By John Mueller
and Mark G. Stewart
In seeking to evaluate the effectiveness of the massive increases in homeland security expenditures since the terrorist attacks of Sept. 11, 2001, the common and urgent query has been, “Are we safer?” This, however, is the wrong question. Of course we are “safer”—the posting of a single security guard at one building’s entrance enhances safety, however microscopically. The correct question is, “Are the gains in security worth the funds expended?”
Is the Risk Worth the Cost?
The key fact is this: At present rates (and including 9/11 in the count), the likelihood a resident of the United States will perish at the hands of a terrorist is 1 in 3.5 million per
year. And the key question, one almost never broached, is this: How much should we be willing to pay to make that likelihood even lower?
We have, in fact, paid — or been willing to pay — a lot. In the years immediately following the terrorist attacks of Sept. 11, 2001 on Washington and New York, it was understandable that there was a tendency to fashion policy and to expend funds in haste and confusion, and maybe even hysteria, on homeland security. After all, intelligence was estimating at the time that there were as many as 5,000 al‑Qaeda operatives at large in the country, and as New York Mayor Rudy Giuliani reflected later, “Anybody, any one of these security experts, including myself, would have told you on September 11, 2001, we’re looking at dozens and dozens and multi-years of attacks like this.”
What IS the Cost Anyway?
The intelligence claims and the anxieties of Giuliani and other “security experts” have clearly proved, putting it mildly, to be unjustified. In the frantic interim, however, the U.S. government increased its expenditures for dealing with terrorism massively. At the 10th anniversary of 9/11, federal expenditures on domestic homeland security have increased by some $360 billion over those in place in 2001. Moreover, federal national intelligence expenditures aimed at defeating terrorists at home and abroad have gone up by $110 billion, while state, local, and private sector expenditures have increased by $200 billion more. And the vast majority of this increase, of course, has been driven by much heightened fears of terrorism, not by growing concerns about other hazards.
Tallying all these expenditures and adding in opportunity costs — but leaving out the costs of the terrorism-related (or terrorism-determined) wars in Iraq and Afghanistan and quite a few other items that might be included — the increase in expenditures on domestic homeland security over the decade exceeds $1 trillion. See the following table.
This has not been enough to move the country into bankruptcy, Osama Bin Laden’s stated goal after 9/11, but it clearly adds up to real money, even by Washington standards.
In our book, we apply conventional cost-benefit and risk analytic approaches to this massive increase in expenditures in an effort to provide an answer to the key question. These approaches have been recommended for many years by the United States Office of Management and Budget, and they are routinely used by such agencies as the Nuclear Regulatory Commission, the Environmental Protection Agency, and the Federal Aviation Administration. In 2004 the 9/11 Commission specifically called on the government to apply these approaches to assess the risks and cost-effectiveness of security measures put in place to deal with terrorism. However, it appears that this simply has not been done.
We Don’t Really Know
Upon taking office in 2005, Department of Homeland Security Secretary Michael Chertoff did strongly advocate a risk-based approach, insisting that the department “must base its work on priorities driven by risk.” Yet, a year later, when DHS expenditures had increased by some $135 billion beyond those already in place in 2001 and when the department had become the government’s largest non-military bureaucracy, one of its senior economists wistfully noted, “We really don’t know a whole lot about the overall costs and benefits of homeland security.” By 2007, RAND President James Thomson was contending that most DHS programs are implemented, “with little or no evaluation” of their performance or effectiveness, and the agency “receives little analytical advice on issues of policy, program, and budget.” And, after an exhaustive assessment, the Congressional Research Service concluded at the same time that DHS simply could not answer the “central question” about the “rate of return, as defined by quantifiable and empirical risk reductions” on its expenditure.
Indeed, at times DHS has ignored specific calls by other government agencies to conduct risk assessments. In 2010, the department began deploying full-body scanners at airports, a technology that will cost $1.2 billion per year. The Government Accountability Office specifically declared conducting a cost-benefit analysis of this new technology to be “important.” As far as we can see, no such study was conducted. Or there was GAO’s request that DHS conduct a full cost-benefit analysis of the extremely costly process of scanning 100 percent of U.S.-bound containers. To do such an analysis would require the dedicated work of a few skilled analysts for a few months or possibly a year. Yet, DHS replied that, although it agreed that such a study would help to “frame the discussion and better inform Congress,” to actually carry it out “would place significant burdens on agency resources.”
Clearly, DHS focuses all or almost all of its analyses on the contemplation of the consequences of a terrorist attack while substantially ignoring the equally important likelihood component of risk assessment as well as the key issue of risk reduction. Moreover, we have been able to find no reference whatever to the likelihood of a terrorist attack beyond rather vague references such as “high,” “imminent,” “dynamic,” “persistent,” and “emerging.” In general, risk assessment seems to be simply a process of identifying a potential source of harm and then trying to do something about it without evaluating whether the new measures reduce risk sufficiently to justify their costs.
Risk Analysis Not Done
This conclusion was strongly supported by a 2010 report of the National Academy of Science. Requested by Congress to assess the activities of the Department of Homeland Security, a committee worked for nearly two years on the project and came up with some striking conclusions. Except for the analysis of natural disasters, the committee “did not find any DHS risk analysis capabilities and methods that are yet adequate for supporting DHS decision making,” and observed that “little effective attention was paid to the features of the risk problem that are fundamental.” Therefore, it concluded, “only low confidence should be placed in most of the risk analyses conducted by DHS.” It also found an “absence of documentation” with the result that it sometimes had to infer details about DHS risk modeling. Indeed, in some cases it was unable to determine what problem was being addressed. It also found “a pattern” of “trusting numbers that are highly uncertain.” And, concluded the committee rather glumly, “it is not yet clear that DHS is on a trajectory for development of methods and capability that is sufficient to ensure reliable risk analyses”: although it found that “there are people at DHS who are aware of these current limitations,” it “did not hear of efforts to remedy them.”
Security Trumps Economics
Overall, it seems, security concerns that happen to rise to the top of the agenda are serviced without much in the way of full evaluation — security trumps economics, as one insider puts it — and such key issues as acceptable risk are rarely discussed while extravagant worst-case scenario thinking dominates, and frequently savagely distorts, the discussion.
It is clearly time to examine massive homeland security expenditures in a careful and systematic way, applying the kind of analytic risk