Michael Chertoff, the new secretary of homeland security, is studying ways to revise the color-coded terrorism alert system created three years ago. He should save himself the trouble and scrap it altogether.
The stated objective of the five-tier advisory system is to increase preparedness among public safety officials, businesses and citizens when there is a threat of a terrorist attack. The reality, however, is that the warning system ends up alienating the very people whose cooperation is essential to managing real danger. If the color-coded system were to meet a timely demise, the only ones to miss it would be comedians on late-night television.
The multicolored advisory was developed in the aftermath of the 9/11 attacks by a harried Bush administration casting about for ways to safeguard a nation of almost unlimited soft targets. Since the system's inception, the threat level has never dropped below the third tier, yellow, or ''elevated.'' On six occasions, the government has raised the threat level to orange, or ''high,'' most recently last August.
There are two fatal flaws with the system. First, the federal government lacks the ability to collect and analyze the intelligence that can produce a timely judgment on where the threat level should be set. As the 9/11 commission and a number of blue-ribbon reports have documented, our intelligence apparatus is badly broken. It will probably be a decade or more before the C.I.A. and F.B.I. have the operatives and informants in place to track an enemy as diffuse as Al Qaeda. In the meantime, the government will not be able to provide adequate warning about the next attack.
Second, it is unreasonable to impose the costs of the upgraded security precautions associated with these alerts on governments and businesses across the country when terrorists cannot strike everywhere at once. The combined total of 11 weeks spent at orange alert since 9/11 have cost cities an estimated $750 million in stepped-up security measures (and that's just the public outlays, not the expenses borne by companies and individuals). The truth is that some targets are much more attractive than others. A nationwide threat system makes no sense in a huge country where the level of risk varies wildly from place to place.
Such broad warnings have another unintended effect. If threats do not materialize, politicians start to look like a version of Chicken Little. When a real emergency arises, citizens may end up second-guessing official guidance just when they need it most.
The multicolored alert system wrongly treats security as something that can be turned on and off like a light switch. The alternative is to start managing security like safety -- something you practice as a matter of routine. When it comes to safety, we require architects and engineers to anticipate how an act of God, human error, or mechanical failure may affect what they are building. We don't expect them to eliminate every risk, but we do insist on appropriate safeguards.
Just as we tailor preparedness for hurricanes, tornadoes and earthquakes to the areas that are most susceptible to them, we need to concentrate our security efforts on the places that are most at risk of a terrorist attack. These tend to be in major urban areas like New York, Chicago, Houston and Los Angeles. Local governments should decide their own levels of readiness. On those rare occasions when the federal government has specific threat information, it should provide the intelligence to local leaders but leave the decision up to them on how they want to respond.
There are already models for this kind of approach. For instance, the Coast Guard has organized a maritime-area security committee in every major port city. These committees are responsible for drawing up security and response plans and setting a three-tier alert level. So Seattle may decide to be at Level 1, for example, while Charleston places itself at Level 2. Each community decides how to bear these costs.
Bolstering our preparedness in the face of a potential terrorist attack will require a sustained national effort. The erratic nature of the color-coded alert system is undermining that effort. Secretary Chertoff shouldn't tinker with something that's beyond repair. Instead, he should eliminate this well-intentioned but ultimately flawed tool for addressing our post-9/11 sense of national vulnerability.
Stephen E. Flynn, a senior fellow at the Council on Foreign Relations, is a former consultant to the United States Commission on National Security.