Ten years ago, before 9/11 made terrorism our national preoccupation, the agencies that now make up the Department of Homeland Security spent about $22 billion a year on public safety and emergency management. Now Homeland Security is the third-biggest department in the federal government, with more than 230,000 employees and a budget of $55 billion a year.
Before 9/11, the United States spent about $30 billion a year on its civilian intelligence agencies; today, such spending has nearly doubled to about $55 billion, more than the entire State Department budget. Add in spending on military intelligence, and the intelligence budget comes to more than $80 billion.
How much additional security have we gained from all that spending? It's impossible to say.
As officials will assure us many times over the coming days, we're safer than we were on 9/11, and they're right. Al-Qaida's central leadership is crumbling. Its members still plot against the United States, but their plans have all been foiled so far. Only a handful of extremist attacks on U.S. territory have succeeded since 9/11, all small-scale actions by American citizens with guns, not international terrorists on airplanes.
Does that mean we can consider de-escalating our massive internal security campaign?
Homeland Security Secretary Janet Napolitano was asked that question recently, and she gave a crisp answer: "No."
"The threat against the United States," she said, is something "that we now have to deal with, I think, throughout the foreseeable future."
But growing numbers of experts and authorities are beginning to ask whether we are spending more than we need to.
"If you ramp up fast, and we felt we had to ramp up after 9/11... you're going to overspend," noted Thomas Kean, the former Republican governor of New Jersey who chaired the official commission on the lessons of 9/11.
"We didn't pay attention to costs," agreed the panel's vice chairman, former Rep. Lee H. Hamilton of Indiana, a Democrat.
Part of the problem was described last week in a Times article by my colleague Kim Murphy. Federal grants have funded hundreds of dubious projects deemed priorities by local law enforcement agencies, from cattle-moving equipment in Nebraska (in case of an attack on livestock) to an 8-foot fence around a veterans hospital in North Carolina (in case of an attack on ailing veterans).
But the issue is bigger than the pork-barrel spending that counter-terrorism grants make possible.
The massive expansion of the homeland security state began in the understandable panic immediately after 9/11, but it continued from there.
The public's desire for safety and Congress' desire to please have combined to make it easy for spending to increase but almost impossible for anyone to argue that it should decrease. Once a program is in place, no one wants to be responsible for killing it, for fear of being blamed in the event of an attack.
"If you sit down and start arguing with somebody about whether a given security step is necessary, you've got a tough burden of persuasion ... because they can give you 101 reasons why a particular step is important," Hamilton said last week.
As a result, we spend money on measures that we hope will make us safer without knowing for sure whether the added amount of safety is worth the extra cost.
Take the Transportation Security Administration's controversial program to install enhanced screening devices in the nation's airports. The advanced machines have drawn criticism from passengers concerned about privacy or radiation.
But the question of whether the scanners' cost provides sufficient benefit was never fully considered, according to the Government Accountability Office.
Most of us would probably favor installing those machines, no matter what a cost-benefit analysis said, because we want airline travel to be as safe as possible. But we're making that call on a gut level.
To take another example, Congress has passed a law that requires Homeland Security to screen 100 percent of the maritime cargo destined for U.S. ports by 2012, no matter what it is or where it's from. Napolitano has said the requirement is too broad and too costly to be worthwhile, and she's right. But Congress hasn't mustered the political will to change the law.
A few members of Congress from both parties have bravely argued that tougher scrutiny is needed, especially in view of the need to reduce the federal deficit. "We can get a better result for less money," Sen. Thomas R. Carper, D-Del., argued earlier this year. But so far, only a few have taken aim at this part of the budget.
There's no such thing as too much security. But there is such a thing as security that's too expensive, such as an 8-foot fence around a low-risk target.
Ten years after 9/11, it's time to face the fact that every risk can't be eliminated - and time to weigh the costs and benefits of security spending more openly.