In this issue:
- GAO on TSA Screening Technology
- Hub Airlines and Congestion Delays
- Bogus Arguments on Private Screening
- Gatwick Airport Sale and Other Privatizations
- Porter Airlines Breaks the Mold
- News Notes
- Quotable Quote
GAO Rips TSA Screening Technology Development
We all want to believe that better technology holds the key to easing our way through passenger screening checkpoints. But the way the Transportation Security Administration is going about doing the R&D, testing, and deployment of better technologies leaves a great deal to be desired. That's the message of the latest report on aviation security by the Government Accountability Office, released last month (GAO-10-128, October 2009).
One key finding is that after years and years of rhetoric about risk-based policy from both TSA and its parent agency, the Department of Homeland Security (DHS), "TSA's strategy does not incorporate some key risk management principles-a risk assessment, cost-benefit analysis, and performance measures." Those principles are required by DHS's National Infrastructure Protection Plan (NIPP). TSA responds blithely that it takes risks into account by analyzing threat information, but GAO points out that the NIPP requires a systematic process of doing risk analysis based on threat, vulnerability, and consequence assessments. TSA admits that it has not yet conducted cost-benefit analysis to set priorities for checkpoint screening or established performance measures for deployed technologies. But as the GAO report points out, without doing these things, "TSA cannot ensure that it is targeting the highest priority security needs at checkpoints, measure the extent to which deployed technologies reduce the risk of terrorist attacks, or make needed adjustments to its [checkpoint] strategy."
At the request of the senior members of Congress who called for this report, GAO used the infamous "puffer machines" as a case study in poor decision-making. TSA deployed 101 of these Explosive Trace Portals (ETPs) to airports in 2006, despite the fact that it had done no testing outside of laboratory conditions. They broke down frequently due to dirt and humidity in the airport environment (not to mention that they slowed down throughput compared with walk-through metal detectors). Most are no longer being used, some have been removed, and another 116 that were bought but never deployed remain in storage.
Going forward, GAO acknowledges that TSA, after repeated prodding, has completed a strategic plan for passenger checkpoints that includes goals and objectives-but it has not conducted a risk assessment as part of that process. It has not assessed the vulnerabilities of the checkpoint technologies currently in place, nor the tactics terrorists could use to bypass or spoof them. "TSA lacks a method to systematically test and identify vulnerabilities in its passenger and baggage screening equipment in an operational airport setting." It has not completed a cost-benefit analysis to help set risk-based priorities and guidelines for developing and selecting among new technologies. And it lacks measures to evaluate the extent to which its checkpoints reduce the risk of terrorist attacks.
A section toward the end of the report also makes dismaying reading. It summarizes DHS's response to the report's eight recommendations, all of which the agency said it agrees with. In most cases, however, the GAO states its concerns that the agency either doesn't really mean it or that the actions it says it will take will not (or not fully) address the intent of the recommendation. In two cases, GAO cannot provide a detailed explanation of its concerns, because "TSA determined our evaluation to be sensitive security information."
These are still early days of the Obama administration. I hope DHS Secretary Janet Napolitano and new TSA Administrator Erroll Southers will take this report seriously.
Do Hub Airlines "Internalize" Congestion Delays?
Although most economists who have studied delays at congested airports recommend some form of pricing to bring peak demands into conformity with runway capacity, the past decade has seen several studies claiming that airlines with a dominant position at hubs (e.g., American at DFW or Delta at Atlanta) "internalize" the resulting delays-i.e., accept them as a cost of doing business at the hub (while appreciating that the congestion also deters new entrants from competing there during peak periods). The most cited of these studies was by Jan Brueckner in 2002, who ended up proposing that airport congestion prices should therefore be lower for the dominant hub carrier than for others.
I only recently came across a 2005 paper by economists Katherine Harback and Joseph Daniel which provides a new empirical test of the internalization argument: "(When) Do Hub Airlines Internalize Their Self-Imposed Congestion Delays?" (University of Delaware Department of Economics, Working Paper No. 2005-08) They developed a model of airport congestion and used flight data from 27 major U.S. airports to test whether dominant airlines internalize such self-imposed congestion-- or ignore it. Most of their statistical tests reject the internalization hypothesis. Another test showed that "flights operating during typical large interchange banks at nearly all the highly congested airports do not internalize delays." These results apply to Atlanta, Charlotte, Washington National, Denver, Dallas, Detroit, Newark, Houston, JFK, Minneapolis/St. Paul, Chicago O'Hare, Pittsburgh, San Francisco, and St. Louis. Two notable omissions from this list are LaGuardia and Los Angeles, neither of which has a large dominant carrier, and both of which have high traffic in relation to capacity essentially all day long.
Harback and Daniel conclude with policy recommendations. They argue that any policy to reduce airport congestion should focus on both increased efficiency and increased competition. "Properly implemented, congestion pricing would improve flight connection times for airlines at their own hub airport, while imposing only minor scheduling delays (often less than 15 minutes) from their most preferred operating times at their non hub airports." Moreover, they write, "Hub airlines should support congestion pricing as a means of reducing self-imposed congestion while pricing other airlines out of their periods of peak bank operations." This approach, they maintain, should be considered pro-competitive because, "while it strengthens the local hub, it means that other airlines will be able to provide moe rapid connections at competing hubs. By improving connecting service, there would be more viable competition or potential competition in many origin-destination markets, putting downward pressure on fares."
If You Outsource Airport Screening, Watch Out for Bogus Arguments
Last month the airport board of Glacier Park International Airport in Montana voted to shift from TSA-provided passenger and bag screening to contract screening, under TSA's Screening Partnership Program (SPP). Airport director Cindi Martin had been complaining for years about inadequate screener staffing, especially during the tourist season. TSA has repeatedly reduced the airport's allocation of screeners, which it re-calculates once a year, based on passenger numbers in October (a quiet month at Glacier Park). Although TSA supplements the screener workforce in the summer as best it can via its National Deployment Officers (a kind of flying squad of screeners), that has not been enough to prevent long lines and passenger complaints.
Unfortunately, much of the news coverage of the airport's decision has focused on fears and complaints from the current TSA workforce. In particular, they cite a "GAO report" that found that screening by TSA-approved contractors costs 17% more than screening done by TSA itself. For example, screener Eric Wood was quoted as saying, "To try to increase and add an extra layer of bureaucracy, it seems ludicrous in this economy. Dollar for dollar, it's going to cost you [the taxpayer] more money, and you're not going to get a better product." Even airport director Martin accepts the alleged GAO finding of 17% higher cost, but thinks the reduced lines will be worth it.
The study they both are referring to says nothing of the sort. I reviewed the report (GAO-09-27R) back in February 2009 (Issue No. 42). What GAO did was to review two reports on the costs and performance of screening under the SPP versus TSA-provided screening, one by Catapult Consultants (hired by TSA to do an independent assessment) and the other by the TSA itself. The Catapult report (which has not been made public but is well-summarized by GAO) appears to have been well done. First, Catapult compared six SPP airports with a matched set of TSA airports. The reported costs were, indeed, 17.4% higher, on average, for the six SPP airports. To get a broader picture, Catapult created a regression model using data from all 450 airports with screening, and four years worth of data. That exercise showed 9% higher costs at SPP airports.
But that's not the end of the story. TSA evidently didn't like the Catapult report, because it found (as had previous outside studies) that "SPP airports' overall performance results are equal to or better than those delivered by non-SPP airports." It also noted that TSA assigns administrative and overhead costs at SPP airport that artificially inflate the reported cost of screening there. So TSA did not release the Catapult study; instead, it produced its own quick & dirty "study" using only one year of data (rather than Catapult's four).
What GAO did, in so many words, was to blow the whistle on this skullduggery, by summarizing and contrasting the Catapult and TSA reports. GAO noted that TSA's recorded costs of screening (used in both of those reports) omit such key items as workers comp, general liability insurance, and some retirement costs for TSA screeners that are not in TSA's own budget. And the cost comparison also neglects the revenue the federal government gets in income taxes from the SPP companies. And because TSA did not compare cost with performance, GAO said that "We believe that TSA should not use [their] study as sole support for major policy decisions regarding the SPP."
In short, the arguments being raised in opposition to Glacier Park's decision are bogus, and it's absurd to cite the well-done GAO report to argue against an airport's decision to go with the SPP option. Frankly, I think the airport trade groups-AAAE and ACI-NA-should have done a better job informing their members about the Security Partnership Program. Small-town airport directors like Cindi Martin should not have to rely on this newsletter to learn the facts about this useful program.
P.S.: Seven smaller Montana airports have already joined SPP this past summer, and both Butte and West Yellowstone have applications pending, in addition to Glacier Park.
Gatwick Sale Revives Airport Privatization
Early in 2009, when Chicago's attempt to lease Midway Airport failed for lack of financing, many observers pronounced the end of a short-lived era of infrastructure privatization, that had begun in 2005 with the lease of the Chicago Skyway. In fact, only last week a poorly researched article in USA Today was headlined "Privately Run Infrastructure Deals Dry Up." Only days before, the financial news media announced that Global Infrastructure Partners had purchased London Gatwick Airport from BAA for $2.47 billion. Doesn't sound very dried up to me!
GIP is a $5.64 billion infrastructure fund, one of nearly a hundred worldwide that have raised substantial sums (in excess of $100 billion) to purchase or long-term lease key infrastructure enterprises. GIP is backed by Credit Suisse and General Electric. It already owns the small but growing London City Airport, a U.S. natural gas pipeline, a British port, and a British waste-management company. It is also considering a bid for one of BAA's Scottish airports.
The fact that the Gatwick deal got done, and at a reasonable price, in today's financial markets is a good sign for other planned infrastructure deals-including other airport privatizations. I've reported in recent issues on announced plans for privatization of New Orleans' Louis Armstrong International and Rio de Janeiro's Galeao International Airport. Both of those planned deals look more credible today than they did a month ago, thanks to Gatwick.
And then there's Sheremetyevo, Russia's second-largest airport (in Moscow). Since the collapse of the USSR, it has lost ground to privately managed Domodedovo, the other main Moscow airport. But last month Russia's Transport Ministry announced plans to privatize Sheremetyevo in 2010, in a deal estimated to be worth €1.5 billion. The winning bidder will be expected to pay for building a third runway as well as terminal modernization projects. Even Kosovo has its principal airport on the market.
To be sure, some of these airports will be easier to finance than others-and some may not attract serious bids, depending in part on how fast credit markets recover. But it's good to see that investor interest in the airport sector is picking up again.
Porter Creates Important Niche Market
For several months I've been following with great interest start-up airline Porter Airlines. It began service just three years ago (October 2006), operating out of Toronto City Center Airport, a small airport adjacent to the Toronto central business district. Today it provides regional service to Montreal, Ottawa, Quebec City, Chicago, Boston, New York (Newark), Halifax, Thunder Bay, and St. Johns, using Bombardier Q400 twin turboprop aircraft. By next spring, its fleet will consist of 20 of these quiet, fuel-efficient planes. As a low-cost but high-service airline (free beer, wine, and snacks on board; coffee and newspapers at the airport), it has led to dramatic fare reductions on routes it has entered these past three years.
Porter is a wholly owned subsidiary of Porter Aviation Holdings, Inc., backed by Edgestone Capital Partners, Borealis Infrastructure, GE Asset Management, and Dancap Private Equity. The company has a 50-year history operating other regional airlines in Canada. It also provides the Fixed Base Operator service at Toronto City Center Airport.
The airport is owned by the Toronto Port Authority, and its continued existence as an airport was in question only a few years ago. Apparently, the city government had considered taking over the property and converting it to other uses, a move that would have been popular with a nearby residents' group called Community Air. That group has opposed-to no avail-Porter's currently-under-way project to expand the terminal to a 10-gate configuration with jet-bridges (which passengers will certainly appreciate during Toronto winters). The first phase is to open this fall, with completion of the $45 million project by next summer.
Porter CEO Robert Deluce told Airport Business (August 2009, p. 16) that the company has identified 17 additional destinations, all within a 500-mile range of Toronto City Center Airport, as possible service additions. Deluce better move fast, in my view, since Air Canada Jazz is considering moving into City Center. The airport's small size limits its capacity, and Porter reportedly already uses a large majority of all available slots.
Porter's success in identifying and serving regional markets-and from a downtown airport close to a huge concentration of business people-causes me to once again reflect on the hubris of those who want to take untold billions from general taxpayers to displace self-supporting short-haul airline service with heavily-subsidized inter-city rail. Other cities could do as Toronto has done, providing airport capacity tailored to short-haul, turboprop (perhaps STOL?) service, with entrepreneurial airlines paying for terminal space and runway access. And because of the flexibility of air service, decisions about which cities to serve would be made by individual carriers, responding to the market demands their efforts would discover. Contrast that with the hugely politicized process of deciding which cities will be on the route of a tax-funded rail service-and the "sunk" nature of the cost of the rail infrastructure if those decisions turn out to be flawed.
My hat's off to Porter-and to the enlightened Toronto Port Authority.
Should AIP Help Implement NextGen?
At the Air Traffic Control Association annual conference last month, the Colorado Dept. of Transportation (which funded a wide-area multilateration surveillance system that will dramatically increase runway throughput at airports in the mountains) argued that the federal Airport Improvement Program should fund such projects. And the Aviation Week Airports weekly for Oct. 27th quoted FAA associate administrator for airports Catherine Lang as agreeing. "What Colorado paid for was not AIP-eligible, but it's a fair question to ask, why not?" she told the newsletter. I agree; a lot of critical NextGen technology can increase runway throughput, and that seems likely to offer many project opportunities with higher ratios of benefits to costs than many current AIP grants.
Delta to Up-Gauge at LaGuardia
How times change! Two years ago as I was drafting Reason Foundation's proposal for congestion pricing at the congested New York airports, Delta was among the carriers denouncing the idea, and one of its arguments was that "up-gauging won't happen" because airlines needed numerous commuter planes to feed traffic to their longer-haul flights. But now that Delta has acquired 125 LGA slots from US Airways, "Delta plans to accomplish its expansion through larger jets," replacing turboprops operated by US Airways Express.
While I'm not a fan of airport slot controls (much preferring runway pricing to deal with congestion), when such controls exist, I support use-it-or-lose-it regulations to prevent incumbent carriers from hoarding unused slots in order to keep out competitors. Hence, I hope the EU Council of Transport Ministers continues to ignore pleas from the Association of European Airlines to extend the summer 2009 waiver of the 80% rule into the winter season and summer 2010. On the other hand, I can see the logic of the FAA agreeing last month to waive its use-it-or-lose-it rule at New York's JFK airport from March through November, while a major runway is taken out of service for a resurfacing and widening project. This will permit carriers at JFK to reduce their schedules during this period so as to prevent much worse delays and congestion.
Manchester Airport Exempts Kids from Body Scanning
Last month the UK's Manchester began a year-long trial of a checkpoint body scanning machine. A group called Action on Rights for Children objected, claiming that using the device on children would violate the Protection of Children Act of 1978, which makes it illegal to "make" or "show" in indecent image of a child. Two days later, the airport caved, announcing that children will be exempted from going through the RapiScan machine. Since UK law defines children as anyone under the age of 18, terrorist groups now have received what amounts to an invitation to use 16 and 17-year-olds as suicide bombers on planes departing Manchester.
"Each item confiscated [at passenger checkpoints] is presumed to have potential explosive, highly flammable, or other characteristics that could endanger life and property. As I understand it, federal health and public safety regulations require that any such suspicious item be isolated until it can be handled by experts, properly attired in protective gear, who would then transport it in an explosion-proof container to a protected laboratory test cell where it would be photographed, x-rayed, measured, weighed, and documented prior to being opened up and examined by remote means to avoid injury to personnel. . . . Why is it that the public safety bureaucrats haven't raised a single eyebrow at the way the TSA folks blithely toss these potential weapons of mass destruction around inside airport terminals full of people? Could it be that they, too, are in on the little secret that this is all just security theater?"
--Aviation reporter (name withheld by request), on an aviation list-serve, Oct. 16, 2009.