National Airspace System: Status of Wide Area Augmentation System Project
(Letter Report, 04/30/98, GAO/RCED-98-79).

Pursuant to a legislative requirement, GAO reviewed the status of the
Federal Aviation Administration's (FAA) Wide Area Augmentation System
(WAAS) project, focusing on: (1) whether the Secretary of
Transportation's report provides a complete assessment of FAA's risks in
developing the WAAS project; and (2) how alternative assumptions would
affect WAAS' benefit-cost analysis of January 1998.

GAO noted that: (1) the Secretary's report provided a complete
assessment of FAA's risks in achieving the WAAS project's performance
and cost goals but not its scheduled goals; (2) in terms of system
performance, the Secretary's report recognized that WAAS' vulnerability
to intentional or unintentional interference from electronic equipment
must be addressed; (3) in January 1998, FAA estimated that it would save
about +$500 million (in net present value) over the WAAS project's life
cycle by fully phasing out its network of ground-based navigation aids;
(4) if FAA retains some portion of this network, these benefits would
decrease; (5) FAA also estimated that aircraft operators could save $350
million by removing ground-based navigation equipment from their
aircraft; (6) these benefits would be reduced to the extent that
operators must continue to keep such equipment on board; (7) by
identifying a range of cost estimates and associated probabilities, the
Secretary's report addressed GAO's past concern that FAA's firm,
discrete-point cost estimates implied a level of precision that could
not be supported; (8) GAO agreed with the Secretary's report that the
greatest degree of uncertainty about the WAAS cost estimates relates to
the costs of the geostationary communications satellites; (9) the
uncertainty exists because FAA does not know exactly how many satellites
will be needed and how much the per-unit costs will be; (10) the
Secretary's report fell short of providing a complete assessment of the
uncertainties FAA faces in achieving WAAS' schedule goals; (11) the
report also did not discuss the risks to the overall schedule if FAA
does not award the contract to lease the satellites by July 1998 as
planned; (12) in January 1998, FAA's analysis found that the benefits to
aviation from WAAS would be three times as great as its costs; (13) GAO
requested that FAA recalculate its benefit-cost analysis to determine
the impact of three alternative assumptions; (14) using these cost and
decommissioning assumptions did not cause much of a decrease in the
benefit-cost ratio or the net benefits; (15) however, the exclusion of
small increments of passenger time savings had a much more significant
impact; and (16) when these alternative assumptions were taken together,
GAO found that the net present value of the project's net benefits
decreased by more than $1 billion but were still about twice as great as
the costs.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-98-79
     TITLE:  National Airspace System: Status of Wide Area Augmentation 
             System Project
      DATE:  04/30/98
   SUBJECT:  Life cycle costs
             Navigation aids
             Cost effectiveness analysis
             Military satellites
             Air transportation operations
             Geosynchronous satellites
             Federal procurement
             Reporting requirements
IDENTIFIER:  FAA Wide Area Augmentation System
             NAVSTAR Global Positioning System
             Very High Frequency Omnidirectional Range/Distance 
             Measuring Equipment System
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Requesters

April 1998

NATIONAL AIRSPACE SYSTEM - STATUS
OF WIDE AREA AUGMENTATION SYSTEM
PROJECT

GAO/RCED-98-79

National Airspace System

(348055)


Abbreviations
=============================================================== ABBREV

  CNS - communications, navigation, and surveillance
  DOT - Department of Transportation
  DOD - Department of Defense
  FAA - Federal Aviation Administration
  GEO - geostationary
  GPS - Global Positioning System
  OMB - Office of Management and Budget
  VOR - very high frequency omnidirectional ranging
  WAAS - Wide Area Augmentation System

Letter
=============================================================== LETTER


B-279202

April 30, 1998

The Honorable Richard C.  Shelby
Chairman
The Honorable Frank Lautenberg
Ranking Minority Member
Subcommittee on Transportation
Committee on Appropriations
United States Senate

The Honorable Frank R.  Wolf
Chairman
The Honorable Martin O.  Sabo
Ranking Minority Member
Subcommittee on Transportation
 and Related Agencies
Committee on Appropriations
House of Representatives

As a key element of its overall program for modernizing the National
Airspace System, the Federal Aviation Administration (FAA) is
planning a transition from ground- to satellite-based navigation by
using satellite signals generated by the Department of Defense's
(DOD) Global Positioning System (GPS).  However, GPS by itself does
not satisfy all civil air navigation requirements, such as the one
requiring that aircraft operators be provided timely warnings of
system malfunctions and another requiring that the system be
available virtually all of the time.  FAA is developing a network of
ground stations and geostationary communications satellites\1 to
enhance GPS so that satellite-based navigation can meet those
requirements.  This network is known as the Wide Area Augmentation
System (WAAS).  The system is expected to improve the safety of
flight operations, allow the fuel-efficient routing of aircraft, and
enable FAA to eventually phase out its network of ground-based
navigation aids. 

In light of concerns about the WAAS project, the conferees for the
Department of Transportation Appropriations Act for fiscal year 1998
and the Senate Appropriations Committee placed reporting requirements
on the Secretary of Transportation and on us.\2 The Secretary was
directed to report on the status and management of the project,
including an identification of baseline performance, cost, and
schedule goals and to provide a risk assessment.  The Secretary's
report was issued on February 11, 1998.\3 We were directed to review
the status of the WAAS project and report by March 1, 1998.  As
agreed with your offices, we delayed the issuance of our report until
we had an opportunity to review the Secretary's report.  We examined
whether the Secretary's report provides a complete assessment of
FAA's risks in developing the WAAS project.  We also examined how
alternative assumptions would affect WAAS' benefit-cost analysis of
January 1998. 


--------------------
\1 Geostationary satellites are located at fixed positions in orbit
22,000 miles above the earth. 

\2 Senate Report 105-55, Department of Transportation and Related
Agencies Appropriations, Fiscal Year 1998 (July 22, 1997), and House
of Representatives Conference Report 105-313, Department of
Transportation and Related Agencies Appropriations, Fiscal Year 1998
(Oct.  7, 1997).  The FAA Administrator is also required to report
quarterly on actual progress made toward baseline performance, cost,
and schedule goals. 

\3 Wide Area Augmentation System (WAAS) Report on Program Status and
Management, Federal Aviation Administration (Feb.  11, 1998). 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

The Secretary's report provided a complete assessment of FAA's risks
in achieving the WAAS project's performance and cost goals but not
its schedule goals. 

  -- In terms of system performance, the Secretary's report
     recognized that WAAS' vulnerability to intentional or
     unintentional interference from electronic equipment must be
     addressed.  This vulnerability may lead FAA to retain an
     independent backup system and to revise its transition plan that
     calls for phasing out all of the agency's ground-based systems
     by 2010.  In January 1998, FAA estimated that it would save
     about $500 million (in net present value) over the WAAS
     project's life cycle by fully phasing out its network of
     ground-based navigation aids.  If FAA retains some portion of
     this network, these benefits would decrease.  FAA also estimated
     that aircraft operators could save $350 million by removing
     ground-based navigation equipment from their aircraft.  These
     benefits would be reduced to the extent that operators must
     continue to keep such equipment on board. 

  -- By identifying a range of cost estimates and associated
     probabilities, the Secretary's report addressed our past concern
     that FAA's firm, discrete-point cost estimates implied a level
     of precision that could not be supported, particularly early in
     the project's development.  We agree with the Secretary's report
     that the greatest degree of uncertainty about the WAAS cost
     estimates relates to the costs of the geostationary
     communications satellites.  FAA estimates these costs at about
     $1.2 billion, or 40 percent, of the $3.0 billion total cost of
     the project.  The uncertainty exists because FAA does not yet
     know exactly how many satellites will be needed and how much the
     per unit costs will be. 

  -- The Secretary's report fell short of providing a complete
     assessment of the uncertainties FAA faces in achieving WAAS'
     schedule goals.  It did not, for example, point out that FAA
     might not find a vendor that is able and willing to complete the
     launching and testing of the satellites by October 2001, as
     called for in the report.  A number of potential vendors have
     cited 2002 or 2003 as a more realistic schedule for putting the
     satellites in orbit.  The report also did not discuss the risks
     to the overall schedule if FAA does not award the contract to
     lease the satellites by July 1998 as planned.  Meeting this date
     is doubtful, however, because (1) negotiations over the terms of
     the contract may become protracted as FAA and the vendor seek to
     minimize their financial risks and (2) FAA may defer awarding
     the contract until it receives congressional approval to extend
     the leasing period from 5 years to 10 years. 

In January 1998, FAA's analysis found that the benefits to aviation
from WAAS would be three times as great as its costs.  We requested
that FAA recalculate its benefit-cost analysis to determine the
potential impact of three alternative assumptions:  (1) accounting
for higher-than-expected satellite costs, (2) reducing the benefits
from phasing out, or "decommissioning," the full network of
ground-based navigation aids in case an independent backup network is
retained, and (3) excluding the benefits derived from small
increments of passenger time savings.  Using these cost and
decommissioning assumptions did not cause much of a decrease in the
benefit-cost ratio or the net benefits.  However, the exclusion of
small increments of passenger time savings had a much more
significant impact.  When these alternative assumptions were taken
together, we found that the net present value of the project's net
benefits decreased by more than $1 billion but were still about twice
as great as the costs. 


   BACKGROUND
------------------------------------------------------------ Letter :2

In the 1980s, FAA began considering how a satellite-based navigation
system might eventually replace the ground-based system that has long
provided navigation guidance to aircraft.  In August 1995, after
several years of research, FAA contracted with Wilcox Electric to
develop WAAS to enhance GPS.  However, because of concerns about the
contractor's work, FAA terminated the contract in April 1996.  In May
1996, the agency entered into an interim contract with Hughes
Aircraft Company (now Raytheon Systems), with the contract becoming
final in October 1996. 

Accuracy, integrity, availability, continuity, and service volume are
the major performance goals for the system to meet.  Accuracy is
defined as the degree to which an aircraft's position as calculated
using the system conforms to its true position.  For precision
approaches to runways, WAAS is expected to provide aircraft operators
with position accuracy within 7.6 meters 95 percent of the time.\4
Integrity is the system's ability to provide timely warnings when its
signals are providing erroneous information and, thus, should not be
used for navigation.  WAAS is expected to provide a warning to
aircraft operators within 5.2 seconds.  Availability is the
probability that, at any given time, the system will meet FAA's
accuracy and integrity requirements for a specific phase of flight. 
For precision approaches, WAAS is expected to be available all but 9
hours per year.  Continuity is the probability that the system's
signal will meet accuracy and integrity requirements continuously for
a specified period.  Service volume is the area of coverage for which
the system's signal will meet availability requirements. 

As shown in figure 1, WAAS is a network of ground stations and
geostationary (GEO) communications satellites: 

  -- Reference stations on the ground (up to 53 units) will serve as
     the primary data collection sites for WAAS.  These stations will
     receive data from the GPS and GEO satellites. 

  -- Master stations on the ground (up to seven units) will process
     the data collected by the reference stations and generate
     accuracy corrections and integrity messages for each of the GPS
     and GEO satellites.  These stations will also validate the
     transmitted corrections. 

  -- Ground earth stations (up to 14 units) will, among other things,
     transmit accuracy corrections and integrity messages generated
     by the master stations to FAA's GEO satellites. 

  -- GEO satellites (up to six satellites) will transmit wide-area
     accuracy corrections and integrity messages to aircraft and also
     broadcast signals that will be similar to the signals broadcast
     by the GPS satellites. 

  -- A ground communications system will transmit information among
     the reference stations, master stations, and ground earth
     stations. 

For pilots to use WAAS for navigation, their aircraft will have to be
equipped with receivers that process the information carried by the
GPS and GEO signals.  The receivers will enable the pilots to
determine the precise time and the speed and three-dimensional
position (latitude, longitude, and altitude) of their aircraft. 

   Figure 1:  WAAS' Architecture

   (See figure in printed
   edition.)

   Source:  Based on an
   illustration from FAA.

   (See figure in printed
   edition.)

By July 30, 1999, FAA expects that WAAS' initial operational
capability will be available for pilots' use.  At that time, WAAS is
expected to support aircraft navigation for all phases of flight. 
However, the initial system will not contain all the required
hardware and software components needed for (1) redundancy in the
event of equipment failures and (2) availability for the nation's
entire airspace.  By December 2001, FAA plans to develop a fully
operational WAAS by adding reference stations and upgrading software
under the Raytheon contract and adding GEO satellites under a
separate contract.  The full system is expected to be capable of
eventually serving as a "sole means" navigation system.  That is, the
system must, for a given operation or phase of flight, allow the
aircraft to meet all navigation system performance requirements. 


--------------------
\4 FAA currently categorizes landing systems according to their
ability to safely guide an aircraft to a runway.  WAAS is expected to
support Category I precision approaches by providing safe vertical
guidance to an aircraft as it descends to a height of not less than
200 feet with runway visibility of at least 1,800 feet. 


   PERFORMANCE AND COST RISKS
   DISCLOSED, BUT SCHEDULE
   UNCERTAINTIES NOT FULLY
   RECOGNIZED
------------------------------------------------------------ Letter :3

The Secretary's report provided a complete assessment of the major
risks FAA faces in achieving the technical performance goals of the
WAAS project.  It also disclosed the cost uncertainties and the range
and probabilities of potential costs.  However, the report could have
done more to disclose the uncertainties associated with FAA's
schedule for making WAAS fully operational. 


      PERFORMANCE RISKS DISCLOSED
---------------------------------------------------------- Letter :3.1

In discussing the risks FAA faces in developing WAAS, the Secretary's
report highlighted the vulnerability of the system's signals to
intentional or unintentional interference from electronic equipment. 
It also discussed mitigation strategies, including the possibilities
of an independent backup system and full access to a second GPS
frequency. 

Concerns about the system's vulnerability to electronic interference
have been highlighted in recent months.\5 In an October 1997 report,
the President's Commission on Critical Infrastructure Protection
warned against relying on satellite navigation as the sole source of
aircraft landing guidance in light of potential interference.\6 That
same month, a group of independent experts from outside FAA, called
together by the agency's management to study the technical issues
facing WAAS, raised concern about the possible intentional jamming of
the signals.\7 In February 1998, the FAA Administrator's task force
on the National Airspace System's modernization recommended that FAA
address the risks posed by electronic interference and gain consensus
among users of the system about the agency's plan to switch from
ground- to satellite-based navigation. 

The Secretary's report recognized that WAAS' vulnerability to
interference must be assessed and appropriate countermeasures must be
in place before FAA can complete the transition to a satellite-based
navigation system.  The report cited several elements of a risk
mitigation plan.  For example, FAA has developed procedures for
reporting and responding to interference that include outfitting
flight inspection aircraft with the capability to locate sources of
interference. 

FAA may employ other risk mitigation strategies as well.  One is the
retention of an independent backup system.  The Secretary's report
noted that FAA is studying the need for such a system, and if the
need for a backup is established, the agency would evaluate various
alternatives.  While the backup system would not have to provide
aircraft operators with the same operational capability as WAAS or
the current ground-based system, it would have to provide, at a
minimum, safe navigation in the event of a loss of service from WAAS. 
Rather than designating WAAS as a sole means navigation system, FAA
may designate it initially as a "primary means" system until concerns
about electronic interference are resolved.  This means that WAAS
would not be expected to fully meet all availability and continuity
requirements for navigation.  As a result, aircraft operators would
either have to be equipped with a backup navigation system or have
restrictions on when and where they could fly. 

FAA and Mitre Corporation\8 officials told us that if an independent
backup system is retained, FAA may decide to deploy fewer WAAS
reference stations and satellites.  In making this decision, the
agency would consider whether civil air navigation requirements could
be met more cost-effectively with a combination of an independent
backup system and WAAS with fewer reference stations and satellites. 

Even if WAAS remains unmodified, the system's benefits to FAA and
aircraft operators could be expected to decrease if some portion of
the current ground-based network is retained as an independent backup
system.  FAA has intended to decommission its entire network of
ground-based navigation aids between 2005 and 2010--with the phaseout
concentrated toward the end of that period.  In January 1998, FAA
found that full decommissioning would result in the agency's saving
about $500 million (in net present value) over WAAS' life cycle.  The
agency also expected aircraft operators to be able to reduce the
proliferation of on-board navigation equipment.  The benefit-cost
analysis estimated that the operators would save about $350 million
by removing such equipment. 

Another risk mitigation strategy to counteract WAAS' vulnerability to
electronic interference (particularly unintentional interference) is
the use of a second frequency.\9 If one GPS frequency was lost
because of interference, a second frequency could be used to provide
service.\10 However, the current WAAS design assumes the use of
single-frequency receivers on board aircraft.  The Department of
Transportation (DOT) and DOD, as joint chairs of the Interagency GPS
Executive Board,\11 are working toward providing aviation and other
civil users with full access to a second GPS frequency on the next
generation of GPS satellites. 

Although the second civil frequency would not be fully operational on
GPS satellites until about 2010, FAA would prefer to build WAAS
ground- and space-based equipment so that users could operate with
"forward compatible" receivers--that is, receivers that can be built
to operate with a single frequency now and also operate with dual
frequencies in the future.  Once a final decision on the second
frequency is made, FAA and industry will need up to 2 years to
develop the minimum operational performance standards so that
manufacturers can begin producing receivers capable of single- and
dual-frequency operations. 


--------------------
\5 While ground-based navigation aids are vulnerable to interference,
satellite-based navigation is especially vulnerable because GPS
signals, which are broadcasted at low power levels, could be easily
jammed. 

\6 The purpose of this commission was to study the nation's
infrastructure, which constitutes the life support system of the
United States, to determine the vulnerability of that support system
and propose a strategy for protecting it in the future. 

\7 Final Report of the WAAS Study Group (Oct.  16, 1997). 

\8 Mitre Corporation operates, under a memorandum of agreement with
funding from FAA, the Center for Advanced Aviation System
Development.  The Center carries out a continuing program of
research, development, system architecture, and high-level system
engineering to support FAA's National Airspace System needs. 

\9 The present GPS satellites broadcast position data for DOD's use
on two frequencies referred to as L1 and L2.  WAAS receivers on
aircraft are currently designed to use position data on the L1
frequency.  WAAS reference stations will receive data on the L2
frequency and use it in combination with data on the L1 frequency to
make corrections. 

\10 Another potential advantage is that FAA could employ fewer WAAS
ground stations in the future if dual-frequency receivers are used to
correct position data that may be distorted as the GPS signal passes
through the ionosphere--that part of the earth's atmosphere beginning
at an altitude of 30 miles and extending outward 300 miles or more. 

\11 The Interagency GPS Executive Board manages the dual civil and
military use of GPS.  The Board includes representatives from the
departments of Transportation, Defense, Agriculture, Commerce, and
the Interior. 


      COST RISKS IDENTIFIED
---------------------------------------------------------- Letter :3.2

In 1997, we expressed our concern that FAA's cost estimates for WAAS
were firm, discrete-point estimates, implying a level of precision
that could not be supported, particularly early in the project's
development.\12 The Secretary's report addressed this concern by
identifying a range of possible costs and associated probabilities. 

The Secretary's report stated at a high confidence level (an
80-percent probability) that WAAS' 15-year life-cycle cost will not
exceed about $3 billion.  Overall, this estimate is $600 million
higher than the agency's September 1997 estimate.  FAA attributes
this increase to the costs of leasing additional GEO satellites being
higher than expected.  (See table 1.)



                                Table 1
                
                  FAA's Cost Estimates for WAAS at 80-
                Percent Confidence Level, 1997 and 1998

                          (Dollars in milions)

Estimate                            September 1997       February 1998
------------------------------  ------------------  ------------------
Facilities and equipment                      $900            $1,000\a
Operations and maintenance
 Satellites                                    500               1,200
 Other                                       1,000                 800
======================================================================
Total                                       $2,400              $3,000
----------------------------------------------------------------------
\a About $100 million for updating WAAS equipment was shifted to the
February 1998 facilities and equipment estimate from the September
1997 operations and maintenance estimate. 

Source:  FAA. 

We agree with the Secretary's report that the greatest degree of
uncertainty about the WAAS cost estimates surrounds the costs of the
satellites.  The uncertainty exists because FAA does not yet know
exactly how many additional satellites will be needed and how much
the per unit costs will be.  The Secretary's report also states at a
high confidence level that the operations and maintenance cost of
satellites will be no more than about $1.2 billion, or about 40
percent, of the project's total cost of $3 billion.  This estimate
includes about $200 million for the cost of maintaining the leases on
the two existing satellites for which FAA currently contracts with
Comsat and $1 billion for leasing additional satellites.  FAA's cost
estimate assumes that the two satellites leased from Comsat will be
retained and two to four additional satellites (with three being the
most probable number) will be obtained.  The annual unit costs for
the added satellites range from about $12 million to $25 million
(with $17 million being the most probable cost). 

The uncertainty surrounding GEO satellite costs is likely to be
reduced as more data become available.  FAA intends to make a
decision on the number of satellites needed for the full WAAS after
determining the placement of satellites in space and how well the GPS
satellites are performing.  The agency will know what the per unit
costs will be after it comes to an agreement with a vendor for
satellite services.  On January 8, 1998, FAA issued a request for
information seeking input from vendors that would be willing to
finance the costs of designing, building, and launching the GEO
satellites.  FAA would commit to a multiyear lease for satellite
services and reimburse the vendor for its costs.\13

According to the Secretary's report, the agency has targeted April
1998 for issuing a request for proposals to solicit vendors' bids and
July 1998 for awarding a contract. 

Two types of leases are potentially applicable to FAA's satellite
leasing strategy:  the operating lease and the capital lease.  An
issue to be resolved is how budget authority for the satellite
leasing costs will be scored.  According to the scorekeeping
guidelines contained in the Conference Report for the Budget
Enforcement Act of 1997, operating leases for physical assets are
primarily intended to meet short-term capital needs and are to be
used to obtain general purpose equipment (that is, equipment not
built to meet a unique government specification or need) and
equipment that has a private sector market.\14 Capital leases for
physical assets, on the other hand, are intended to be generally
longer term and used to obtain equipment built to meet unique
government-specified needs or uses and leased to the government for
most of its useful economic life. 

FAA's satellite lease would likely be scored as an operating lease if
FAA signs a long-term lease through which the agency leases space on
"hosted" GEO satellites.  In other words, FAA's WAAS payload would
share space on satellites with other users.  Scorekeeping guidelines
require that an agency have sufficient budget authority to cover at
least the cost of the first year of the contract plus any
cancellation fees.  According to WAAS' funding profile, FAA expects
no leasing costs in fiscal years 1999, 2000, and 2001 if the vendor
agrees to cover the costs of building and launching the satellites
and to wait until 2002 for FAA's first payment on the contract. 
FAA's Assistant Chief Counsel, Procurement Law Division, told us that
the agency may enter into contracts without budget authority for its
cancellation fees because FAA has multiyear contracting authority
that exempts it, under certain conditions, from the Anti-Deficiency
Act.\15

FAA's satellite lease would likely be scored as a capital lease if
FAA signs a long-term lease for "dedicated" satellites that would be
built to meet WAAS' specifications and used primarily, if not
exclusively, for WAAS' operations.  In that case, scorekeeping
guidelines require enough up-front budget authority to reflect the
estimated net present value of the entire lease, about $290 million,
in fiscal year 1999, the first year of the contract.\16 Congressional
approval of this amount would result in less budget authority being
available for other programs funded through the appropriations
process in that fiscal year. 


--------------------
\12 The history of FAA's WAAS cost estimates are detailed in two GAO
publications:  National Airspace System:  Questions Concerning FAA's
Wide Area Augmentation System (GAO/RCED-97-219R, Aug.  7, 1997) and
National Airspace System:  Observations on the Wide Area Augmentation
System (GAO/T-RCED-98-12, Oct.  1, 1997). 

\13 Beyond 2009, FAA carried the estimated lease cost forward while
adjusting it for inflation through the remainder of the WAAS life
cycle. 

\14 House of Representatives Conference Report 105-217, Balanced
Budget Act of 1997 (July 29, 1997). 

\15 See 31 U.S.C.  1341(a)(1)(B) and 49 U.S.C.  40111 and 40112. 

\16 This amount represents the present value in 1999 dollars of the
future costs associated with leasing the satellites. 


      SCHEDULE UNCERTAINTIES NOT
      FULLY RECOGNIZED
---------------------------------------------------------- Letter :3.3

Although the Secretary's report discussed risk factors that could
affect the achievement of FAA's schedule goals for developing WAAS,
it fell short of providing a complete assessment.  For example, while
it assigned a 99-percent degree of confidence in meeting various
milestones during fiscal year 1998, the report did not assign
probabilities for milestones for fiscal year 1999 and beyond.  The
agency has set schedule goals for the development of the initial and
full system but has provided no range or confidence levels for
achieving those goals. 

The conferees for the DOT Appropriations Act for fiscal year 1998
required the Secretary of Transportation to provide by February 15,
1998, a detailed report on FAA's plans to provide satellite
communications for WAAS.\17 According to the Secretary's transmittal
letter to the Congress, his Department's report of February 11
included these plans.  In our view, however, the report could have
done more to discuss the uncertainties FAA faces in obtaining the
required GEO satellites. 

As already noted, FAA released a request for information from
satellite vendors on January 8 and has evaluated this information. 
According to the Secretary's report, the agency expects to issue a
request for proposal by April 1998, award a contract to a satellite
provider by July 1998, and complete the launching and testing of the
satellites by October 2001.  By December 2001, only 2 months later,
WAAS is scheduled to become fully operational. 

One major uncertainty is whether FAA will find a vendor willing and
able to complete the launching and testing of the satellites by
October 2001.  In responding to the January 8 information request, a
number of potential vendors pointed to 2002 or 2003 as a more
realistic schedule for putting the satellites in orbit. 

If the GEO satellites are launched after 2001, the resulting delay
would be likely to have implications for the project's benefits and
costs.  Benefits would decrease, for example, because users would not
have a fully operational system available for navigation as early as
expected.  Aircraft operators would not realize some portion of the
$350 million (in net present value) that FAA estimates operators
would save by removing ground-based navigation equipment from their
aircraft.  At the same time, the project's costs would be likely to
increase.  In April 1998, FAA's WAAS program office estimated that a
12-month delay would cost an additional $6 million.\18 This amount
would be needed to pay Raytheon to retain a core staff of system
engineers to complete the integration and testing of the GEO
satellites. 

Another major uncertainty centers on the time needed to award a
contract for the satellites.  If the satellite contract is not
awarded by July 1998 as planned, the remainder of the schedule is
likely to slip.  Contract award by that date, however, is doubtful
for two reasons.  First, negotiations over the terms of the contract
might become protracted as FAA and the vendor seek to minimize their
financial risks.  For example, FAA expects the vendor to invest
hundreds of millions of dollars to cover the costs of building and
launching the satellites.  However, while FAA expects to pay a
premium for the vendor to finance the satellite costs, the vendor may
not wish to carry the costs until FAA begins paying, as planned, in
fiscal year 2002--more than 3 years after the contract is awarded. 
FAA and the vendor will be negotiating on the extent of the
government's financial guarantees.  These guarantees are likely to
take the form of cancellation fees that FAA would pay in the event
the contract is terminated. 

Second, FAA may defer contract award until it receives congressional
approval to enter into a 10-year lease for the GEO satellites.  Under
49 U.S.C.  40111 and 40112, the agency is currently limited to
contracts with a 5-year base period with 3 option years.  To reduce
its costs for the satellite lease, FAA would like to be able to
extend the satellite leasing period from 5 years to 10 years and
intends to seek the authority to enter into multiyear contracts for
an unlimited number of years. 


--------------------
\17 House Conference Report 105-313 (Oct.  7, 1997). 

\18 This estimate assumed that Raytheon would complete software and
hardware development as planned.  If Raytheon's development efforts
are slowed, FAA could incur additional costs.  For example, FAA may
be liable for penalty fees if the agency does not meet the terms of
the Raytheon contract.  According to FAA, a 12-month delay would also
cause FAA to fund satellite and communications costs totaling about
$25 million out of its facilities and equipment account rather than
out of its operations account, as expected, because until a system is
fully operational for as many as 2 years, the agency uses the
facilities and equipment account as the source of funds. 


   UNDER ALTERNATIVE BENEFIT AND
   COST ASSUMPTIONS, WAAS REMAINED
   COST-BENEFICIAL, BUT NET
   BENEFITS WERE REDUCED
------------------------------------------------------------ Letter :4

In making investment decisions, FAA conducts benefit-cost analyses to
determine if the benefits to be derived from acquiring new equipment
outweigh the costs.  FAA's analyses dating back to 1994 have always
found WAAS to be a cost-beneficial investment--that is, the benefits
clearly exceeded the costs.  (See app.  I for details on FAA's
benefit-cost analyses for the WAAS project in 1994, 1996, 1997, and
1998.)

In FAA's benefit-cost analyses, the costs for WAAS included the
future life-cycle costs for facilities and equipment as well as
operations and maintenance costs and the costs for decommissioning
the current ground-based navigation aids, such as very high frequency
omnidirectional ranging (VOR) units.\19 The system's benefits to FAA
included the savings from reduced maintenance of the navigation aids
that are to be decommissioned and the avoidance of capital
expenditures for replacing those aids with new ground-based
equipment.  Aircraft operators--the users of WAAS--also benefit.  The
users' benefits included the reduction of accident-related costs
(from death, injury, and property damage) because the system's
landing signals would be available at airports or runways that
currently lack precision landing capability.  Also, aircraft
operators could benefit by reducing the proliferation of on-board
navigation equipment and receiving savings that result from the
shorter flight times on restructured, more direct routes that
aircraft could fly using WAAS.  Shorter flight times from these more
direct routes also benefit passengers.  Nonaviation benefits were
excluded from FAA's analyses.\20

FAA's investment analysis group prepared the agency's most recent
benefit-cost analysis, in January 1998, to assist FAA in evaluating
whether WAAS was a sound investment.  Unlike previous analyses, FAA's
January 1998 analysis used a risk assessment methodology that
recognized uncertainties and placed confidence levels on each
outcome.  The base case analysis assumed that the two existing
satellites will continue to be leased throughout the WAAS life cycle
and that additional dedicated satellites will be necessary according
to the following probabilities:  two more satellites, 20-percent
probability; three more satellites, 65-percent probability; and four
more satellites, 15-percent probability.  The base case analysis also
assumed that there is a 100-percent probability that all ground-based
navigation systems will be decommissioned by 2010.  In its analysis,
FAA also included the value of the time passengers would save,
assuming a range of savings that generally varied from about 20 to 60
seconds, with the most probable amount being 30 seconds, in
calculating the benefit-cost ratios. 

This analysis found (1) a 20-percent chance (the low confidence
level) that the WAAS benefit-cost ratio could be 4.0 or greater and
(2) an 80-percent chance (the high confidence level) that the ratio
could be 3.0 or greater.\21 Expressed another way, the net benefits
(dollar value of benefits minus costs) of WAAS were $3.4 billion or
greater at the low confidence level and $2.4 billion or greater at
the high confidence level.\22

As discussed previously, it is possible that satellite costs could
increase and that FAA would decide to retain some of its ground-based
navigation systems.  To understand the impact of these possibilities,
we asked FAA's investment analysis group to perform alternative runs
of their benefit-cost analysis using the methodology that they
followed.  The scenarios we requested made the following assumptions: 

  -- a 20-percent probability that the two existing leased satellites
     will continue to be leased throughout WAAS' life cycle and an
     80-percent probability that they will be replaced with one, more
     expensive, dedicated leased satellite;

  -- a 50-percent probability that three additional satellites will
     be needed and a 50-percent probability that four additional
     satellites will be needed; and

  -- a 50-percent probability that 125 VOR units will never be
     decommissioned and a 50-percent probability that 650 VOR units
     will never be decommissioned.\23

DOT's guidance, dated April 9, 1997, directs departmental staff to
include passenger time savings in benefit-cost analyses.\24 The
guidance notes that a controversy exists over whether small
increments of time savings, such as a few minutes or less, should be
valued at the same hourly rate as larger increments.  However, it
concludes that assuming "a constant value per hour for large and
small time savings is probably appropriate." The Director, FAA's
Office of Aviation Policy and Plans, told us that while only small
increments of passenger time savings may result from any one FAA
project, more significant--and clearly valuable--time savings may
result from aggregating the small increments.  Because FAA develops
and implements many aviation projects over a number of years, the
agency would not know the total impact of these projects on passenger
time savings unless all increments were captured in its benefit-cost
analyses.  An official of the Office of Management and Budget (OMB)
told us that her office does not provide specific guidance to federal
agencies about the valuation of small increments of passenger time
savings.  She said that while OMB has not formally endorsed DOT's
April 1997 guidance, OMB's staff do not have any major concerns with
it. 

Concerned that passengers might not perceive and value time savings
of as little as 30 seconds, we reviewed the economic literature about
the validity of using small increments of time and found that no
consensus exists.\25 (See app.  II for a discussion of the
literature.) In the absence of a consensus among experts, we
requested that FAA's investment analysis group perform an alternative
run of its January 1998 benefit-cost analysis base case excluding the
value of small increments of passenger time savings.\26

The results shown in table 2 reflect the use of the alternative
assumptions compared with those in FAA's 1998 base case analysis.  We
found that our alternative cost and decommissioning assumptions alone
did not cause much of a decrease in the benefit-cost ratios and net
benefits.  Excluding small increments of passenger time savings
caused a more pronounced decrease.  For example, we found that at the
high confidence level, net benefits declined by only $0.2
billion--from $2.4 billion or greater using FAA's base case
assumptions to $2.2 billion or greater using our alternative cost and
decommissioning assumptions.  However, the exclusion of small
increments of passenger time savings alone led to a $1 billion
decline in net benefits.  Nevertheless, when the alternative
assumptions are taken together, the system's benefits still exceed
the costs by nearly a 2-to-1 ratio. 



                                         Table 2
                         
                          Effects of Alternative Assumptions on
                            WAAS' Estimated Benefits and Costs

                                  (Dollars in billions)

                           Benefit-cost ratio                    Net benefits\a
                   ----------------------------------  ----------------------------------
                   Low confidence    High confidence   Low confidence    High confidence
                   level             level             level             level
-----------------  ----------------  ----------------  ----------------  ----------------
FAA's 1998     se assumptions
base ca

Including all      4.0 or greater    3.0 or greater    $3.4 or greater   $2.4 or greater
passenger time
savings

Excluding small    2.7 or greater    2.2 or greater    $1.9 or greater   $1.4 or greater
increments of
passenger time
savings


GAO's assumptions used for satellite needs and decommissioning
-----------------------------------------------------------------------------------------
Including all      3.6 or greater    2.8 or greater    $3.3 or greater   $2.2 or greater
passenger time
savings

Excluding small    2.4 or greater    1.9 or greater    $1.7 or greater   $1.1 or greater
increments of
passenger time
savings
-----------------------------------------------------------------------------------------
\a "Net benefits" refers to the net present value of WAAS' benefits
minus its costs. 

Source:  FAA. 


--------------------
\19 VOR units provide signals to aircraft for navigating between
airports and for making nonprecision approaches to airport runways. 
These units are presently the primary radionavigation aid in the
National Airspace System for civil aviation operations. 

\20 Although WAAS will benefit nonaviation users, such as farmers,
boaters, and truckers, these benefits--some of which could be
substantial--were not included in FAA's analyses.  In August 1997,
the federal Differential GPS Executive Steering Group estimated, for
example, that the agricultural industry has the potential to realize
savings estimated at a net present value of $2.45 billion during the
15-year life cycle of a nationwide system augmenting GPS by improving
the ability of farmers to precisely apply herbicides, insecticides,
and fertilizer.  If these additional benefits were included, the
benefit-cost ratio for WAAS would increase. 

\21 A benefit-cost ratio is a measure of the relationship between the
present value of the benefits of a project and its costs.  Since
benefits are divided by costs, any ratio above 1.0 indicates that the
project is cost-beneficial, and any ratio below 1.0 indicates that
the project is not cost-beneficial. 

\22 As an alternative to the benefit-cost ratio, for which the
present value of benefits is divided by the present value of costs,
analysts sometimes calculate the present value of net benefits.  This
value is equal to the present value of benefits minus the present
value of costs.  When using alternative assumptions for calculating
benefits and costs, the present value of net benefits can be a useful
tool for making comparisons. 

\23 In our analysis, we did not take into account what the effect of
retaining ground-based navigation systems would be on other related
benefits.  For instance, if aircraft operators have to retain
equipment as a backup system to mitigate the risk of electronic
interference, equipment and fuel savings linked to removing such
equipment might not be realized.  We did not make any assumptions
about these user benefits, because FAA is still in the early stages
of assessing what a backup system may entail and the impacts on
users. 

\24 The Department values passenger time from $22 to $33 per hour,
depending on the nature of the air travel. 

\25 We previously questioned FAA's use of small amounts of time
savings for passengers in our April 1993 testimony before the
Subcommittee on Transportation, House Committee on Appropriations. 
See Department of Transportation and Related Agencies Appropriations
for 1994, Hearings, Part 6, p.  819. 

\26 Small increments of passenger time savings accounted for most of
the time savings included in FAA's benefit-cost analysis for WAAS. 
Larger increments--such as 15 minutes--accounted for some of the
passenger time savings.  The sensitivity analyses conducted for us
excluded only the small increments. 


   CONCLUSIONS
------------------------------------------------------------ Letter :5

The Secretary's report adequately discussed the risks FAA faces in
achieving the performance and cost goals for the WAAS project.  It
could have done more, however, to recognize the schedule
uncertainties, particularly those related to obtaining the GEO
satellites. 

More information would help the Congress and the administration in
deciding on future investments in the WAAS project.  Information on
the range of milestones for making the system operational and the
probabilities attached to those milestones would aid decisionmakers
in determining the timing of the investments.  Also, a detailed
explanation of FAA's strategy for leasing GEO satellites would help
them in understanding the cost and budgetary implications. 
Particularly useful would be information on (1) the
cost-effectiveness of the hosted and dedicated satellite options and
(2) the estimated premium to be paid for a vendor's financing of the
building and launching of the satellites. 

Even under our alternative assumptions, WAAS' benefits clearly
outweigh its costs.  However, the continued investment in WAAS must
compete with other demands on FAA's capital and operating budgets. 
When more is known about the likely costs for obtaining GEO
satellites and the extent to which the agency may retain existing
ground-based navigation aids, an updated benefit-cost analysis would
help the Congress and administration in making future investment
decisions.  The analysis would be more useful if the agency compared
an investment in WAAS with alternative uses of FAA's resources and
explained the effects of including small increments of passenger time
savings on the benefit-cost ratio and net benefits of the system. 


   RECOMMENDATION
------------------------------------------------------------ Letter :6

To assist the Congress in making future funding decisions for the
Wide Area Augmentation System project, we recommend that the
Secretary of Transportation direct the FAA Administrator to report
information to the Congress on

  -- the range of milestones for making the initial and full Wide
     Area Augmentation System operational and the probabilities
     associated with those milestones;

  -- a detailed explanation of the agency's strategy for leasing
     geostationary satellites; and

  -- updated benefit-cost analyses, including a comparison with
     alternative investments of FAA's resources and an explanation of
     the effects of including small increments of passenger time
     savings. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :7

We provided a draft of this report to the Departments of
Transportation and Defense for review and comment.  We met with
officials from the Office of the Secretary of Transportation and FAA,
including the Director, Communications, Navigation, and Surveillance
(CNS) Systems; the Chairman, Satellite Operational Implementation
Team; the WAAS Program Manager; and the Manager, CNS/Facility
Investment Analysis.  We also spoke with the Assistant for GPS,
Positioning, and Navigation Policy, Office of the Deputy Under
Secretary of Defense (Space).  DOT and DOD generally agreed with our
draft report's findings, conclusions, and recommendations.  They gave
us information and suggestions to help make the report clearer and
more accurate.  We incorporated their suggestions where appropriate. 

DOT expressed concern that the wording in our draft report could
leave the impression that we believe FAA improperly calculated the
WAAS benefit-cost ratio because of the inclusion of small increments
of passenger time savings.  The agency noted that DOT's guidance
directs departmental staff to include all increments--large and
small.  We did not intend to suggest that FAA should not follow DOT's
guidance, and we have added language to the report to clarify this. 
However, our review of the economic literature found a lack of
consensus among experts on the validity of using small increments of
passenger time savings and our sensitivity analysis found that the
inclusion of small increments was significant for WAAS' benefit-cost
ratio and net benefits.  Taken together, these findings argue for
informing decisionmakers about the effects of including small
increments of passenger time savings when future benefit-cost
analyses are conducted for the WAAS project. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :8

To obtain information for this report, we interviewed (1) officials
at FAA headquarters and DOD, including DOD's National Reconnaissance
Office; (2) representatives from Raytheon (previously Hughes
Aircraft), the prime contractor on WAAS; and (3) officials from the
Mitre Corporation who provide technical advice to FAA.  We reviewed
agency documentation on the current schedule, life-cycle costs, and
performance goals for WAAS.  We also reviewed technical reports from
the WAAS contractor and outside experts that discussed the risks and
challenges facing the project.  To identify the potential impact of
differing assumptions on the benefit-cost ratio for WAAS, we asked
FAA to run alternative analyses. 

We performed our review from October 1997 through April 1998 in
accordance with generally accepted government auditing standards.  We
did not assess the reliability of all cost information.  However,
with regard to satellite costs--the major cost item contributing to
increased life-cycle costs--we did satisfy ourselves that the
information being used by FAA is in general agreement with those
estimates provided by outside sources, such as DOD.  Also, while we
did not perform an extensive review of the model used to calculate
benefit-cost ratios, the model used by FAA is widely recognized as an
appropriate economic analysis tool for providing risk-adjusted
benefit-cost ratios. 


---------------------------------------------------------- Letter :8.1

We are sending copies of this report to interested congressional
committees, the Secretaries of Transportation and Defense, and the
Administrator of FAA.  We will also make copies available to others
on request. 

If you or your staff have any questions or need additional
information, please call me at (202) 512-3650 or send email to
[email protected].  Major contributors to this report are
listed in appendix III. 

Gerald L.  Dillingham
Associate Director,
 Transportation Issues


RESULTS OF FAA'S BENEFIT-COST
ANALYSES, 1994, 1996, 1997, AND
1998
=========================================================== Appendix I

The results of the Federal Aviation Administration's (FAA)
benefit-cost analyses of the Wide Area Augmentation System (WAAS)
project in 1994, 1996, 1997, and 1998 are summarized in table I.1. 
On the benefit side, benefits to the government accrue from the
reduced maintenance of the existing, ground-based network of
navigation aids and the avoidance of capital expenditures for
replacing these aids.  Benefits to users--the aircraft
operators--fall into five categories: 

  -- Efficiency benefits derive from having precision landing
     capability at airports where it does not now exist. 

  -- Avionics cost savings reflect how WAAS will enable users to
     reduce the proliferation of avionics equipment in their
     cockpits. 

  -- Fuel savings reflect the use of less fuel to fly aircraft that
     carry less avionics equipment. 

  -- Safety benefits stem from the reduction in accident-related
     costs (death, injury, and property damage) because of the
     availability of WAAS landing signals at airports that presently
     lack a precision landing capability. 

  -- Direct route savings result from the shorter flight times
     associated with restructured, more direct routes that aircraft
     can fly because of WAAS. 



                                        Table I.1
                         
                         FAA's Analysis of the Net Present Value
                            of Benefits and Costs for the WAAS
                           Project, 1994, 1996, 1997, and 1998

                                  (Dollars in millions)

                                                   1997\b                  1998\c
                                           ----------------------  ----------------------
                                                  Low        High         Low        High
Category                 1994      1996\a  confidence  confidence  confidence  confidence
-----------------  ----------  ----------  ----------  ----------  ----------  ----------
Benefits
Government              1,385         943         754         754         680         500
User
Efficiency              1,051         768         148         286         320         270
Avionics                1,312       1,109         546         546         550         340
Fuel                       98          95          13          13          10          10
Safety                    560       1,384         624         624         650         560
Direct route                        5,489         637       4,299       2,820       1,860
=========================================================================================
Total benefits          4,406       9,789       2,722       6,521       4,650       3,600

Costs
-----------------------------------------------------------------------------------------
Research and               \d          \d         540         540         620         730
 development,
 facilities and
 equipment
Operations and             \d          \d         720         720         420         550
 maintenance
=========================================================================================
Total costs             1,081       1,051       1,260       1,260       1,090       1,230
Benefit-cost              4.1         9.3         2.2         5.2         4.0         3.0
 ratio
-----------------------------------------------------------------------------------------
\a Does not add to total because of rounding. 

\b The analysis was released in September 1997. 

\c The analysis was released in January 1998.  Because the
methodology used in this analysis establishes confidence intervals
for each benefit, each cost, and each total, the benefit and cost
items to not add to the totals.  Furthermore, the methodology also
establishes confidence levels for the benefit-cost ratios, and as a
result, the ratios shown here differ slightly from what a direct
calculation would show.  For example, a direct calculation for the
low and high confidence level benefit-cost ratios would be 4.3 and
2.9, respectively. 

\d Not applicable. 

Source:  FAA. 

FAA's September 1997 benefit-cost analysis took a more conservative
approach than previous analyses in estimating the benefit-cost ratio. 
That is, compared with the previous analyses, the assumptions
underlying the September study increased the expected costs of WAAS
and simultaneously reduced the expected benefits, which resulted in a
lower benefit-cost ratio than found in the previous versions of the
study.  The higher costs in the 1997 analysis were largely due to the
inclusion of the costs of decommissioning ground-based navigation
systems that were not included in any earlier versions of the study. 
On the benefit side, several changes in key assumptions led to
reduced expected benefits, including (1) a shorter life cycle for the
project, (2) a reduction in the assumed "saved" costs from phasing
out ground-based navigation systems,\27 (3) a reduction in estimated
safety benefits based on the use of more recent accident data,\28 and
(4) a reduction in the expected flight time savings resulting from
more direct routes.  In addition, the high benefit-cost ratio in the
1997 analysis included passenger time savings.  The low benefit-cost
ratio in the 1997 analysis excluded passenger time savings. 

The January 1998 analysis used a different methodology in calculating
benefit-cost ratios than did the 1997 analysis.  The 1998 analysis
assessed how multiple events, such as a combination of numbers of
satellites and ranges of satellite costs, could affect the
benefit-cost ratio.  This analysis then produced benefit-cost ratios
at high and low confidence intervals.  That is, at the high
confidence level, there is an 80-percent chance that the benefit-cost
ratio could be between 3.0 and 1.0.  Conversely, at the low
confidence level, there is a 20-percent chance that the benefit-cost
ratio could be between 4.0 and 1.0.  This analysis did not, however,
exclude passenger time savings.  The 1997 analysis assessed how
individual events, such as increased satellite costs, could affect
the benefit-cost ratio.  While this analysis did not assign
confidence levels to its benefit-cost ratios in arriving at the 5.2
high estimate and the 2.2 low estimate, it did exclude passenger time
savings in the low estimate. 


--------------------
\27 Specifically, the analysis assumed that old equipment would have
been replaced at a slower rate so that savings from not having to
replace that equipment were reduced. 

\28 Prior to the September 1997 benefit-cost analysis, older data on
accident rates were used.  Since rates of accidents have been
declining with time, use of the most recent data reduced the expected
safety benefits from WAAS. 


ISSUES REGARDING THE VALUATION OF
SMALL INCREMENTS OF PASSENGER TIME
SAVINGS
========================================================== Appendix II

A sizable portion of the calculated benefits of WAAS are from the
time aviation passengers are expected to save once the system is in
place.  However, most of these savings come in small increments of
time--a minute or less per passenger trip.  Concerned that passengers
might not perceive and value time savings of a minute or less, we
requested, as a sensitivity analysis, alternative runs of the WAAS
benefit-cost analysis that excluded these passenger time savings.  We
made this request because we found that there is considerable
controversy and no consensus in the economic literature about whether
travelers perceive and value very small time savings.  We do not
suggest that these benefits should be excluded from the benefit-cost
analysis of WAAS or that FAA should undertake an analysis that is not
in accordance with Department of Transportation's guidance that
directs its staff to include small increments of passenger time
savings in benefit-cost analyses.  However, we believe it is useful
to understand how sensitive the benefit-cost results were to the
inclusion of small increments of passenger time savings. 

This appendix provides information on (1) the value of passenger time
in the WAAS benefit-cost analysis and (2) the issues discussed in the
economics literature on the value of very small increments of time to
travelers. 


   A SIGNIFICANT PORTION OF THE
   CALCULATED BENEFITS OF WAAS ARE
   ATTRIBUTABLE TO PASSENGER TIME
   SAVINGS
-------------------------------------------------------- Appendix II:1

FAA's most recent WAAS benefit-cost analysis found, as is the case
for many transportation improvement projects, considerable benefits
attributable to reduced travel time for passengers.  In the case of
WAAS, about 40 percent of the calculated benefits, or approximately
$1 billion in the base case benefit-cost analysis, are due to time
savings that would accrue to travelers because of slightly reduced
flight times.  FAA officials told us that, on average, these time
savings would probably be about 30 seconds per flight.  FAA's
guidance regarding the valuation of small increments of passenger
time savings suggests that there is no reason, based on either
empirical findings or theoretical concepts, that these small
increments should not be valued at the same per hour rate as larger
increments of time savings. 


   FEW EMPIRICAL STUDIES ON SMALL
   INCREMENTS OF PASSENGER TIME
   EXIST; CONCEPTUAL ARGUMENTS ARE
   CONTRADICTORY
-------------------------------------------------------- Appendix II:2

We reviewed several studies, including an overview study prepared for
FAA, on the issue of the value of small increments of passenger time. 
As FAA's analysis points out, there is limited empirical work on this
issue.  Several studies from the 1970s suggest that travelers may
place little value on very small time savings, such as 1 minute.\29
However, the findings of these studies may have limited applicability
to WAAS.  First, these studies, and most other analyses of this
issue, are focused on intracity commuter travel.  The nature and
characteristics of such travel are very different from intercity air
travel and, accordingly, results from the studies may have little
applicability to how intercity air travelers value time.  Secondly,
the study prepared for FAA discusses the considerable methodological
problems and limitations in these empirical studies.  Because of
these problems, and the lack of studies focused on air travel, the
empirical literature does not provide definitive evidence about how
small increments of time savings are valued by travelers who would
benefit from WAAS. 

Our review of the conceptual arguments regarding the value of small
increments of passenger time revealed a mixed message.  There appear
to be sound conceptual points on both sides of this debate:  Some
suggest that small increments of passenger time savings should be
valued on a pro rata basis just as larger increments of time are;
others suggest that less value should be placed on very small time
savings. 

Those who argue that small time savings have little value suggest
some key reasons.  First, people cannot perceive very small time
savings, and if they cannot perceive them, they do not value them. 
Second, even if a savings of, for example, 1 minute is perceived, it
will not be of value to a person unless that time can be put to some
alternative use.  Because it is likely to take some threshold amount
of time to have value in an alternative use, very small increments of
time cannot be used and are therefore not valued.  Moreover, as the
amount of time savings increases, more potential uses of that time
become available. 

Conversely, several conceptual arguments suggest that there is no
basis for valuing small time savings at less than their pro rata
share of the value of larger time savings.  First, some analysts have
suggested that even if people do not perceive a time savings, they do
place value on it if they put that time to an alternative use. 
Additionally, the issue of needing a threshold block of time for an
alternative use may be true, but this would suggest that people may
always have some "spare" time that cannot be used, and if so, very
small increments of time may, in some cases, push them over the
threshold level and give them a usable block of time.  This not only
suggests that small increments of time may, in some cases, have
considerable value, but it also points out that even spare, or
unusable, time will be valued because there is the possibility of
time savings from some other source that will meet the threshold for
a usable time block.  The final argument for valuing even small time
increments is that transportation improvement initiatives are
somewhat arbitrarily divided into recognized "projects." That is,
across both time and geography, a variety of projects may be
providing incremental time savings that may each be only a small
amount, but when added together become significant.  Hence, it is not
appropriate to view the savings of a given project in isolation of
other projects that might occur a year later or at a different
location. 


--------------------
\29 The studies reviewed in the study prepared for FAA included
Thomas C.  Thomas and Gordon I.  Thompson, "The Value of Time Savings
for Commuting Motorists as a Function of Their Income Level and
Amount of Time Saved," Costs and Benefits of Transportation Planning,
Highway Research Board, Highway Research Record, Vol.  31, No.  4,
(1970), pp.  1-14; David A.  Hensher, "The Value of Commuter Travel
Time Savings," Journal of Transport Economics and Policy, Vol.  10,
No.  2 (May 1976), pp.  167-176; and Ian Heggie, ed., "A Diagnostic
Survey of Urban Journey to Work Behavior," Modal Choice and the Value
of Travel Time (Oxford, England:  Clarenden, 1976). 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix III

Amy D.  Abramowitz
Leslie Albin
John H.  Anderson, Jr.
Robert E.  Levin
John T.  Noto
E.  Jerry Seigler


*** End of document. ***