Improving Management of the Aviation Security Workforce

Testimony of
Robert W. Poole, Jr.

House Committee on Homeland Security

Subcommittee on
Economic Security, Infrastructure Protection, and Cybersecurity

July 28, 2005

Reason Foundation
3415 S. Sepulveda Blvd., Suite 400
Los Angeles, CA 90034
310-391-2245
www.reason.org

1
My name is Robert W. Poole, Jr. I am the Director of Transportation Studies at the Reason
Foundation, a public policy think tank based in Los Angeles. As a former aerospace engineer, I
have been studying transportation issues for more than 20 years. My career also includes public
safety and criminal justice research, and since Sept.11, 2001, I have focused considerable
attention on issues involving aviation security. I was in Washington advising Members during
the House debates on the Aviation & Transportation Security Act of 2001, which created the
Transportation Security Administration (TSA) and is perhaps best known for “federalizing”
airport security.

I have recently completed research on an alternative to the model of airport screening mandated
by ATSA and put into practice by TSA. My testimony today draws on portions of that
forthcoming Reason policy paper.

Two Basic Problems: Centralization and Conflict of Interest

The most fundamental problem is the highly centralized way in which TSA has interpreted its
charge under ATSA. This is at odds with the great variation in size, design, and function of
America’s more than 400 commercial-service airports. In addition, because of its legislated role
as the principal provider of airport screening services, TSA is in the conflicted position of being
both the airport security policymaker/regulator and the provider of some (but not all) airport
security services. My testimony will address both problems.

1. Overcentralization
From the outset, TSA has been plagued by the conflict between centralization and
decentralization. Part of the rationale for “federalizing” airport security was to provide a
consistently high level of security nationwide, regardless of the myriad differences among
airports (which range from huge to tiny, from primarily origin & destination [O&D] to primarily
transfer hubs, and from centralized terminals to multiple terminals). These differences crucially
affect numerous aspects of both passenger and baggage processing. Early on, TSA officials
verbally acknowledged this vast diversity by repeatedly saying, “If you’ve seen one airport,
you’ve seen one airport.” But their highly centralized approach has revealed that sentence to be
mostly lip service.

One example is how TSA allocates screeners among the 446 airports it is responsible for. Once a
year, it reallocates the screening workforce, to take into account changes in airline activity, using
a confidential algorithm. These allocations may be tweaked occasionally during the course of a
year, but airport directors have no idea how the algorithm works and little ability to influence the
allocations.

The problem is that commercial aviation is an inherently dynamic industry. As part of our
analysis, we looked at how much air service changes at individual airports. Tables 1 and 2 are
drawn from a database of monthly enplaned passengers at the top 100 airports. For the sample
year 2003, the tables illustrate the month-to-month volatility in passenger numbers at these
airports, which account for the lion’s share of passengers and screeners.
Table 1

2
 Monthly Changes in Enplaned Passengers, Top 100 U.S. Airports, 2003
Month No. airports No. airports Airport with Amount of
with +/-10% with +/-15% greatest change change
January 77 54 Pensacola -26%
February 7 1 San Juan -19%
March 95 81 Myrtle Beach 76%
April 24 6 Salt Lake City -18%
May 29 15 Palm Springs -37%
June 20 7 Anchorage 57%
July 19 10 Islip 26%
August 11 0 Wichita -15%
September 82 56 San Juan -38%
October 64 35 Palm Springs 39%
November 23 9 St. Louis -47%
December 14 3 Myrtle Beach -22%
Source: U.S. DOT T-100 carrier reports

Table 2
Examples of Monthly Airport Enplanement Volatility, 2003 (percent change)
Airport Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Ft. Myers 7 8 38 -11 -32 -20 3 -8 -23 38 28 6
Seattle -21 -4 17 1 9 16 11 2 -25 -4 -4 12
Source: U.S. DOT T-100 carrier reports

A screener staffing allocation decided a year in advance is simply not a good fit for this dynamic
airline environment. When a single airline begins serving, or withdraws from serving, such an
airport, the change can happen in a matter of a month or two, but it may take TSA six months or
more to catch up with it (if it is under sufficient pressure to make a change prior to the next
annual screener re-allocation). During those many months, the airport will operate with too few
or too many screeners.

A second example is the highly centralized way in which TSA has interpreted the provision in
ATSA that allowed five airports to opt out of TSA-provided screening as a pilot program. What
airports expected, and what most people would assume to be the way to implement such a
program, would be for TSA to define criteria for such firms, certify those that met the criteria,
define the rules for airports to implement outsourced screening, and then let those airports with
acceptable plans issue RFPs and select the firm (from those on TSA’s list) submitting the best
proposal. The airport would then contract with the firm, under the supervision of the TSA’s
Federal Security Director who oversees all other security operations at that airport.

That was not how TSA implemented the pilot program, however. While it did certify a handful
of firms, it did not allow airports to issue RFPs, select their preferred bidder, or enter into a
contract. Rather, after TSA selected the five airports that would participate as the pilot sites, it
assigned one of its certified firms to each airport. The TSA itself entered into a contract with
each firm and directly supervised its operation at each airport. Moreover, when the November
2004 date specified by ATSA approached, after which point all airports would be free to opt out

3
of TSA-provided screening in favor of contract operations, TSA defined its Screening
Partnership Program along the same highly centralized lines.1

And the centralization does not stop there. As the Government Accountability Office (GAO)
noted in an April 2004 assessment of the pilot program, because TSA runs the program in such a
centralized manner, “private screening contractors have had little opportunity to demonstrate and
achieve efficiencies.”2 Among other things, the GAO report notes that the contractors lack the
authority to determine staffing levels and conduct hiring. Their head-count is part of the TSA’s
overall 45,000, allocated as part of the overall process. And actual hiring by the contractors must
be coordinated through TSA headquarters. Before new staff can be hired by a contractor, TSA
must authorize this, and it must set up an assessment center in the area, using TSA’s national
assessment contractor. According to GAO, this process typically takes several months. Their
report notes a case at one of the pilot program airports where a staff shortage went on for
months, waiting for TSA’s process. The inability to hire screeners during this time “contributed
to screener performance issues, such as absenteeism or tardiness, and screener complacency,
because screeners were aware that they are unlikely to be terminated due to staffing shortages.”3

GAO also reported that Federal Security Directors (FSDs) at non-pilot program airports
expressed similar frustrations at TSA’s centralization of hiring and training. In a survey of all
155 FSDs, GAO found that “the overwhelming majority . . . reported that they needed additional
[local] authority to a great or very great extent.”4

2. Conflict of Interest
Congress decided to “federalize” airport screening after concluding that the prior institutional
arrangements included both regulatory failure and conflict of interest. Prior to 9/11, the Federal
Aviation Administration was in charge of airport security, and its rules required that access to
airport concourses be limited to those who cleared a basic screening process at checkpoints. The
FAA delegated this screening responsibility not to the airports (which own the premises) but
rather to the airline that had the largest presence on each concourse (generally a “signatory”
airline that had signed a long-term use and lease agreement with the airport). The structural
failure was that the airlines had no real incentive to make security a priority. Since operating this
function was a cost item for airlines, and airlines operate in a very competitive business, their
interest was to meet whatever requirements FAA laid down at minimal cost. Over time, that led
to the well-documented situation in which the screening companies paid not much more than
minimum wage, did only modest amounts of training, and suffered turnover rates sometimes in
excess of 100 percent per year.

The regulatory failure was that the FAA essentially set no standards for hiring and training of
screeners. Moreover, the FAA was de-facto satisfied with the relatively low level of performance
of those screeners, when challenged by “Red Teams” that attempted to get prohibited items past

1
Transportation Security Administration, “Guidance on Screening Partnership Program,” June 2004.
(www.tsa.gov/interweb/assetlibrary/SPP_OptOut_Guidance_6.21.04.pdf).
2
General Accounting Office, “Aviation Security: Private Screening Contractors Have Little Flexibility to
Implement Innovative Approaches,” GAO-04-505T, Washington, DC, April 22, 2004.
3
Ibid, p. 7.
4
Ibid, p. 10.

4
the screeners. GAO called for implementation of performance standards for screening in 1987,5
but the agency failed to act. In the 1996 FAA reauthorization act, Congress required FAA to
“certify companies providing security screening and to improve the training and testing of
security screeners through development of uniform performance standards.”6 Three years later,
in January 2000, FAA issued a proposed rule, Certification of Screening Companies, which
would have held companies to minimum performance standards. When the rule had not been
finalized by November 2000, Congress directed FAA to issue a final rule no later than May 31,
2001.7 The FAA failed to meet this deadline, so Congress then required it to report twice a year
on the status of each missed statutory deadline. That was the situation as of Sept. 11, 2001.

In response, Congress took responsibility for airport security away from FAA and gave it to the
newly created TSA, an appropriate response to regulatory failure. But in response to the
structural failure, instead of doing as nearly every other country in the world does—making each
airport responsible for securing its operations under national regulatory supervision—Congress
instead vested in TSA not only the regulatory responsibility but also the service provision duties
of airport screening. Note that TSA was not given responsibility for carrying out access control
or perimeter patrols or law enforcement functions at the airports. Those security functions were
still the airport’s responsibility, under the watchful eye of the TSA’s Federal Security Director
(FSD) assigned to that airport. But for baggage and passenger screening, TSA was to be both the
regulator and the operator.

This dual role is a potentially serious conflict of interest. As one airport director said to a
Chicago Tribune reporter in the early days of TSA, “The problem inherent in the federally
controlled screening process is that you end up having a federal agency sitting in the middle of
your terminal, essentially answerable to nobody.”8 This point was underscored in BearingPoint’s
report on the five pilot-program airports. “Because the screeners at a private contractor [pilot
program] airport are not government employees, the FSD is able to take a more objective
approach when dealing with screener-related issues raised by stakeholders such as airport
management or air carriers.”9

The classic example of a federal agency with this kind of dual-role conflict was the Atomic
Energy Commission, created after World War II to encourage peaceful uses of nuclear power. In
carrying out this mission, the AEC became both a promoter of nuclear energy (funding research
& development, doing educational/marketing work, etc.) and the regulator of all civilian nuclear
reactor operations. Eventually, public criticism of the conflict of interest—that the AEC could
not serve as an objective regulator if it was also the chief promoter of nuclear power—led
Congress to split those functions. It created a purely regulatory body, the Nuclear Regulatory
Commission, for that role. And it shifted the R&D functions into the newly created Department
of Energy.

5
General Accounting Office, “Aviation Security: FAA Needs Preboard Passenger Screening Performance
Standards, GAO-RCED-87-182, Washington, DC July 24, 1987.
6
Sec. 302, P.L. 104-264.
7
Sec. 3, P.L. 106-528.
8
Jon Hilkevitch, “Airports Not Sold on Federal Screeners,” Chicago Tribune, April 6, 2002.
9
BearingPoint, “Private Screening Operations Performance Evaluation Report, Summary Report,” Washington, DC:
Transportation Security Administration, April 16, 2004.

5
Rethinking TSA’s Role in Screening

Early in 2005 separate reports were made to Congress, one by the DHS Inspector General’s
Office and the other by the GAO. Based on testing of airport screening operations, both
concluded that screening performance today, several years after TSA took over, is no better than
it was shortly before or shortly after 9/11. In other words, the new agency with a budget of $5.5
billion per year, nearly half of which is devoted to baggage and passenger screening, has not led
to improved protection of planes from dangerous objects.

The GAO report also concluded that the performance of screeners at the five pilot-program
experts was modestly better, on average, than that of TSA screeners—enough of a difference to
be statistically significant. Given that TSA provision of screening services entails a conflict of
interest, those findings serve to strengthen the case for separating such service provision from
TSA’s inherently governmental role as security policy-maker and regulator in aviation. That
would permit the actual provision of airport security to be devolved to each airport, as it is in
Europe and most of the rest of the world, under TSA oversight via the FSDs. Airports would be
free to provide those services either in-house, with their own workforces, or by contracting with
a TSA-certified security company.

A. Separating Policy-Making and Regulation from Operations

The dual-role nature of TSA stems directly from the ATSA legislation. Thus, this problem can
only be corrected by new legislation to overhaul TSA in the interest of improving its
performance, thereby increasing aviation security.

We can gain useful perspective on this issue by looking at how European governments have
addressed this issue. Europe began confronting hijackings and terrorist attacks on airports in the
late 1960s. Risk analysis identified the need for a comprehensive approach that included
background checks of airport personnel, passenger and baggage screening, and airport access
control. The initial approach in most nations was to use national government employees to beef
up airport security, either from the transport agency or the justice agency. But beginning in the
1980s, European airports began developing a performance contracting model, in which
government set and enforced high performance standards and airports carried them out—usually
by hiring security companies, but occasionally with their own staff. Belgium was the first to
adopt this model, in 1982, followed by The Netherlands in 1983 and the United Kingdom in
1987, when BAA was privatized. The 1990s saw a new wave of conversions to the public-
private partnership model, with Germany switching in 1992, France in 1993, Austria and
Denmark in 1994, Ireland and Poland in 1998, and Italy, Portugal, Spain, and Switzerland in
1999.

Table 3 provides a breakdown of outsourced passenger and baggage screening at 33 large

European airports as of late 2001. Of these, only Zurich and Lisbon airports were not using the
performance contracting model, and in both nations, efforts to shift to this model were under
way.

Table 3

6
 Outsourced Passenger and Baggage Screening in Europe

RANK BY
PASSENGER & HAND Private HOLD BAGGAGE Private
TOTAL INT'L CITY (AIRPORT CODE)
BAGGAGE SCREENING /2 Screeners? SCREENING Y Screeners?
PAX /1

1 LONDON (LHR) BAA Y ADI Initial, SIS (CIVAS) Y

2 PARIS (CDG) SIFA/Brinks/ICTS Y ICTS/ASA/SIFA Y

3 FRANKFURT/MAIN (FRA) FRAPORT Y FRAPORT and others /3 Y

Randon Securicor-ADI &
4 AMSTERDAM (AMS) Group 4 Falk Y Y
Group 4 Falk
5 LONDON (LGW) BAA Y ICTS; Initial Y

6 BRUSSELS (BRU) Securair Y Securair Y

7 ZURICH (ZRH) State Police State Police /4 See note 4 below

8 COPENHAGEN (CPH) Copenhagen Airport Security Y Copenhagen Airport Security Y

9 MANCHESTER (MAN) Manchester Airport plc Y Securicor/ADI Y

10 MADRID (MAD) Vinsa, State Police Y State Police

11 MUNICH (MUC) SGM (Airport Company) Y various private companies /3 Y

Aeroporto di Roma; physical

12 ROME (FCO) Y Aeroporto di Roma Y
searches handled by police

13 DUSSELDORF (DUS) ADI Y ADI Y

SEA; physical searches

14 MILAN (MXP) Y SEA Y
handled by police

15 DUBLIN (DUB) Aer Rianta (Airport Authority) Y Aer Rianta (Airport Authority) Y

16 STOCKHOLM (ARN) Group 4 Falk Y Group 4 Falk Y

/3
17 VIENNA (VIE) VIASS Y VIASS and others Y

18 PARIS (ORY) ASA, SIFA Y ICTS, Brinks Y

19 BARCELONA (BCN) Prosegur, State Police Y Prosegur,State Police Y

20 LONDON (STN) BAA Y ADI (Securicor) Y

/5 /5
21 LISBON (LIS) State Police See note 5 below State Police See note 5 below

22 OSLO (OSL) ADECCO, Olsten Y ADECCO, Olsten Y

80% Securitas/20% State 80% Securitas/20% State
23 MALAGA (AGP) Y Y
Police Police
n/av GENEVA (GVA) Airport Authority Y ICTS Y

n/av ATHENS (ATH) ICTS/Wackenhut/3D Y Hermis/Civas Y

n/av NICE (NCE) ICTS, SGA Y ICTS, SGA Y

n/av HELSINKI (HEL) Securitas Y Securitas Y

n/av BIRMINGHAM (BHX) ICTS & AAS Y ICTS & AAS Y

n/av BERLIN (BER) Securitas Y Securitas Y

n/av STUTTGART (STR) FIS Y FIS Y

n/av COLOGNE (CGN) ADI Y ADI Y

n/av HAMBURG (HAM) FIS Y FIS Y

n/av HANNOVER (HAJ) FIS Y FIS Y

/1
Based on 1999 Int'l Airport Traffic Statistics from ACI.
/2
As of October 2001.

7
/3
These airports do not have centralized baggage screening, but airlines hire private companies to x-ray bags.
/4
Public/private partnership underway.
/5
Legislation proposed to permit public/private sector partnership.
Source: Aviation Security Association

The GAO visited five nations in 2001 to examine their security screening practices—Canada and
four European nations (Belgium, France, The Netherlands, and the United Kingdom).10 Its report
focused on the superior performance of the European airports, all of which use the performance
contracting model. GAO reported significant differences between their screening practices and
that of then-current U.S. airports in four areas:

• Better overall security system design (allowing only ticketed passengers past screening,
stationing law enforcement personnel at or near checkpoints, etc.);

• Higher qualifications and training requirements for screeners (e.g. 60 hours in France vs. 12
hours as then required by FAA);

• Better pay and benefits, resulting in much lower turnover rates; and,

• Screening responsibility lodged with the airport or national government, not with airlines.

Most of these lessons were incorporated by Congress into the ATSA. What was largely ignored,
however, was the fact that under the European conditions of high standards and oversight,
performance contracting (hiring private security firms, paying them adequately, and holding
them accountable for results) is the model adopted by nearly all European airports over the past
two decades. Israel and a number of other nations in the Caribbean and the Far East also use this
model.

Companies that do not meet the standards and perform effectively are not simply fined but
actually have their contracts cancelled. Since these are typically long-term (e.g., up to six-year)
contracts, losing such a contract is a serious loss of business, creating a strong incentive for high
performance. Companies often bid on a whole package of security services, not just passenger
screening, paid for via a single monthly charge. This avoids undue cost pressures being put on
any one element.

The alternative recommended here is not “privatization”—which would be the case if all airports
were required to use private contractors. Rather, it is devolution. The idea would be to remove
TSA’s conflict of interest by devolving the actual provision of screening to the airport level,
which is where all other aspects of airport security (such as access control and perimeter
protection) already reside. Airports would then have the option of complying with federal
screening requirements either with their own TSA-approved screening workforce or by hiring a
TSA-certified screening contractor. This approach has strong support among airport directors,
and is also embraced by the leading congressional champion of TSA reform, Rep. John Mica (R,
FL), chairman of the House Aviation Subcommittee. Mica has called for “a decentralized

10
General Accounting Office, “Aviation Security: Terrorist Acts Demonstrate Urgent Need to Improve Security at
the Nation’s Airports,” Testimony of Gerald L. Dillingham before the Senate Commerce, Science, and
Transportation Committee, Sept. 20, 2001.

8
screening program with federal oversight,” citing the TSA’s conflict of interest a case of “the
regulator regulating itself.”11

B. Airport-Centered Security

How would devolution work? As in Europe, the airport director would be in charge of securing
the airport premises, under the supervision of the TSA Federal Security Director (FSD) assigned
to that airport. I will discuss four key aspects a devolved model of airport screening.

1. Make-or-Buy Authority
The most fundamental aspect of devolution is that the responsibility for carrying out the
screening of baggage and passengers would be shifted from TSA to each individual airport. And
as with all other airport services, it would be up to the airport to decide how to carry out the
screening functions. Like most businesses, airports outsource some services and perform others
using their own paid staff. In the case of screening, as with other security functions, the
operations would have to comply with all TSA requirements.
But with TSA no longer being in the business of screening, its requirements would have to be
reconfigured for the new circumstances. To gain the flexibility advantages that go along with
devolution, the hiring and training of screeners should be devolved rather than being centralized
in Washington and carried out by a national TSA contractor. Rather, TSA would provide training
requirements and a core curriculum which could be used by airports, TSA-certified screening
contractors, and TSA-certified screener training firms operating on a decentralized basis in
various parts of the country.

2. Funding Allocations
Under current law, passenger and baggage screening are paid for by TSA, whether provided by
its own workforce or by TSA-certified contractors. This funding would presumably continue
under devolution, but in order to take advantage of the flexibilities provided by devolution, two
key changes should be made in how the funding is done. First, the allocations should be made far
more frequently than once a year; ideally every month but at least quarterly. This should be done
in accordance with a transparent workload formula arrived at with significant input from the
airport organizations, AAAE and ACI-NA. Second, each airport should receive a lump sum
amount which it can use as it sees fit for TSA-approved screening operations.

Why monthly allocations rather than the current more-or-less annual allocation? The idea is to
better match resources with workload. As Tables 1 and 2 illustrated, today’s dynamic, highly
competitive airline industry is characterized by rapid change. USAirways downsizes its hub at
Pittsburgh; JetBlue orders 100 new larger-size regional jets to add service to many smaller
airports; America West and USAirways merge, very likely leading to further cutbacks at some
airports; and one of more legacy carriers may well liquidate (Chapter 7 bankruptcy), leading to
significant changes in service. With funding allocations adjusted every month among the 446
airports with screeners, and the local flexibility to increase and decrease staffing as needed, there
will be a much better match of screening workforce to actual workloads.

11
U.S. Rep.John L. Mica, “Screening Reform,” Aviation Week & Space Technology, June 6, 2005.

9
In addition to keeping funding in pace with passenger flow, the devolved system should leave the
funds unencumbered by many of the current requirements. Currently, TSA screeners are paid on
a national wage scale, regardless of local living costs. And TSA-certified screening contractors
must, per ATSA, pay the identical wages and benefits to their screeners. While the intent of these
provisions in ATSA was to prevent a return to minimum-wage screeners with high turnover, that
was a brute-force solution to a problem caused by the lack of FAA standards for screener
selection, training, and performance. With hiring and operations under the control of each
airport, the airport or its contractor should be free to innovate, using whatever mix of job
functions and compensation approaches will best get the job done, while meeting all TSA
training and performance standards. Thus, especially at smaller airports, the same employee
might do passenger screening during peak morning hours and do access-control or perimeter
patrol during the remainder of her shift. Some airports (or their contractors) might develop
workable split-shift approaches to cover morning and afternoon peaks without paying for a lot of
unproductive time in between. The point is to let airports and their contractors decide on the best
use of the screening money, to get the most bang for the buck.

3. Incentives for In-line Baggage Systems

The imposition, in ATSA, of extremely tight deadlines for implementing 100 percent explosive-
detection inspection of all checked baggage also led to brute-force solutions. Large and medium
airports mostly installed huge EDS machines in their ticket lobbies or in available spaces in their
baggage areas; in either case, they had to be loaded by hand, one bag at a time. Between the
inherently slow processing time and this hand-feeding, processing rates were often as low as 100
bags/hour. Hence, in order to prevent massive delays, large numbers of $1 million apiece EDS
machines were required. Smaller airports were equipped mostly with explosive trace detection
(ETD) machines as their primary means of compliance with the inspection mandate; in addition,
thousands of ETDs were installed at large and medium airports for secondary screening of bags
identified as suspicious by EDS. As of June 2004, some 1,228 EDS and 7,146 ETD machines
had been installed at U.S. airports.12

These brute-force approaches are very labor-intensive. If EDS machines are integrated into a
conveyor-fed baggage processing system (in-line system), and especially if go/no-go
assessments are made at a remote display terminal (on-screen resolution), the bag processing
rates go way up and the labor involved goes way down. The latest GAO report on the subject
finds that under ideal conditions, an in-line EDS system can process 425 bags/hour compared
with 180 bags/hour under ideal conditions for stand-alone EDS. And replacing an ETD operation
with stand-alone EDS changes throughput from 36 bags/hour to 180 bags/hour.13

The savings in labor are equally impressive. According to the GAO report, a typical lobby-based
EDS installation has one EDS plus three ETDs, requiring a workforce of 19 screeners. This can
be replaced by an in-line EDS requiring just 4.25 screeners—a 78 percent reduction. For the nine
large airports that have implemented in-line systems, TSA’s retrospective analysis found a
reduction in bag screeners and supervisors of 78 percent. Similar GAO calculations analyzed
replacing a 3 to 5-unit ETD installation with one stand-alone EDS plus one ETD for alarm

12
Government Accountability Office, “Airport Security: Systematic Planning Needed to Optimize the Deployment
of Checked Baggage Screening Systems,” GAO-05-365, March 2005.
13
Ibid.

10
resolution. The former would require 12.3 to 20.5 screeners, while the latter needs only 6.75. If
we take the intermediate case of a 4-unit ETD installation, the reduction in staff from 16.4 to
6.75 is 59 percent.

Because of numbers like these, several airports that have switched from stand-alone, lobby-based
EDS to in-line systems with on-screen resolution have reported a payback period of little more
than one year. TSA’s analysis of nine airports shifting to in-line system reached a similar
conclusion, the GAO reported. In other words, the one-time investment in in-line EDS quickly
pays for itself in reduced payroll costs. (It should be noted that GAO’s review of TSA’s
aggregated analysis found that the results held true for eight of the nine airports; modification
costs were so high at Seattle’s SEA-TAC that there were no net cost savings from the
conversion.14)

If the screening funds were devolved to airports as proposed above, it would clearly be in an
airport’s interest to finance the investment in new screening systems so as to achieve these
ongoing savings. And once the costs of the equipment and facility modernization were paid off,
the savings could be used for other security improvements, such as more passenger screening
lanes and screeners, if needed. Over time, as overall screening costs came down, smaller annual
allocations from TSA would be needed, thereby producing federal budget savings.

4. Liability
One of the issues that have held back many airports from participating in the post-November
2004 Screening Partnership Program is liability. With TSA as their provider of screening
services, if a terrorist incident having any connection with passenger or baggage screening
occurs at the airport, then TSA is the party most likely to be at risk for lawsuits. But if the airport
opts for a TSA-certified contractor, and such an incident occurs, there has been concern that the
airport might be at greater risk for not having gone with the standard approach.

This liability issue arose first in connection with EDS machines and other technologies needed in
security protection. In response, Congress passed the Support Anti-terrorism by Fostering
Effective Technologies Act, better known as the SAFETY Act. It provides a process by which
companies providing homeland security technologies or services can become certified by DHS
and win a limit on their liability. FirstLine and Covenant, two of the leading private screening
companies, have recently received this designation.

However, if TSA withdraws from the provision of screening services and this function is
devolved to airports, the same liability concern may arise. Under that new set of alternatives, it
would maintain more of a level playing field between in-house and contracted screening services
if airports were made eligible to receive the same extent of SAFETY Act protection as
designated screening companies.

Summing Up: Benefits of Reform

I have argued for two basic changes in the model of airport security that has been employed in
the United States since the passage of the ATSA legislation in 2001. Those changes are (1) to
14
Ibid, Appendix IV.

11
remove the TSA’s conflict of interest by making it the policy-maker and regulator, but not the
provider, of airport screening; and (2) to devolve screening responsibility to the airport level,
under the supervision of TSA’s Federal Security Director in each case.

Those changes would improve airport security in several ways. By making all on-airport security
functions the responsibility of the airport, this approach would lead to a more integrated
approach, with the FSD supervising everything. Removing TSA’s conflict of interest, and
making the airport responsible for all aspects of security (as in Europe) should also increase
accountability for results.

This approach should also produce meaningful savings in annual payroll costs for screening
functions, as well as permitting a shift of screeners from baggage to passenger screening. The net
savings will free up scarce airport security resources for other needs such as lobby security,
access control, and perimeter control. Over time, those savings at airports may permit TSA and
DHS to spend relatively more on protecting vital non-aviation infrastructure.

This concludes my presentation. More details on this subject will be available this fall, when
Reason publishes the policy study on which these comments are based.

12