercommittee. Congressional committees and subcommittees have reasons to want to keep jurisdiction and to exercise power, and it is hard to distinguish turf from representation of the interests of constituents. One is tempted to say that if the task is extremely critical—and combating terrorism seemed to be such—the risk of centralizing power in the super-committees needs to be run.

(The committee or supercommittee members might fail in quite other ways. They could easily put narrow local interests over those of the nation or could be so indebted to large campaign donors as to fail to represent all segments of their constituency. But this is quite another matter. These narrow interests would be present in the two large oversight committees the Gilmore Commission recommended, or, if there were no supercommittee, in the eighty-eight committees. They reflect a problem with our form of government that goes beyond that of restructuring agencies. I will discuss this in the final chapter.)

The Third Option: Keeping Congressional Control

It was the third option that prevailed—congressional control of a new department with Congress keeping its existing budgetary and personnel and mission control of the agencies and programs thrown into the new department. Based on the Hart-Rudman Commission findings of 2001, it preserved much of congressional committee power. (Gary Hart had been a Democratic senator from Colorado and Warren Rudman had been a Republican senator from New Hampshire, so the interests of Congress were likely to be well considered.) A new cabinet-level department would be created. The power of the president would consist in recommending to the Senate a nominee to be the secretary of the department, and exercising the normal amount of control the president has over any cabinet head, such as for agriculture or defense. While considerable, of course, presidential power would be less than in the first or the second option. Congress is the clear winner.

Political scientist Donald F. Kettl put it as follows, referring to Tom Ridge:

 

What the members of Congress left unsaid was that if Ridge remained a presidential appointee without congressional confirmation, they would have little control over his operations. If they could pass legislation authorizing the office, setting out its powers, gaining the right to confirm him in office, and controlling the office’s budget, they could dramatically shift the balance of power. Many members of Congress saw this as one of the biggest new initiatives in decades, and they wanted to ensure that they could control its direction. Bush turned them down, saying through a spokesman that the president did not need congressional action to do what was required. (Kettl 2004, 44)

 

Many agencies would be moved from their previous departmental location to the new Department of Homeland Security, but the agencies would still be under the supervision of Congress. Thus, the Secretary of the DHS would not have the power to use presidential authority to establish new units (e.g., a counterterrorism threat center that would not be under the control of Congress) or to force coordination through threatened budget cuts of, say, the immigration services, that the first option provided. All the congressional committees and subcommittees would still have considerable say over agency budgets and line authority; no supercommittees would override them, and Congress made sure none with any real power would ever exist.

Nevertheless, even this option created a high degree of conflict within Congress as it considered the new department. One senator was quoted as saying, “Hell hath no fury like a committee chairman whose jurisdiction has been taken away,” and the reshuffling would alter some jurisdictions. (Wise 2002b) Congress would lose some budgetary authority, and the power to make some line appointments, but not as much as under the other two options. A bill to establish the DHS was introduced by two senators, with Connecticut’s Joseph Lieberman taking the lead. (Democrats still had a majority in the Senate but Lieberman’s power was limited to heading a committee concerned with governmental organization that was, until he seized the initiative, of minor importance.) The president eventually reluctantly gave in and introduced a similar bill of his own.

If the president opposed the Lieberman bill, why did he give in after a few months and propose his own bill, which was very similar except that it included even more agencies than the Lieberman bill had? Clarke argues that the White House was about to have two disasters: (1) an unmanageable department that both houses of Congress strongly supported and (2) “the major new piece of legislation in response to September 11 would be named after the man whom the majority of voters had wanted to be Vice President just twenty months earlier.” It was better to have one of those two outcomes rather than both, so the President sent up a bill that would be called the Homeland Security Act, not the Lieberman Act. (Clarke 2004, 250)

The president’s bill was drafted in secrecy by five staffers unfamiliar with many of the programs they were moving, and “some of the decisions were made almost at random,” according to a stunning review of the department in December 2005 by Washington Post staff writers Susan Glasser and Michael Grunwald. The plan caught many agencies by surprise when it was revealed to the cabinet, and behind-the-scenes campaigns by agencies began. Powerful agencies such as the Justice Department and the intelligence community (FBI, CIA, NSA, etc.) refused to allow offices or staff to be moved out of their agencies. Health and Human Services did not want to lose the national drug stockpile. Energy blocked the transfer of the its Livermore lab (atomic weapons), and it made no sense to transfer the lab’s radiological detection teams when it was pointed out that the teams consisted of employees with regular jobs who mobilized only during emergencies. A key facility in a large agency is a policy office, but the Budget Bureau vetoed putting one in the DHS on the grounds that we should not foster big government. What policy there was came from the White House’s Homeland Security Council, which micromanaged the details of the DHS, asking for regular updates about uniforms for border guards, the curriculum for teaching border inspections, the selection of a single firearm for DHS training academies, and the like. The powerful Bush adviser Karl Rove vetoed a plan by Ridge to secure large chemical plants because it would give new regulatory power to the Environmental Protection Agency and was opposed by the chemical industry. “We have a similar set of concerns,” Rove wrote to the president of BP Amoco Chemical Co. Everyone, it seems, foresaw a disaster; one aide to a cabinet member said, “we all expected an ineffectual behemoth, and that’s what we got.” (Glasser and Grunwald 2005)

But what they got included extensive congressional oversight, written into the new law, and this added to the ineffectualness of the behemoth. I have characterized this option as “keeping congressional control, “ but it is more complicated than that. As a colleague at the Stanford Law School, Laura Donohue, points out, in many ways Congress did not achieve new control, but only maintained control over already existing agency functions and does not control many new initiatives. The executive office has substantial resources. The consequential actions related to homeland security were undertaken by executive agencies, rather than ones controlled by Congress. The Department of Justice expanded the Patriot Act without effective oversight by Congress; the intelligence agencies were unchanged; and the executive branch expanded the national security letters, Foreign Intelligence Surveillance Act warrants, and various collection powers, all of which have been used extensively. Congress dismantled the controversial Total Information Awareness program—intended to turn truck drivers, postal workers, TV cable installers, and others into counterintelligence agents—but the White House continued the course with its Highway Watch, Marine Watch, Neighborhood Watch, and other programs. Though defunded by Congress, the Defense Department’s Defense Advanced Research Projects program with its eighteen data-mining operations were transferred to the National Security Agency, CIA, and FBI. (Donohue 2006; Harris 2006) The NSA’s data-mining operation on millions of U.S. citizens with the help of all but one of the major telephone companies was a major political issue in 2006.

Thus the question of who won and who lost—the executive branch or Congress—is not as clear-cut as one might think. But Congress, as we shall see, kept substantial access to the barrels of pork that homeland security disgorged.

EXECUTING OUR DEFENSE

A Rough Start and Poor Reviews

Given the ignominious birth of the department, we should not expect it to achieve much homeland security for at least a few years, and the initial record sadly confirms that expectation. Even if it had strong presidential backing, the difficulty of merging diverse tasks, funding the new responsibilities, coping with congressional interests and with the inevitable uses to which organizations can be put severely limited the effectiveness of the department.

For starters, the launch was rough and premature. President Bush had resisted congressional efforts to establish the department, but once Congress passed the law, he set an unreasonably ambitious four-month deadline to open its doors to twenty-two agencies. It had a hard time finding any doors to open and initially, befitting the lack of enthusiasm on the part of the White House, was to be placed far out in the Washington suburbs. Ridge, appointed head of the agency, successfully fought that, but the agency still ended up stuck in the basement of a Navy building several miles from federal Washington without room to house personnel that were to be transferred there. It received little help in staffing; the secretary’s staff was very sparse; for weeks some offices lacked phones; the budget was so small that finding funds was a constant preoccupation. Touted as receiving a $40 billion allocation, it received far less in new money. One-third went to other agencies such as the Pentagon and most of the other $27 billion was not new money. Five of the twenty-two agencies had a total budget of $19 billion, which they brought with them, and was counted in the $40 billion figure. (See Matthew Brzezinski’s scathing and disheartening details on the failings of the department: Brzezinski 2004a; 2004b.) Congress’s dozens of committees still have oversight claims on the department, through their old ties to the agencies.

In such a situation, strong leadership on the part of Tom Ridge may not have been enough to rescue the failing launch. Still, according to the research of Susan Glasser and Michael Grunwald (2005) it was not forthcoming. Ridge did not insist on including a policy shop and spent a good deal of time worrying about the image of the department. Organizations such as the DHS should be designed to work with an average leader; we cannot count on getting an extraordinary one. We didn’t.

Glasser and Grunwald conclude, in their review:

 

To some extent, the department was set up to fail. It was assigned the awesome responsibility of defending the homeland without the investigative, intelligence and military powers of the FBI, CIA and the Pentagon; it was also repeatedly undermined by the White House that initially opposed its creation. But the department has also struggled to execute even seemingly basic tasks, such as prioritizing America’s most critical infrastructure. (Glasser and Grunwald 2005)

Congress has watched the agency very carefully. Secretary Ridge said that in the first year, he and his top assistants testified 160 times, about every day and a half, before congressional committees, and counting staff in general there were more than 1,300 briefings on the Hill. It takes from twenty-four to forty-eight hours to prepare for a briefing, he said. And this did not count the “hundreds and hundreds” of Government Accountability Office inquiries. (U.S. Congress 2004) The executive coordination option may have had fewer resources than a department, but they could have been better focused.

The agency has been watched carefully by several public interest groups such as the Council on Foreign Relations, the Heritage Foundation, and the Brookings Institution; by quasi-government groups such as the Rand Corporation and the Center for Strategic and International Studies; and most carefully by the Government Accountability Office (formerly the General Accounting Office), which is asked to do studies for Congress. There are some differences among the reports. For example, the Heritage Foundation emphasizes surveillance of citizens, and recommends that local law enforcement personnel “submit annual assessments of the events, activities, or changes in demographics or patterns of behavior of groups in their jurisdiction,” as a Rand Corporation summary of recommendations by commissions and public interest groups notes. (Parachini, Davis, and Liston 2003, 17) (This would reinstall Admiral John Poindexter’s Total Information Awareness program in the DHS.) The Brookings Institution, in contrast, has an economic tilt, and is the only one to weigh the budgetary implications of its own recommendations. It is easy to recommend actions as if they were costless.

The previous recommendations of the public interest agencies, like those of the dozens of commissions (Bremer, Hart-Rudman, Gilmore, etc.), suffer from heavy generalizations that evade the tough issues, urge more spending than is feasible, and urge actions without any guidance on setting priorities, as the Rand Corporation summary notes with sadness. (Parachini, Davis, and Liston 2003, 37) Some pay passing attention to first responders, but far too little, and I do not recall a single one that seriously considered the role of the average citizen.

The collaboration issue is a key one; the failure of agencies to connect has been the most prominent of the many 9/11 failures. The GAO and the public interest group reports drone on about the need to cooperate, network, collaborate, or link the disparate agencies within the DHS, as well as link DHS agencies with the powerful intelligence and defense agencies outside of it. But cooperation and collaboration are not costless, and appear to be the exception in government, not the norm. The authority over the agencies that are to cooperate (i.e., the White House) calls for cooperation but does not expend the resources (attention, monitoring, political capital, clear signals) to insure it. Every agency wants the others to cooperate with it but is reluctant to cooperate with them. The fear is that autonomy is lost unless cooperation is on one’s own terms, which isn’t cooperation at all. The reports all fail to confront this issue. Witt was able to get many agencies to cooperate within FEMA, and a good part of this may have been his skills, but the largest part could have been the support he had from the White House. Such support can turn even an administrator of modest talents into what appears to be unusual talents. (Organizations can be neither designed nor led in a way that requires exceptional personnel.) Therefore, the reports might have noted that given the interests of individual organizations, collaboration requires determined oversight and insistence by the head of the executive branch, the president. Asking the organizations to cooperate more, as the reports do, evades placing responsibility for their failure to cooperate where it belongs.

The GAO itself issued one hundred reports on homeland security even before 9/11, and in the first three years after that issued more than two hundred additional critical reports. Every week or two it cited a string of failures (and a sprinkling of successes) of the department and, while acknowledging the insufficient funding, it criticized the DHS’s poor fiscal management and waste, as did the DHS’s own inspector general office. Within a year of its creation (the legislation was signed November 25, 2002) the GAO designated it as a “high risk” agency, indicating serious performance problems.

It was not only large (180,000 employees) and diverse (twentytwo agencies with 650 separate computer systems to integrate), but so many of the agencies it took in were already high-risk agencies by GAO standards—that is, highly inefficient and poorly meeting the challenges in security and nonsecurity functions as well. The moves were unlikely to increase their performance, since some nonsecurity functions (e.g., fishing rights, computer crime, tariffs) might have benefited from staying close to other agencies that were not brought in to the new department. The new department merged agencies that, along with their security roles, had responsibilities for such unrelated activities as fisheries, mapping floodplains, river floods, animal diseases, livestock inspections, a national registry for missing pets (yes!), energy reliability, computer crime, citizenship training, tariffs on imports, drug smuggling, and the reliability of telephone networks. The potpourri of unrelated activities exceeded that of any previous large government mergers.

Interorganizational cooperation or coordination is familiar in the business world and among voluntary organizations. It takes work, but there is nothing impossible about it. Large corporations cooperate routinely on political and legislative matters (which is why some theorists refer to the “power elite”). Voluntary organizations, even when competing for funds or clients, cooperate informally and through formal superorganizations that they fund and join. Networks of small firms in Northern Europe, Japan, and to some extent the United States (Silicon Valley, biotech firms) cooperate extensively and productively, even while competing. And there are radically decentralized firms such as Johnson & Johnson that take steps to insure cooperation and coordination among their hundreds of small, independent units. It is far from impossible. But these cooperative enterprises are networks; DHS is not that form of organization. It consists of agencies with diverse histories thrown together, with only parts of each agency tasked with, or assigned, roles in a common enterprise, and subject to the oversight of several committees. In such a case, cooperation will be extremely difficult, perhaps more difficult than before the reorganization.

Public sector transformations are more difficult than those in the private sector, the head of the GAO, the comptroller general, wrote in a letter to a congressional committee, because organizations in the public sector must contend with more power centers and stakeholders, have less management flexibility, and are under greater scrutiny. Furthermore the top officials are typically political appointees who do not stay for long. (Walker 2004) Even in the private sector, where things are easier, his letter notes that more than 40 percent of executives in acquired companies leave within the first year, and 75 percent within the first three years. It even takes from five to seven years, according to research, to make mergers and acquisition work in the private sector, with all that sector’s advantages.

In December 2004, Clark Kent Ervin, the inspector general of the DHS (a watchdog division within the department) was not reappointed. He was a Harvard University–trained lawyer from Texas who had worked in the first Bush administration. But he had issued many critical reports accusing DHS officials of ineptitude and of fraud, including a charge that almost $50 million in excess profits were paid to the Boeing Company. (Ross and Schwartz 2004) He repeated some of his criticisms in an op-ed piece in the New York Times shortly after he was fired (Ervin 2004) and gives more blow-by-blow details in his book about his time in the department. (Ervin 2006) The failure to reappoint him was an alarming development, and, as we shall soon see, the first of many instances of using the terrorism crisis to countenance favoritism and corruption.

It is not clear as yet what the reorganization meant to the agencies that were moved, except that they were expected to take on new duties or increase their security efforts. In many cases it may only have meant a change in the letterhead, and the personnel continue to use their contacts with other agencies and go about their business. One cannot imagine any great changes in, say, the Plum Island Animal Disease Center, other than increasing their research on the deadly substances that terrorists might use. The Federal Protection Services, moved from the General Services Administration, may have received some new weapons and was told to have some meetings with the Office for Domestic Preparedness, moved from the Justice Department, but one can’t imagine much new synergy from such contacts or much new energy. (An exception, pointed out me by Laura Donohue, is the Exercise Division, transferred from the Department of Justice, which runs three hundred high-quality exercises a year. Another was to bring together several offices in various agencies that deal with border control.) But to the extent that new security responsibilities are added to the agencies, as must be the case with almost all of them, one can imagine that all the agencies that are folded into the department will be having a harder time doing the jobs they were originally designed to do.

The improvements that have been made, in border security, airline security, immigration checks, and a small beginning in port security, all could have been made without any major reorganization, if so directed by the White House. Indeed, the former inspector general wonders, considering all the failures and the loss of an intelligence role, “why it is, exactly, that we still have a Department of Homeland Security.” (Ervin 2006, 189)

New Roles for Old Agencies

Even if the DHS had not been formed and a coordination unit in the White House were responsible for our response instead, there had to be problems with asking many agencies to add terrorism to their task list. The Immigration and Customs Enforcement service is designed to let people in to the country, particularly those whose skills are deemed to be in short supply. It is now asked to deny entry to suspected terrorists but has little expertise to do this. The Border Control agencies are designed to prevent smuggling and illegal entry but have little capability to identify terrorists. Thus new tasks must be learned, and connections be made to parts of organizations they had little contact with (the FBI’s counterterrorist divisions and the CIA).

One basic problem is making fruitful connections between the relevant agencies. Finding the right organization is not easy. If you suspect a threat to your chemical plant, do you go to the local police, the FBI, or the EPA—or the Coast Guard or border control if it’s near the water or a border? Whichever one you pick then has the problem of deciding which of the others, and how many, should be involved. (Information should be shared, and dots connected.) If there is a dispute among them as to what actions to take, or what resources a unit should supply, what is the next level that would resolve the dispute? The organizations contacted may not accept your authority if you demand specific resources, or you may not be sure who has the resources you think necessary.

Even the question of what person you contact in the organization may be problematical. Take, for example, the frustrations of the Centers for Disease Control and Prevention (CDC). Though (fortunately) they were not transferred to the DHS, the example illustrates the interorganizational problems of combating terrorism. Shortly after 9/11, an anthrax attack occurred, sending the lethal substance through the mail, where contaminated sorting machines spread it randomly to hundreds of addresses in addition to the intended addresses. The experts who understood anthrax were at the CDC, so it was sensible to call them in, but they were caught off guard. What they knew about anthrax came almost exclusively from agricultural settings, where it occurs naturally and infects animals and farm workers. “Everything we knew about the disease just did not fit with what was going on. We were totally baffled,” said an official of the public health system. (Altman and Kolata 2002) Anthrax had always been an agricultural problem, but now it was a criminal one, thus requiring contact with the FBI. But, an official at the CDC told me, there was no reason for them to have any established links with the FBI. The FBI dealt with criminals, the CDC with microbes and victims. Still, one might think that networking should be easy; just get on the phone and call the FBI. (“Listen to the options carefully; they have changed.”) But who would they call? Since the FBI had no experience with what looked like a criminal epidemic, its help lines were of little use. (“Your call is important to us; please try again.”)

This official had, by chance, struck up a friendship with an FBI agent as a result of a totally unrelated conference a year before. She had the agent’s number; she called and explained the problem, and he was able to run interference for her up the FBI hierarchy to the appropriate level and office. (It is unclear how the Centers for Disease Control escaped being folded into the new DHS; given the threat of bioterrorism, they would seem to belong there more than, say, the Plum Island Animal Disease Center. Making them a part of the DHS would not have made it any easier for them to contact the FBI, even if the FBI had been also made a part of the department. Given their unique character and extraordinary importance and expertise, one can be grateful that the CDC remain reasonably independent.)

Since the attack was unpredicted, there were no established routines to call into play. Not only had the CDC seldom dealt with the FBI, they had had no reason to deal with the post office, state and local police agencies, and so on. And not only were there layers of phone numbers that needed to be searched, but the officials dealing with the attack had no authority over other agencies such as local post offices or police to get them to do what they thought needed doing once they found them. Agencies all have their organizational interests, jurisdictions, and clout. Networks are not hierarchical; the units have autonomy. An official of the CDC said they were unprepared “for layers and levels of collaboration among a vast array of government agencies and professional organizations that would be required to be efficient and successful in the anthrax outbreak.” (Altman and Kolata 2002)

Even when you successfully network, what authority do you have? A Connecticut state health officer, where there was a mysterious case of anthrax illness, said “we were very much aware that we had no jurisdiction over federal facilities whether it was the V.A. or the post office.” (Altman and Kolata 2002) Merging twenty-two agencies will not solve problems such as these, nor will injunctions to establish “clear lines of authority” or to “cooperate.” As the ample disaster and emergency response literature shows, coordination and cooperation requires (after the central authority lays out a meaningful mission and exercises oversight) frequent drills, exercises, simulations, and meetings where diverse agencies get a chance to see each other’s point of view, establish personal contact, and build trust. This is new work, and in turn requires increased budgets. Even the efficient agency is not likely to have slack funds to handle a new unfunded mandate. Yet, as noted above, the first-responder agencies are grossly underfunded. Creating a DHS may have made it harder to direct the new terrorism funds to official first responders—larger interests got there first— and appears to have even diverted FEMA funding from them.

Organizational Problems: Displacement of Missions

A big chunk of the new Department of Homeland Security contains an agency we discussed in the last chapter: the Federal Emergency Management Agency. What would happen to its traditional concern, natural disasters? The fate of programs concerned with natural disasters under DHS was a concern from the beginning. FEMA director Joe Allbaugh was asked early in 2002 about this and told Congress that the traditional role of FEMA would not be affected. But even Republican lawmakers were not convinced. Representative Don Young (R-AK), chairman of the House Transportation and Infrastructure Committee, said that if the homeland security secretary wanted to redirect the agency and focus on preventing terrorist attacks, he could reduce “other [FEMA] missions and direct those resources entirely to security.” Congressman Young had good reason to think this possible: This is what the first director that President Reagan appointed did, as we have seen. To forestall this, the chair of the Select Committee on Homeland Security, Richard Armey (R-TX) redrafted the White House proposal to keep FEMA primarily an agency dealing with natural disasters. The White House objected, suggesting that a displacement of its mandate could indeed have been in the cards. Some senators and the highly regarded former FEMA head, James Witt, were opposed to putting FEMA in the new department, but it was moved. (Pincus 2002) Nevertheless, when Florida was hit by a succession of four major hurricanes in 2004, there was no outrage about FEMA’s performance. As we shall see, there were strong political reasons to pay attention to those hurricanes and be prepared. But otherwise, the agency had been weakened.

Organizational Challenges:

New Tasks, Few New Resources

A further problem was what happened to the agencies when they were transferred to DHS, and given added homeland security tasks, but expected to continue with their usual ones, often with no significant budget increase, if any. In Pittsburgh, where the Coast Guard helps control traffic on the busy Ohio River, homeland security activities had accounted for 10 percent of Coast Guard activity; it now grew to 50 percent. This meant cutting other activities, such as assisting boaters and acting as traffic cops on the crowded river. The Coast Guard’s effort in drug interdiction declined by 60 percent after 9/11, and time invested in preventing an encroachment on American fishing territories and enforcing fishing rules shrank 38 percent. (Kettl 2004, 39–40) One might say that at least the Coast Guard changed its priorities, but increased funding for security could have left the old activities in place if they were deemed necessary.

The U.S. Border Patrol is an example of inadequate funding coupled with new tasks. It hasn’t had much to do on the 4,000-mile border with Canada. But a terrorist was caught (almost accidentally) bringing explosives over the border to be used in an attack on the Los Angeles airport. Then it was learned that some terrorists used Canada as a port of entry (though not any of the 9/11 terrorists). The Border Patrol, in 2002, had only 330 agents supported by one analyst to intercept illegal crossings of the 4,000 mile border. In the preceding twenty years two hundred had been cut in government downsizing efforts. Half the inspection booths were simply closed. (Wise 2002a) As of April 2004, miles and miles of the border consisted of dense, overgrown brush where before there had been cleared spaces. In 2004, it finally received some more resources for an impossible task.

A serious problem has emerged that concerns the critical area of official first responders—police, fire, and emergency medical personnel and various voluntary and homeowners associations. As mentioned previously, the title of a 2003 Council of Foreign Relations task force report summed up the problems: “Emergency Responders: Drastically Underfunded, Dangerously Unprepared.” (Rudman, Clarke, and Metzl 2003) The task force declared the underfunding by government at all levels to be extensive and estimated that combined federal, state, and local expenditures would have to be tripled over the next five years to address this unmet need. Covering this funding shortfall using federal funds alone would require a fivefold increase from the current level of $5.4 billion per year to an annual federal expenditure of $25.1 billion. Nor would these funds provide gold-plated responses; they would go to essentials. For example, the executive summary gave these examples of deficiencies:

 

• On average, fire departments across the country have only enough radios to equip half the firefighters on a shift, and breathing apparatuses for only one-third. Only 10 percent of fire departments in the United States have the personnel and equipment to respond to a building collapse.

• Police departments in cities across the country do not have the protective gear to safely secure a site following an attack with weapons of mass destruction.

• Public health laboratories in most states still lack basic equipment and expertise to adequately respond to a chemical or biological attack, and 75 percent of state labs report being overwhelmed by too many testing requests.

• Most cities do not have the necessary equipment to determine what kind of hazardous materials emergency responders may be facing. A study found that only 11 percent of fire departments were prepared to deal with the collapse of buildings with more than fifty inhabitants, 13 percent were prepared for chemical or biological attacks, and only 25 percent had equipment to communicate with state or federal emergency-response agencies. (Shenon 2003a)

 

Furthermore, the funds that the federal government did allocate for emergency responders were sidetracked and stalled due to a politicized appropriations process, the slow distribution of funds by federal agencies, and bureaucratic red tape at all levels of government, according to GAO reports.

Congress has played a substantial role itself in hampering the effort. The $3.5 billion promised by the White House in January 2002 for first responders in the state and local governments fell victim to partisan squabbles in Congress, not being approved until more than a year later, in February 2003. Given congressional control over the DHS, this was to be expected.

A glaring example of thoughtless, on-the-fly reorganization occasioned by the birth of the new department concerned a particularly important sector of first responders—the effective National Disaster Medical System (NDMS). It deployed and coordinated volunteer teams of doctors, nurses, and other medical personnel in a crisis, some seven thousand volunteers. It had been moved from the Department of Health and Human Services to FEMA by the original DHS designers. HHS felt it belonged in HHS, not in FEMA, and tried to wrest control of the NDMS during Hurricane Isabel (September 2003). It did not do well in FEMA and was starved of resources. Its paid staff had shriveled from 144 to 57 and did not even include a physician. NDMS volunteers complained about FEMA’s unpaid bills, faulty equipment, and intransigent leadership. The National Association of NDMS Response Teams sent a harsh letter to Ridge’s successor, Michael Chertoff. It warned that two years after their move to FEMA, they were less prepared than ever: “We feel that the identity of the NDMS is being lost via FEMA’s efforts to ‘swallow’ NDMS functions, rather than support them. . . . During transition, it has been fragmented, reduced, and relegated to a position without the authority, staff, resources . . . or systems in place at FEMA to move forward with the most fundamental of readiness and critical mobilization issues.” (Grunwald and Glasser 2005) This is one of many instances wherein the DHS has made us less safe.

Organizational Uses

Organizations, as I have often noted, are tools that can be used by those within and without them for purposes that have little to do with their announced goals. A new organization such as the DHS invites use. As soon as the department was established, the corporate lobbying began. Four of Secretary Tom Ridge’s senior deputies in his initial position as Assistant for Homeland Security at the White House left for the private sector and began work as homeland security lobbyists, as did his legislative affairs director in the White House. The number of lobbyists that registered and listed “homeland,” “security,” or “terror” on their forms was already sizable at the beginning of 2002, numbering 157, but jumped to 569 as of April 2003. One lawyer for a prominent Washington DC law firm was up-front about corporate interests. He mentions in his online résumé that he authored a newsletter article titled “Opportunity and Risk: Securing Your Piece of the Homeland Security Pie.” (Shenon 2003b) It is a very large pie indeed.

A Web document, “Market Opportunities in Homeland Security,” introduces buyers to the “$100 billion” homeland security marketplace, for $500.00 plus shipping. Less exuberant in its predictions, a Frost & Sullivan report indicates the industry generated $7.49 billion just in 2002, with total market revenues of $16 billion estimated for 2009. Frost & Sullivan is an “international growth consultancy,” found at www.frost.com. A report from Govexec.com by Shane Harris, titled “The Homeland Security Market Boom,” issued less than six months after 9/11, documents the aggressiveness of U.S. business in flocking to the new funding source. “Every good company out there can take what they do and reposition it for homeland defense,” says Roger Baker, the former chief information officer of the Commerce Department, who is now with a private company. (Harris 2002) In 2006, less than three months after registering as a lobbyist, former attorney general John Ashcroft was developing a practice centered on firms that want to capitalize on a government’s demand for homeland security technology, and had banked at least $269,000 from just four clients. Observers said it was virtually unprecedented for a former U.S. attorney general to start a lobbying firm. (Staff 2006a) An article in Rolling Stone lists a number of products that lobbyists procured for their client: Tiptonville, Tennessee (population 7,900) received $183,000 for an all-terrain vehicle, defibrillators, and hazmat suits for the volunteer fire department. The mayor explained: “If I were Al Qaeda and wanted to get to Memphis . . . I’d come to Tiptonville. No one would expect me there.” Mason County, Washington (population 54,000), got $63,000 for a biochemical decontamination unit that no one has been trained to use. Bennington, New Hampshire (population 1,450), got nearly $2,000 for five chemical weapons suits. The police chief said he saw no specific threats but it was being offered so he figured that he would get on the bandwagon. (Klinenberg and Frank 2005) Converse, Texas, used its money to transport lawnmowers to the annual lawnmower race, and the mayor of Washington DC, Anthony Williams, certainly a city that is a legitimate terrorist target in contrast to these others, spent $100,000 to produce a rap song on emergency preparedness and $300,000 on a computerized car-towing service. (Ervin 2006, 186)

As bad as all this was, what Clark Ervin’s chapter on fraud and waste within the DHS, principally by the Transportation Security Administration (TSA) and the bureau of Immigration and Customs Enforcement (ICE), reveals is over-the-top. Contractors big and small, some of them national figures such as Boeing or Unisys Corporation swarmed over the agency, charging extravagant fees with cost-plus contracts that had no records of expenditures. The failures of DHS managers seem almost willful, and in some cases involved outright fraud. Ervin’s office, according to his account, was stymied by DHS top management in his attempt to establish accountability. (Ervin 2006, Chap. 9) It continued, of course, after he was not reappointed, with the TSA literally losing track of millions of dollars every month in 2005. (Phillips 2005)

There were other within-government uses, too. Our chapter on natural disasters noted that presidential declarations of disaster areas, and the federal funds that followed, varied directly with the political importance of the area to the president of the time. Shortly after 9/11, Congress passed the USA Patriot Act. Senator Patrick Leahy (D-VT), then chairman of the Senate Judiciary Committee, wrote in the criteria for distributing some $13.1 billion among the states. His committee used a formula long in use for distributing much smaller funds, one that favored the small states. The small states now resisted any change in the formula, and could do so since they had the power in the Senate. The funding was almost exactly in reverse order of the threat. (The degree of threat has been assessed by a nongovernmental research organization using sophisticated probability models.) The ten highest amounts went to states and districts with the least threat, except for Washington DC, where the congresspeople work. Thus Wyoming received $61 per person, but California only $14. Alaska, hardly a target for terrorism, received $58 per person, while New York, the target of six separate plots by Islamic terrorists in the last decade, got only $25. (Ripley 2004)

This point deserves elaboration. The formula meant that 40 percent of the funds had to be divided equally among the states, regardless of population. The executive branch had discretion over the remaining 60 percent, and it did at least somewhat better by distributing it according to each state’s population; but it too did not distribute the funds according to the risk the state’s population was exposed to. In early 2003, Congress announced a plan that might rectify the situation, a new $100 million grant for “high threat” urban areas only. New York City, for example, would get 25 percent of it. Immediately, Congress pressed the administration to increase the size of the lucrative handout and also increased the number of cities at risk. Disasters are funding opportunities. Soon fifty cities, perhaps politically important to the administration, were designated as “high threat,” and while the size of the grant grew to $675 million, New York City received only 7 percent instead of 25 percent. Its funds were doubled, but many more low-risk cities were now funded. Democrats charged that the Bush administration allowed this to happen because it doesn’t have a constituency in the big cities. (Ripley 2004, 37)

Despite charges that the basic formula in use distorted federal outlays and was only partially rectified by the grants to specific cities, the formula was still in place at the end of 2004. The House approved a bill to have the funding formula reflect the risks the states faced, and the White House, to its credit, made a similar request in its 2005 budget. But the Senate would have none of it. Senator Leahy is a member of the powerful Homeland Security Appropriations Subcommittee, and his state of Vermont gets a handsome $54 per capita. He curtly reminded Secretary Ridge of the power of small states such as his. “I have to say, I was really disappointed that the President’s proposed budget . . . drops the all-state minimum formula,” he said. “That would affect all but, I think, one or two in this subcommittee.” He charged that the administration “wants to shortchange rural states.” The funding bill, according to Senate aides, “would go nowhere,” and it didn’t. (Ripley 2004)

Wyoming is the number one recipient of homeland security money per capita. Asked about this, the typical remark was that “our citizens deserve the same kind of protection that they are afforded in other places in the country.” This was from the chief of police in Douglas (population 5,238), who had just received a new $50,000 silver RV that serves as an emergency operation command center, paid for with federal dollars. Firefighters in Casper, Wyoming, even denied they were less at risk than, say, New York City residents. “No one can say Casper can’t be a terrorist target.” Wyoming had the largest budget surplus, as a percentage of budget, of any state in the nation. Yet the seriously in-debt state of California spends five times as much, per person, of its own money on homeland security—taxes its citizens pay—as does Wyoming. (Ripley 2004, 37) The unrepresentative character of the Senate (Dahl 2003) and the parochial interests of the citizens of small states, who expect their senators to bring in the federal dollars, will make it difficult to respond to our vulnerabilities.

Finally, in December 2004, the DHS was able to get around Senator Leahy’s bill, and announced a new formula that focused on cities rather than states, over the protests of small states, that went partway to matching funds to threats. New York City was the biggest winner, going from a $47 million grant to $208 million, and Washington DC, Los Angeles, Chicago, and Boston got smaller increases. (Lipton 2004) Senator Diane Feinstein (D-CA) tried again to further increase the proportion of funds to risky areas in 2005, but Senators Collins (R-ME) and Lieberman (D-CT) blocked it. As of this writing, the issue is at least still alive.

Another Organizational Use: De-unionizing

The new department offered opportunities to further presidential agendas unrelated to the terrorist threat. Though President Bush did not favor the department and its massive movement of twenty-two agencies, it provided an opportunity for what appeared to be an attack on civil service. President Bush immediately demanded that Congress strip all employees who would be transferred of civil service status. Liberals and labor saw this as an attack on the eighteen different government unions, and that it would reduce the amount of union membership in the government significantly. The president argued that because of the unique, nonroutine nature of defense of the homeland, the department needed to be free of civil service and union restrictions on terminating employees. The matter dragged on until March 2003, when the last day for final comment on the proposal arrived. A ninety-one-page comment from three powerful unions representing about one-quarter of the department’s workers arrived. The unions had squared off for a fight.

The DHS and the Office of Personnel Management proposed regulations that would cover 110,000 of the department’s 180,000 employees, affecting how they would be paid, promoted, and disciplined. It would become a model for revamping civil service rules in the rest of the federal government. Pay would be linked to performance (political performance and less aggressive bargaining, the union argued), union bargaining rights in several areas would be restricted (e.g., deployment of workers and use of technology), and the government would speed up and tighten the disciplinary process. (Barr 2004) There was a lengthy standoff, and as of October 2005, the issue is still in dispute since a federal judge, for the second time, ruled the new personnel rules invalid. (Staff 2005e) Disasters are opportunities.

Departure of Key Personnel

The departure of seasoned terrorist experts started almost immediately. Rand Beers had thirty-five years of experience in intelligence; he had replaced Oliver North, who was the director for counter terrorism and counternarcotics in the Reagan administration. He spent seven months in the new department, and five days after the Iraq invasion in March 2003, he resigned. Three months later, he told a Washington Post reporter of his disaffection with the counterterrorism effort, which was making the country less secure, he said. The focus on Iraq, he said, “has robbed domestic security of manpower, brainpower and money.” (Blumenfeld 2003) Agreeing with many counterterrorism experts, he saw the minimalist Afghanistan war as only dispersing Al Qaeda and not pursuing it enough to disable it, and the maximalist Iraq war as recruiting terrorists. Another disaffected expert, Richard Clarke, left the NSC in February 2003, just before the Iraq invasion, saying the same thing. His revelations about the misdirected, underfunded, and bureaucratically incompetent response to the terrorist threat, Against All Enemies, made the bestseller lists in April 2004.

Other experts departed or would not be recruited. A New York Times story in September 2003, six months after the start of the department, reported two top officials leaving. “So few people want to work at the department that more than 15 people declined requests to apply for the top post in its intelligence unit—and many others turned down offers to run several other key offices, government officials said.” (Mintz 2003) The administration announced that 795 people in the FBI’s cyber-security office would be transferred to the DHS, but most decided to stay with the more reliably funded, higher-status FBI, and only twenty-two joined the new department. (Mintz 2003) Flynt Leverett, who served on the White House National Security Council for about a year until March 2003 and is now a fellow at the Brookings Institution, observed, “If you take the (White House) counterterrorism and Middle East offices, you’ve got about a dozen people . . . who came to this administration wanting to work on these important issues and left after a year or often less because they just don’t think that this administration is dealing seriously with the issues that matter.” (Drees 2004) In a union survey of eighty-four union personnel in the DHS, 80 percent said it was a “poorer agency,” and 60 percent said they would leave if they could get the same salary in another agency; and the GAO rated morale at the DHS as one of the lowest of any government agency. (Elliston 2004) For other examples of departures, see Against All Enemies. (Clarke 2004) The “brain drain” continued into 2006 and appeared to be increasing. (Hall 2006; Lipton 2006f)

Centralizing to Combat a Decentralized Enemy

A final concern with the new department was the emphasis on centralization. Unfortunately, the immediate response of most politicians and even some administrators to signs of poor coordination, indifference to changing environments, and new tasks is to rein everyone in, centralize, and give specific tactical orders. I expect this is what Senator Lieberman and others had in mind when they wrote their proposed bill, which was similar in this respect to the one the president’s aides later drafted. It was easy to react this way to incredible stories of bungling and mismanagement. Bringing twenty-two agencies under one command seemed quite sensible, since they rarely had worked together. In the intelligence reorganization legislation, which we will come to later, creating an intelligence czar was the easiest response to the credibility-challenged CIA and the balkanized fiefdoms of the FBI.

In contrast, some commentators recommended that the structure of the new department and of the intelligence agencies should match the structure of the enemy. I do not know of any compelling theoretical arguments to support this view, but it makes considerable sense nevertheless and is worth exploring. The threat comes from highly decentralized terrorist networks, while the response comes from two newly centralized, hierarchical agencies. Current Islamic terrorism of the Al Qaeda type involves unpredictable acts by cells in loosely coordinated networks. To defeat the networks requires on-the-ground operatives with maximum autonomy that can infiltrate the networks and also exercise close surveillance, striking only when an operation is imminent. A centralized response is to bomb any suspected targets, raid them, and round up anyone who looks suspect. To have the best response to domestic threats is to allow considerable autonomy to border control agencies, to airlines and other transportation agencies, and, when the attack comes, to official first responders. (For devastating critiques of the 9/11 Commission’s view of the role of citizens and official first responders and its recommendation to centralize what should be decentralized, see K. Tierney 2005.)

Since intelligence was to remain outside the DHS, I would recommend a coordinating role for the head of intelligence, instead of a czar, whose responsibilities would be to collate information from the intelligence community; conduct frequent reviews, surveys, and evaluations of the separate agencies; and recommend budgetary changes to the White House. The intelligence coordinator should not have distracting operational duties, such as George Tenet did as also directing the CIA. Intelligence would remain decentralized. The head office of the DHS would behave similarly— coordinating, reviewing, evaluating, and handling the budgets of the diverse agencies. Indeed, most of the twenty-two agencies would not need to have been moved. With strong White House support for the head of intelligence and the head of a homeland security coordinating office, a decentralized response need not mean chaos or balkanization, but agency empowerment.

FEMA UNDER A BUSH

We will leave the DHS as a whole at this point, and turn to FEMA, a cabinet-level small agency that had been swallowed up by the DHS whale. Though FEMA appears to have done well in the succession of hurricanes in Florida in 2004, two things made this response unusual. Florida has yearly hurricanes, and Governor Jeb Bush had a well-organized state response team that was able to assert independence from FEMA when necessary. Second, it was also a politically sensitive state with a Republican governor, offering great opportunities for spreading federal grants, even over areas that were not touched by the hurricane. This was a crucial election year for President Bush, and Michael Brown, the head of FEMA, did his job well. In a six-week period in August and September 2004, emergency-supply trucks were pre-positioned to deliver ice, water, cots, blankets, baby food, and building supplies. The magazine Government Executive reported: “Seldom has any Federal agency had the opportunity to so directly and uniquely alter the course of a presidential election. . . . Seldom has any agency delivered for a president as FEMA did in Florida this fall.” (Klinenberg and Frank 2005)

Katrina was different. It was enormous, the state was not politically important, there was little pre-positioning, and FEMA had had another year in which to decay. Michael Brown, appointed head of FEMA in 2002, became the target of the news channels, late-night talk shows, and Comedy Central. During Katrina he managed to make a fool of himself with e-mails about what time he needed to have dinner and his success at shopping at Nordstrom while New Orleans drowned. But Washington Post writers Michael Grunwald and Susan Glasser (2005) offer a far more complex view of this contradictory personality and his agency: Brown was dedicated to FEMA and to keeping it independent (after all, his career was riding upon it), and he alienated the White House and other DHS agencies and its leadership with his harsh analysis of their failings. Just as Katrina was an outsized storm, Brown and his agency are outsized failures, going far beyond the normal inevitable failures of organizations that are the subject of so much of this book. As flawed as the response to 9/11 was, all the responsible government agencies (except the Coast Guard) appeared to perform considerably worse in Katrina. This failure requires exploration and interpretation. Though unusual in their magnitude and tragic consequences, the failures of the DHS and FEMA and its director are clearly within the realm of the possible, thus strengthening the theme of this book that prevention and mitigation will always fall short, sometimes alarmingly so, and we should begin to reduce the size of our vulnerable targets.

Joseph Allbaugh was once a third of President Bush’s “Iron Triangle,” along with Karl Rove and Karen Hughes, but was “exiled,” Grunwald and Glasser say, to head up FEMA after the 2000 campaign. Then in 2002 Bush announced that FEMA would lose its cabinet-level status and be placed under the new DHS, so Allbaugh quit. He arranged to have his deputy, Michael Brown, succeed him. Initial press reports in Time and The New Republic indicated Brown did not have a background in emergency management and had exaggerated his teaching experience in his résumé, but these were successfully contradicted by affidavits he assembled for testimony before a House select bipartisan committee investigating the hurricane months later. (U.S. Congress 2005) Brown had gathered experience in bureaucratic politics in his two years working for Allbaugh. He was determined to keep FEMA as independent from DHS as possible, but the first head of DHS, Tom Ridge, wanted it to be integrated into DHS and to be a key player in DHS, and spoke of putting FEMA “on steroids.” To do so, however, meant that its focus would have to shift from an emphasis on natural disasters to one on terrorism. Ridge stripped FEMA of its control over the millions of dollars worth of preparedness grants concerned with 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 and schools that had been retrofitted for earthquakes with FEMA funds were protected from high-impact structural hazards. The day of that quake was also the day that the new president, George 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. Three out of every four grants for mitigation 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 probably was a major blow to states such as Louisiana that are prone to weather disasters.

But the states and counties themselves may have weakened their disaster programs. Political scientist Patrick Roberts details the extensive cuts in federal funds for natural disasters, but he also makes another telling point. “State and local emergency management agencies reorganized to meet the terrorism threat to a much greater degree than has FEMA. . . . Many state emergency management agencies may have simply been too small and weak to withstand the funding and attention shift toward the terrorist threat. These agencies depend on federal and state grants for their operational budgets, and when grant criteria emphasized the terrorist threat, state and local agencies shifted their priorities. In addition, the law enforcement culture, which is more concerned about terrorism than is the natural hazards culture, is stronger in some state and local agencies than at FEMA.” (Roberts 2005, 443) This may help account for some of the alarming failures at the state and local level.

Ridge also seized the Office for Domestic Preparedness (ODP) from the Justice Department. This was a major source of funding, so it was politically valuable. The OPD did not go to FEMA, however, but to his own office in DHS, and FEMA’s preparedness grants went there too. Brown objected. He sensibly noted that at the state and local level, the people responsible for preparing for disasters were the same who responded to them; it did not make sense to pry them apart. Brown appealed all the way to the White House but was overruled. (Grunwald and Glasser 2005)

FEMA also lost its grant program for fire departments, its terrorism training program, and still other grant programs. Ridge’s office got the job of creating a “National Preparedness Goal” that would create likely scenarios, something FEMA had expected to do. Finally, Ridge’s office got the granddaddy of them all, the National Response Plan. By the time Katrina hit, the plan had not been exercised, nor had its all-important appendix dealing with first responders been drafted.

In this environment we would not expect an appropriate response to the hurricane, but the details of the response are still alarming. For example, FEMA director Brown 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—and shown on national TV news—since Monday, but Brown told an incredulous TV interviewer, Paula Zahn, that Thursday, “Paula, the federal government did not even know about the convention center people until today.” (Lipton and Shane 2005)

FEMA and Brown also did not know where the ice was. It was not pre-positioned, as it had been in Florida. 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 four thousand trucks, costing the taxpayers more than $100 million. Most of it was never delivered. In an age of sophisticated tracking (FedEx, DHL, WalMart, etc.) FEMA’s system broke down. Asked about the vital ice, 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 the food, drugs, and medicine for the victims fresh, to treat people with heat exhaustion, and to keep the sick, old, and frail cool. There was plenty of it for Florida the year before. The drive to privatize was signaled earlier by Brown, before he was made FEMA head. At a conference in 2001 he said: “The general idea—that the business of government is not to provide services, but to make sure that they are pro-vided—seems self-evident to me.” (Elliston 2004)

For days following Katrina empty air-conditioned trucks with no supplies drove aimlessly past “refugees” who had no water or food or protection from the sun. Reporters came and went, but food and water and medical supplies did not. (Staff 2005d) The Red Cross was not allowed to deliver goods because that might discourage evacuation. (American Red Cross 2005) Evacuation by air was slowed to a crawl because FEMA said that post 9/11 security procedures required finding more than fifty federal air marshals to ride the airplanes and finding security screeners. This search was prolonged. At the airport’s gates, inadequate electric power for the detectors prevented boarding until officials relented, but they still required time-consuming hand searches of desperate and exhausted people. (Block and Gold 2005) The only food— emergency rations in metal cans—was confiscated because the 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. (J. Tierney 2005) Without functioning fax machines to send the required request forms, FEMA would not send help that local officials begged for. (Apparently no one at FEMA dared violate the rules, even in such an emergency.) Perhaps a fifth of the New Orleans police force simply quit—exhausted and discouraged, their own homes gone—or were themselves looting. A large National Guard force hid behind locked doors in the convention center, saying “there were too many of them for us to help” and when they went forth on their mission they sneaked out to avoid the hungry evacuees, saying they needed their food for themselves. (Haygood and Tyson 2005) A Navy ship with transport and hospital facilities idled offshore, waiting for days to be called. Almost five days after the next hurricane, 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. (Rita came ashore just a few days after Katrina.) For other tales of similar failures, see “Hurricane Katrina as a Bureaucratic Nightmare,” by Vicki Bier. (2006) As noted in chapter 2, the failures were not only official ones. Because Murphy Oil failed to prepare its tanks properly for the storm, 1,800 homes were made uninhabitable and the St. Bernard Parish of New Orleans was polluted by 1.1 million gallons of oil.

A particularly disturbing report about the new DHS came out in the congressional investigations of the hurricane response. The department had set up a Homeland Security Operations Center that costs about $56 million a year to run and provides up-to-date news on impending disasters for more than thirty agencies. With Brigadier General Matthew E. Broderick, a veteran of Vietnam and Somalia, in charge, the center watched Hurricane Katrina bear down on the Gulf Coast on Monday, August 29. The dispatches begin rolling in on Monday morning, reporting major flooding in some parts of the city, people calling for rescue from rooftops and attics, ten feet of water already in some areas, and the flooding increasing. Warnings of breached levees came in from the Coast Guard, the Army Corps of Engineers, and the Transportation Security Administration during the day. But before he left to go home to bed about eleven o’clock that evening, Broderick saw a television report showing that there was drinking and partying in the French Quarter, which was on high ground and never actually flooded. That report apparently reassured him. He told investigators that he was not surprised to hear of flooding during a hurricane, that was expected, but that the operations center’s job was to “distill and confirm reports.” “We should not help spread rumors or innuendo, nor should we rely on speculation or hype, and we should not react to initial or unconfirmed reports which are almost invariably lacking or incomplete.” So he did not place a call Monday evening to Mr. Chertoff at the DHS or to the White House. It is probably irrelevant that the head of FEMA unilaterally decided not to work with the center and didn’t tell Broderick directly what FEMA had witnessed that day, since Brown himself was ignoring television coverage. (Lipton 2006a) The Department of Homeland Security Operations Center is a poster child of the organization, used to advertise the country’s preparedness.

Katrina also triggered the biggest deployment in the National Disaster Medical System’s history. One official called the result “a national embarrassment.” “In an after-action report, a NDMS team leader Timothy Crowley, a doctor on the Harvard Medical School faculty, called the deployment a ‘total failure.’ Crowley’s team was summoned late, then sent to Texas instead of Louisiana, then parked in Baton Rouge for a week while New Orleans suffered.” The team was finally sent to the disaster zone and was immediately overwhelmed by the demands for help, but no additional help was available. It later found out that a host of other teams “had been sitting on their butts for days waiting and asking for missions” said Crowley. (Grunwald and Glasser 2005)

Accounting for the Katrina Response

FEMA was not the only organization to fail so massively in Katrina and Rita, but it, and its parent organization, the Department of Homeland Security under Michael Chertoff, was certainly the key one. Can we attribute the failure to the evisceration of FEMA under the Bush administration? Did its enfeeblement also enfeeble the response of the National Guard units, the military when it was called in, and local and state agencies? Or was it the size of the hurricane? We have some data points for comparison. The response to Rita, arriving a few days after Katrina, has been declared much better by some news stories (Hsu and Hendrix 2005; Block and Gold 2005), and almost as bad by others. (Staff 2005f) Rita should have been easier. It was less destructive; citizens were more likely to evacuate early based on the experience with Katrina; no major cities were hit; top FEMA officials would be unlikely to again be unable to get the president’s attention; and state guards and the military were already mobilized. The failures in Rita were not encouraging.

Another data point is the response of FEMA to the four hurricanes that hit Florida in 2004. The only really critical news stories about the response refer to the large amounts of money distributed to areas that suffered no hurricane damage, widely attributed to Republicans’ interest in keeping that state in its column. FEMA approved payments in excess of $31 million to Florida residents who were unaffected by the 2004 hurricanes, for example. (Leopold 2005) (Staff 2005a) (As noted in chapter 3, political scientists have found that nearly half of all disaster relief is delivered on a political basis rather than by need.) But FEMA was actually blocked from playing more than a role as a resource provider in the state, on the order of Governor Jeb Bush. Florida had more experience with hurricanes than Louisiana, was wealthy enough, and was well-connected enough to fund extensive programs and exercises to cope with familiar disasters; and while four hurricanes in a row was unprecedented, none were of the magnitude of Katrina. As noted, the hurricanes occurred just before the critical national presidential election, and FEMA made sure that it was ready. One possible explanation for the failure is the indifference of the White House as the accurately forecasted hurricane headed toward New Orleans and swept through the three states. The president was on a golfing vacation, was briefed on the danger in a conference call, but assumed all would be taken care of in the biggest threat to a large city in recent memory. None of the cabinet members changed their routines. Michael Chertoff, the head of the DHS, saw no need to declare a federal emergency until days after the storm hit; and the head of FEMA said his staff did not inform him about the deplorable conditions in the convention center, even though some of them must have been watching, along with the rest of the nation, the endless TV news coverage. It is hard to explain this indifference, especially in what is said to be a very politically astute administration that could capitalize on a disaster with a vigorous response.

One explanation could simply be the disarray in the DHS in general—a department the administration did not want, but when forced on it, used it to reward political loyalists, privatize government, defeat civil service, and give out contracts to business friends. But there was more than disarray; the uses to which it was put shaped its response. The DHS had been fabricated in haste and neglected after its launch. But it also reflected deeper political values. As noted earlier, Brown shared the administration’s view that less government was better than more. It was “self-evident” to Brown that the government was not to provide services but only to see that the private sector provided them. FEMA would not be proactive in the face of the predicted disaster—an assertive role for government interference in the private sector—but would wait until help was properly requested (except in politically sensitive Florida). With that view of the mission of FEMA at the top, it is hardly so surprising that urgency was lacking at the bottom.

There was also more than an expression of conservative political values at work. One of the most extraordinary things about this disaster is the extent to which people at the lower levels of the responding organizations simply did not use common sense. Neither panic nor being “overwhelmed” will answer the following questions: Why did they insist upon full documentation when this was not possible because the fax machines were not working? Why did they force qualified medical personnel to stand aside because they could not legally engage in medical care in Louisiana? Why did they insist that air marshals had to be found to fly on the airplanes that carried out desperate evacuees? Why did they insist on time-consuming searches because the electronic monitors had no power? Why did they remove tins of emergency food from the evacuees because of regulations that were irrelevant to this emergency? Why did they not allow empty buses to pick up refugees? Why did they not require that National Guard trucks idly driving by thirsty evacuees carry water with them and distribute it? And the list goes on and on.

These are not instances that can be explained by the overwhelming nature of the disaster; the officials could have behaved differently, and the force of the storm or the destruction is irrelevant. They also go well beyond the issue of “prosaic failures” that all organizations are subject to. (Clarke and Perrow 1996) These behaviors do not involve panic, enormous overload, or unfamiliar tasks or settings, conditions that usually account for failures in unprecedented events. They involved going by the rules. Rather than being flexible and innovative, which is possible even when the challenge is overwhelming, these personnel appeared to revert to rote training, insistence on following inappropriate rules, and an unusual fear of acting without official permission. This is what needs explanation.

I would suggest that as the top ranks of the agency lost experienced personnel with high morale and commitment, and were replaced by political appointments with no professional experience in emergency management, 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 had increased the number of political appointees for government agencies by 15 percent between 2000 and 2004. (Writers 2005a) (In President Clinton’s second term, the percentage of political appointments declined.) FEMA has always had political appointees; most agencies do and some political appointees may even have experience in their field. But it was widely believed that the number of inexperienced political appointees rose dramatically during the Bush administration. Even if the increase in FEMA was only the average 15 percent, it would have an impact. The problem continues as of April 2006. Looking for a replacement to head FEMA, several people with extensive experience were recruited but all of them turned the Department of Homeland Security down. Most of them cited the failure of the White House to establish clear goals and a clear role for FEMA within the DHS, including ambiguity about the importance of natural disasters as compared to terrorist threats. (Lipton 2006f)

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, sticking to the rules to protect your career may be better than breaking them even if the rules are inappropriate. This defensive posture might spread to allied agencies, such as the Transportation Security Administration, which is already less concerned with safe transit than terrorists’ potential to use transportation as a weapon. A hypothetical situation could prompt this question: 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 fax machines are out of order. The message may be that in perilous times it is best to go by the book. (While not unreasonable, this is not substantiated by research that I am aware of). 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. The centralization would reinforce a tendency to go by the rules even if the situation suggests they are inappropriate. Maximizing the ability of the lowest level to extemporize and innovate will minimize the bureaucratic responses that so characterized FEMA. A frequent criticism of the reorganized 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) (Even worse is a centralized organization with leaders that ignore the call to arms.)

The failures of the National Guard are harder to account for. The head of the National Guard Engineer Battalion hiding in the convention center ordered, and got, more ammunition and barricaded the entrances. He admitted they could have gotten the center under control, if so ordered, but senior commanders ruled out that possibility. (Haygood and Tyson 2005) So they left in the night. Were senior commanders taking cues from the top and deciding not to bend the rules? Could it be that in the absence of an energized response from the White House, the heads of state guard units and of the military in Washington did not feel they should tell their units that this was a disaster of such magnitude that they should use their best local judgment? The heads of military units near New Orleans awaited orders to help that never came, as did a naval hospital ship in the gulf. How can we explain the many cases elsewhere during the crisis where the guard and the regular military performed well? Were they cut off from their commanders, and this allowed them to innovate and respond?

This account of the failings of the DHS in the instance of the Katrina Hurricane relies heavily on the characteristics of the Bush administration. But if the United States is to protect itself from devastating hurricanes, we must envision the possibilities of having more administrations such as this one. Just as organizations must be structured such that even with leadership that is just average (and even below average, as one cannot rely on having above-average leadership any more than the fictional town of Lake Wobegon can expect only above-average children), so must we prepare for organizational failures. The best preparation for hurricanes is to prevent the concentration of populations in risky areas. Even under the best of administrations, the devastations from a storm such as Katrina will be enormous.

THE DHS AND INTELLIGENCE

Most commentators see the biggest failing of the largest reorganization of the government in recent times—the creation of the DHS—as the failure to connect it with the intelligence agencies. The hub of DHS’s dot-connecting efforts was to be a new intelligence center for tracking terrorists. Just four days after Ridge was sworn in as the first secretary of homeland security, he found that DHS would not control this center. “Ridge and his aides thought the center was one of the key reasons the department had been created, to prevent the coordination failures that helped produce Sept. 11. Not only had the White House undercut Ridge, it also let him find out about his defeat on television.” (Glasser and Grunwald 2005) The importance of this decision cannot be overemphasized.

The new agency was no match for the agencies in the intelligence community (IC). The Department of Homeland Security only managed to get one office of the FBI (the National Infrastructure Protection Center) and assurances that a few DHS members could sit in on the coordinating committees in the IC. It can ask for information but has no assurance it will get it. Since intelligence is critical for security, for deciding where to put resources, for information on what kind of threat is likely, for alerts that a threat is imminent, for knowledge of the strengths and weaknesses of terrorist groups, and so on, the DHS is almost totally dependent on an intelligence system that is not decentralized but fragmented. Of course, many things are obvious: cockpit doors of aircraft must be hardened; container ports are vulnerable, as are national landmarks and nuclear power plants and chemical plants, among others. But intelligence is needed to decide how much money and effort should go to each type of target, since perhaps only a third of the funds needed to do the most obvious things are available. What are terrorists most likely to attack?

Even when the DHS gets warnings from the IC it has had trouble communicating with its own agencies. Both federal and state agencies said they were informed of “orange alerts” only by watching CNN, not through notification by the DHS, which issued them. And some governors and mayors refused to respond to the orange alerts since they were so vague and response was so expensive. The DHS appears to find it difficult to be responsive itself. Testing the capabilities of its police force, the U.S. Park Police deliberately left a suspicious black bag on the grounds of the Washington Monument. The police failed to respond quickly or effectively. One officer reportedly was caught sleeping. When a test official called the Department of Homeland Security to warn them about the bag, he got this priceless recording from our protectors: “Due to the high level of interest in the new department, all of our lines are busy. However, your call is important to us and we encourage you to call back soon.” (Shernkman 2004)

Should the DHS have gotten more control over those security agencies that are not clearly related to military strategy and battlefield tactics? (The bulk of the estimated $40 billion spent yearly on security is military-related.) Should it have at least gotten the FBI, which is primarily concerned with domestic security, though it does operate abroad to some extent? Aside from the problem of increasing the sheer size of the DHS even more, the consensus is that the security agencies were far too powerful for even parts of them to be moved. As Amy Zegart argues, agencies concerned with foreign affairs, such as the intelligence agencies, are oriented toward the president, rather than Congress, and controlled by the president to a greater degree than domestic agencies. (Zegart 1999) The interests of the president and of the intelligence agencies thus were “aligned,” as political scientists put it. Even if it were wise to give the nonmilitary intelligence agencies to the DHS, that was not likely to happen if the DHS could not even get control of the FBI. Worse yet, while DHS had the statutory responsibility for establishing a common watch list out of the twelve that existed, even this task was given to the Justice Department, over the objections of Clark Kent Ervin, its beleaguered inspector general. Ervin has a graphic chapter on the DHS’s failure to do the main job it was set up to do: coordinating disparate intelligence efforts. (Ervin 2006; Mintz 2004)

Since the DHS was unable to coordinate the efforts of the twenty-two diverse agencies over which it was given nominal control, it was highly unlikely to be able to coordinate the efforts of those agencies over which it had no control. Regardless of the structural reasons that would seem to make it logical for the agency to incorporate the FBI, it is not a happy thought. Quality leadership at the top of the DHS was not available, and the insular FBI would resist the kind of incorporation that would guarantee its cooperation with DHS headquarters: most of its focus remained on drug interdiction and nonterrorist criminal activity; it would have little or nothing to do with natural and industrial disasters; and it had powerful friends and constituencies in Congress and law enforcement agencies that would resist changes. But coordination of intelligence outside of the DHS appears to be just as difficult.

When the 9/11 Commission released its report in the summer of 2004 (Staff 2004), it triggered another burst of government reorganization. The commission recommended a radical change in the intelligence community. It was to be headed by an intelligence director with cabinet-level status and the authority to determine the budgets and key personnel of all of the fifteen agencies that made up the IC. The commission was on well-trodden ground with its recommendation. In just the past ten years there have been thirteen major studies and reports concerning our national intelligence system. They all recommended reorganizations, particularly to centralize controls over the disparate activities. Political scientist Thomas H. Hammond asks, “Why is the intelligence community so difficult to redesign?” (Hammond 2004) Bureaucratic politics and power of the sort we have been examining play a role, he admits, but there are more basic structural reasons.

Hammond argues that because of dilemmas that are inherent in any structural set up, any reorganization plans are bound to have enough faults in them to prevent any agreement on basic changes. For example, the intelligence community both collects information and integrates and disseminates it. A structure that is good for collection may be poor for integration and dissemination, and vice versa. Furthermore, a structure that favors rapidly acting on intelligence in any situation short of an imminent attack—say, acting on the August 2001 warnings and information about flight schools, etcetera—has costs. It may disrupt the source of information and prevent further surveillance that could identify more terrorists and their organizations. (This is the classic tension between the FBI and the CIA.) Finally, a structure that is appropriate for dealing with one kind of threat, for example, state-sponsored terrorism, will not be appropriate for another kind of threat, such as that presented by Al Qaeda. A major criticism of the Bush administration’s handling of terrorism from 2000 to 2001 was that it was still preoccupied with state-sponsored threats from North Korea, Iran, Syria, and even to some extent Russia, whereas the IC should have been reorganized to deal with the mounting threat of Islamic terrorist organizations.

But neither structure might be appropriate for domestic terrorism as represented by the Oklahoma City bombing, by leaderless groups, or by individuals loosely connected for an action on abortion clinics, power grids, logging operation, or animal rights—we have had terrorists in all of these—and then dissolving. All three forms of terrorism, state-sponsored, foreign jihadists, and domestics, are still present dangers, but we can hardly have three separate structures to deal with them. The appearance of non-statesponsored groups like Al Qaeda has not meant the disappearance of state-sponsored terrorism or even of anthrax mailings. So how should the structure be oriented?

The complex trade-offs required have produced a kind of structural conservatism on the part of intelligence policy makers, Hammond argues. No alternative structure has seemed clearly superior to the present one. And of course the costs of tearing organizations apart and disrupting career paths are substantial.

Hammond illustrates the difficulties when he outlines six functions and policy areas of the intelligence community, and they give pause to any easy solution. Recognition of the variety (and possible incompatibility) of these functions and policy areas does not seem to have been addressed by the December 2004 intelligence reform bill signed by the president. They are:

 

• Determining the intentions and monitoring the capabilities of the former Soviet Union, China, North Korea, Iran, etcetera

• Monitoring the nuclear proliferation technologies, capabilities, and intentions of foreign state and nonstate organizations

• Counteracting terrorism at home and abroad

• Providing intelligence support for antidrug campaigns

• Providing intelligence support for U.S. government policy in Iraq

• Supporting U.S. combat operations

 

The intelligence reorganization of December 2004 primarily involved the creation of a new director of intelligence that has budgetary control over all fifteen agencies. This prompted strong opposition from the Pentagon, which was eventually defeated when the public outcry finally brought the reluctant president to force a congressional committee chairman allied to the Pentagon to back down. (Another chairman made the issue of illegal immigrants hostage to the bill but only partially succeeded.) The details of the reorganization and budgetary changes were “still to be determined” at the end of 2004, and as of this writing, Spring 2006, it is far from clear what will emerge from the rather vague legislation. It does not appear to address the dilemmas that Hammond identifies.

Senator Robert Byrd (D-WV), one of only two members of the Senate to vote against the bill, delivered a scathing speech dealing with the limitations of the bill. Congress in general wished to give more powers to the head of the new agency, but the Pentagon and other interests got the powers watered down. However, Byrd was primarily concerned about the hasty passage of a long, complicated bill whose latest version the Senate had only twenty-four hours to review: the secrecy the bill provided to the new agency, closing off ombudsman reviews and whistle-blower protection; the failure to deal with prison scandals associated with intelligence interrogation and the successful attempt to limit inquiries into possible prison abuses; the change from mandates to promises regarding new resources; and the successful attempt to reinstate and reinforce powers under the Patriot Act that Congress wanted reconsidered, parts of which the courts had thrown out. The intelligence bill was described by civil liberties groups as a Trojan horse, using the opportunity for reform of intelligence failures to greatly weaken civil liberties. (See, for example, Eggan 2004.) Senator Byrd’s remarks touch on many of the organizational issues we have been dealing with. It is a sobering litany of why we should not expect much from the government organizations that are supposed to protect us and the uses to which our organizational efforts might actually be put. (Byrd 2004)

An impressive lineup of experts argued in vain against hasty consideration. Byrd pointed to the following flaws, among others: the thoroughly politically partisan Office of Management and Budget can screen intelligence testimony before it is presented to Congress; the requirement for inspector general and ombudsman positions was removed, and it was left to the discretion of the intelligence director whether to appoint them and what their powers would be; an independent Civil Liberties Board, recommended by the 9/11 Commission, was made dependent on the Office of the President; the Foreign Intelligence Surveillance Act was being modified to give the executive branch power to undertake electronic surveillance, allowing the president to monitor domestic telephone calls.

The problem of congressional oversight remained after the reorganization of intelligence. An attempt to establish a single super-committee in each branch of Congress fell prey to the interests of existing committee chairmen. The chairman of the Rules Committee of the House complained that giving jurisdiction over the Transportation Security Administration (newly established) and the border control to a supercommittee left him with “scars.” “I will be dining alone,” he said. Another representative, Curt Weldon (R-PA), said, “but when you read the legislative language, it guts all the authority,” leaving it in the president’s hands, which was one of Senator Byrd’s points. Another Republican representative said: “I think we’re fighting tonight for the soul of Congress. It’s turf battles, it’s people who want to go back to September 10” in terms of congressional oversight. (Kady 2005)

I think that the most important factor will be the intentions of the president. He has appointed a head that presumably shares his political vision, but the Pentagon is suspected of starting its own domestic intelligence agency that will be out of reach of the new director. Structural changes are needed, and one may rejoice that a terrorist threat integration center will be a substantial part of the new agency. But the new head of the IC is unlikely to solve the dilemmas, even with a new terrorist threat center. A February 2006 news story reported that lawmakers were worried that the director of national intelligence, John Negroponte, had not moved quickly enough to establish his leadership and had not been able to exert effective control over the Pentagon. It was also feared that rather than being a lean operation, the agency was becoming another bureaucratic layer. (Pincus 2006) A terrorist threat integration center established in the Clinton administration was moderately successful; the same center in the Bush administration was not. A very great deal depends on executive leadership, much more than structural reorganizations, as important as they may be. It is an observation I have made in connection with the diversion of the goals of FEMA in the Reagan administration, and will see again with the Millstone nuclear plant failures and others in future chapters.

DREARY CONCLUSIONS

There is no doubt in my mind that the nation is somewhat safer since the 9/11 attack. Suspects have been apprehended, the Federal Aviation Administration has made changes, so has Immigration and Customs Enforcement. But the first two improvements were made outside of the new department of Homeland Security, and the third easily could have been without its appearance. As we shall see in a later chapter, the department has had very limited success in making our chemical and nuclear piles of vulnerability more secure. Our borders are still so porous that it would be sheer luck if a guard happened onto a terrorist. A few of the thousands of containers that daily enter our ports are said to be under some surveillance, and the department has been active there but has been thwarted by large shippers, principally Wal-Mart, which has given $191,500 to current House Homeland Security Committee members since 2000 (one of the advantages of being on such a committee). One of Wal-Mart’s friends, the National Customs Brokers and Freight Forwarders Association, put it bluntly: “The private sector needs to continue to get emerging government figures to swear on a stack of Bibles that commercial operations are an important responsibility that cannot be subordinated wholly to security interests.” (AFL-CIO 2006) Barry Lynn provides a detailed account of Wal-Mart’s opposition to strong port-security legislation. (Lynn 2006) But the new surveillance (and more breaches of basic privacy, unfortunately) of populations that might harbor terrorists is handled by Justice. Billions have been spent to improve intelligence and first responder capabilities, but intelligence funding is outside of the Department of Homeland Security. That does not leave us with much to be grateful from the department.

And we have no idea how many more billions would need to be spent, and where to spend them, in order to close all the holes in our open society. I think it is foolish to think our society will ever be safe from determined terrorists, but it is possible that we have raised the bar just enough to make it a bit more difficult for them, and this may be at least a small part of the explanation as to why we have not been successfully attacked on our soil since September 11, 2001—more than five years at this time of writing. (Better explanations for the hiatus on attacks will be explored in the last chapter. In essence, they argue that the United States has been shown to be vulnerable, and that is enough. There is room for small attacks, of course, but more pressing concerns for the Islamic extremists are getting “infidel” troops out of Islamic nations and destabilizing Islamic regimes that are “corrupted” and shaky by driving out all infidels and installing fundamentalist regimes.)

Our efforts have, in some cases, made us safer from our two other sources of disaster, natural and industrial ones. Certainly, strengthening first-responder capabilities will often mitigate natural and industrial disasters. We need better intelligence for these types of incidents as well as for terrorist attacks, improved medical response to epidemics as well as biological attacks, and training and simulated emergencies for all three threats. But there is a danger in the “all hazards” justification for the structure of the DHS. The department focuses on terrorism, and the most expensive parts of its program will have little to do with industrial or natural disasters: upgrades for police departments are favored over those for fire departments; coping with biochemical attacks receives substantial funding, but not coping with epidemics, which are much more likely to occur and more devastating; intelligence and surveillance activities (watch lists, border security, cameras in subways and on streets, surveillance of antiwar groups and mosques, etc.) are extensively funded but have nothing to do with natural disasters. No one has done an accounting that I am aware of, but I suspect that most of the money tries to protects us from only one of our three disaster sources.

But we have a porous society, less protected (and less inconvenienced) than those of our European allies and Israel. A few suicide bombers coordinated to blow up tunnels, bridges, and airports would panic our government, and domestic or foreign terrorists are capable of shutting down the Northeast power grid for weeks with a few well-placed, small explosions. Suitcase bombs in a chemical plant (which are still unprotected, as we shall see in Chapter 6) could put seven million people at risk if the weather cooperated. A drive-by attack on the spent-fuel cooling tank at a nuclear power station would be harder to pull off but quite possible and could release more radiation than is held in the core in minutes. These are targets of our own making, made large and vulnerable for reasons of small economies and unwillingness to have a few inconveniences. Very little is being done, and little can be done, about all three threats, other than basic reductions of our vulnerabilities. We are unprepared, and should not be prepared, to have perhaps 15 percent of our employed population engaged in protecting our infrastructure. With our form of federalized government and the problems of campaign financing and corporate power, government can only muddle through its large and its small problems, as Charles Lindblom argued almost fifty years ago, including its new problem of a Department of Homeland Security. (Lindblom 1959) Still, it is possible that the criteria by which I and others have judged the first five years of the department may set too high a bar. Perhaps the bar should be “Have we made terrorists, even suicidal ones, pause just a bit, maybe for five years, before trying more attacks?”

Nevertheless, it is useful to analyze the supposed failures of the DHS, as well as its troubled origins and those of the intelligence community. We can always do a bit better. And through this analysis, we can better understand what we are up against when we organize and reorganize. It makes the plea to reduce our basic vulnerabilities all the more compelling. Such a reduction would also be a very difficult project, perhaps more difficult than (and almost as improbable as) a reasonable defense against terrorism, but I think not. As we proceed through other disaster-prone aspects of our society in this book, I will slowly build the case for such a reduction.

¹ An earlier version of this appeared in Perrow 2006a.