Wednesday, September 7, 2011

Air Passenger Security: Tradeoffs in the Decade after 9 /11

K. Jack Riley of RAND has been thinking over the time and cost tradeoffs of airline passenger security. A short overview is here; a chapter-length discussion can be found in the RAND book, The Long Shadow of 9/11: America’s Response to Terrorism, which can be read as an e-book here. Riley points out how the costs of air security could be reduced--and how doing so could save hundreds or thousands of American lives each year by encouraging people to fly rather than drive. Here are a few excerpts (with footnotes deleted) from the book chapter:

The overall safety of air travel since 9/11:
"In the ten years since 9/11, about 6 billion enplanements (instances of passengers boarding a commercial plane) have occurred in the United States, and perhaps an additional 14 billion have occurred worldwide. Among those 20 billion passengers, according to the Aviation Safety Network, 7,019 died in aviation accidents between 2001 and 2009. In addition, roughly 200 more have died on airplanes or at airports as a result of terrorism since 9/11.... And since 9/11, no passengers have died in terrorist acts from enplanements originating in the United States. In short, despite the tragedy and loss of life on 9/11, air transportation is overwhelmingly a secure means of transportation, especially in the United States. At least three factors contribute substantially to the improved security since 9/11. First, and perhaps most importantly, passengers know now that they must be vigilant. ... Second, airlines have reinforced cockpit doors in a way that strictly limits access to the cockpit. Most crews have also modified their procedures to ensure that there is a barrier between the cockpit door and
the passengers when the cockpit door needs to be opened. ... Third, changes to the visa approval process represent a relatively unheralded but important contribution to air transportation security.
All 19 of the terrorists involved in the 9/11 attacks were in the United States on legitimate visas."
What about inspecting air cargo and TSA employees?
"Until recently, though, the starkest contrast of all to the treatment of U.S. airline passengers was the treatment of U.S. commercial cargo on those same flights boarded by the passengers. It was not
until August 2010 that all of the commercial cargo loaded onto domestic passenger planes was scanned or searched. Shipments on cargo jets from international destinations are not all currently screened, although the date for implementing the plan for screening them has been moved
up from 2013 to 2011. Another departure from the “inspect everyone and everything” approach involves TSA’s own workforce at the airports. TSA employees are not screened when they enter the secure area of an airport throughout the course of the day, because they are trusted employees who have had a background check. They are thought to be at particularly low risk for coercion or conversion to radicalism. At many airports, certain other employees also have all-access badges that allow them to bypass security. Thus, what is considered the “sterile area” of the airport is in fact
not sterile. Substantial volumes of people (and, until recently, cargo) have made it into the sterile area without inspection. There have been no terrorist incidents associated with these leakages, however, suggesting that the cargo and employee risks have been appropriately managed for years (even before full cargo screening began) and begging the question of what kinds of risk management improvements might be available for passengers."

The case for scaling back security checks for planes originating in the United States:
"There is very little reason to be concerned about suicide bombers being present on flights originating in the United States. The security improvements noted above—passenger vigilance, cockpit security, and visa screening—go a long way toward preventing radical jihadists from entering the country or, having entered, from being able to commandeer a plane to conduct a spectacular attack. Moreover, the radical threat resident in and willing to conduct a suicide attack on the United States is extremely small. ... Recognizing the security of flights originating in the United States and thus returning all passengers to the domestic procedures that existed before the recent additions would save, at minimum, about $1.2 billion annually. ... It would also reduce the deadweight losses that domestic travelers incur from arriving at airports early, waiting in lines, and undergoing intensive scrutiny."

The case for a "trusted traveller" program:
"The current security regime applies the same procedures to all 700 million passengers who board planes each year in the United States. That we have not developed a reasonable way to reduce that inspection workload is perhaps the biggest missed opportunity of the past decade. A trusted traveler program could be configured in a variety of ways. Recent conversations with airline industry executives suggest that a very small fraction of fliers account for a very large proportion of trips. In all likelihood, then, a trusted traveler program could be relatively small (with 5 million enrollees or less) and could still provide significant benefits. No program will be bulletproof, but such a program does not need to be given the extremely low odds of encountering a suicide terrorist on a flight originating in the United States. A trusted  traveler program could initially be organized around these characteristics or combinations of characteristics:
• Possession of a security clearance issued by a U.S. government agency. ...
• A profile that involves frequent travel. An individual traveling
100,000 miles per year is, conservatively, spending 200 hours on airplanes a year. That is 10 percent of a standard 2,000-hour work year, suggesting that such travelers can be trusted with the basic screening that was in place prior to the deployment of WBI [whole body image] machines and pat-downs. ...
• Willingness to submit to the equivalent of a security-clearance process. Some travelers would find it well worth the time and expense to obtain such a credential in exchange for the ability to move
through an airport more quickly. Several programs, including Global Entry, NEXUS, and SENTRI, already allow certain travelers to be pre-approved for expedited clearance for entry at U.S. borders. Global Entry members pay a fee, undergo an interview and background check, and provide fingerprints as part of seeking approval. SENTRI and NEXUS operate in a similar fashion at Mexican and Canadian ports of entry, respectively. The combined programs cover hundreds of thousands of frequent travelers. The marginal costs of implementing this approach would be relatively low, since more than a million travelers have already paid for these entry/exit credentials. Extending the privileges of these programs from entry into the United States to security at U.S.airports would be a relatively trivial and easily justified action."

The opportunity cost of not taking such measures is measured in preventable deaths: 
"Researchers have estimated that the 9/11 attacks generated nearly 2,200 additional road traffic deaths in the United States through mid-2003 from a relative increase in driving and reduction in flying resulting from fear of additional terrorist attacks and associated reductions in the convenience of flying.20 If the new security measures are generating similar, or even smaller, substitutions and the driving risk has grown as hypothesized, the new methods could be contributing to more deaths  annually on U.S. roads than have been experienced cumulatively since 9/11 from terrorism against air transportation targets around the world."