The National Center of Excellence
Federal Aviation
Administration
For Aviation Operations Research
2nd National Airspace System
Infrastructure Management Conference
NAS Infrastructure in Transition
June 13, 2006
Richard Golaszewski
GRA, Incorporated
University of California Washington Center
Washington, DC
115 West Avenue • Jenkintown, PA 19046 • USA
℡ 215-884-7500 • 215-884-1385
richg@gra-inc.com
Acknowledgement
The following presentation includes work performed by
a number of organizations and persons supporting
the JPDO Evaluation and Analysis Division
GRA, Incorporated
GRA, Incorporated
June 13, 2006
1
Background
4 Questions on value of NGATS
ƒ Funding bodies
ƒ Users
4 Potential need for regulatory changes
ƒ Require benefit-cost analysis (BCA)
4 Justification for investment program
ƒ NASA R&D
ƒ FAA JRC
ƒ OMB 300
4 Sound program management must understand benefits and
costs to government and users
4 Need to consider interdependencies
ƒ User equipage
ƒ Benefits estimates
GRA, Incorporated
June 13, 2006
2
Approach to Long-Term
Investment Analysis
4 Understand performance of future system without NGATS
investment
ƒ What is already underway (e.g., OEP)?
ƒ Secular improvement in ATM system productivity
4 Models abstract futures that can be highly divergent
ƒ 2 X capacity in 2025 vs. 3 X capacity in 2025
4 Order of investment can affect results (some capacity benefits
overlap)
4 Marginal delays without investment can be quite large—expand
demand while holding capacity constant
4 In sector with very long lived physical assets (such as aircraft),
sequencing and coordination of implementation plans are essential
GRA, Incorporated
June 13, 2006
3
Cost-Benefit Approach for JPDO
4 Differences from internal FAA cost benefit analysis
ƒ Multi-agency
ƒ Longer time horizon
ƒ Not only discounted “cash-flow” analysis (traditional internal FAA
process)
4 Provides reference and context for all agencies participating in
NGATS
4 Provides platform for input of industry on specific
product/segment costs
4 Uses modeling and simulation platforms to estimate benefits of
acquiring package(s) of improvements
4 Allows JPDO to respond to multiple needs
ƒ
ƒ
ƒ
ƒ
FAA (ATO-F, ATO-P, APO, JRC, etc.)
NASA (PART)
Other agencies
OMB
GRA, Incorporated
June 13, 2006
4
Identifying Benefits and Costs
4 Both intangible and tangible benefits and costs should be
recognized
4 Calculation of net present value should be based on
incremental benefits and incremental costs
4 Possible interactions between the benefits and costs being
analyzed and other government activities should be
considered
4 Analyses should focus on benefits and costs accruing to the
citizens of the United States; impacts outside of U.S. economy
noted separately
4 There are no economic gains from a pure transfer payment
matched by the costs borne by those who pay for it
GRA, Incorporated
June 13, 2006
5
Measuring Benefits and Costs
4 The principle of willingness-to-pay
4 Market prices
4 Externalities, monopoly power, and taxes or subsidies can distort
market
4 Inframarginal benefits and costs—The economist’s concept of
consumer surplus measures the extra value consumers derive from
their consumption compared with the value measured at market prices
4 Indirect measures of benefits and costs—Most reliable when they are
based on actual market transactions
4 Multiplier effects—Employment or output multipliers that purport to
measure the secondary effects of government expenditures on
employment and output should not be included in measured social
benefits or costs
GRA, Incorporated
June 13, 2006
6
NGATS Benefit-Cost Framework
With NGATS
Without NGATS
GRA, Incorporated
Noise
Emissions
Congestion
Safety
Security
Spillovers
Unpriced
Benefits
and Costs
Passengers
Shippers
Aircraft Operators
Airports
ATM Providers
Aircraft Manufacturers
Other
Private
Benefits
and Costs
Net
Benefits
of
NGATS
(NPV of all
benefits
and costs)
June 13, 2006
7
Issues in Estimating NGATS Costs
4 Projecting O&M costs needs estimates of labor input to future
systems
4 What will be primary and backup systems?
4 What will be required to operate in various categories of
airspace?
4 Will equivalent throughput increase?
ƒ Reduced separation
ƒ Reduced runway occupancy time
GRA, Incorporated
June 13, 2006
8
Illustration of Cost Tradeoff for NAS
NAS Cost
$/Op
affect
Capital Stock
Investment
Labor Productivity
Wage Rates
NGATS
For a given
NAS Capacity,
Operations
and Demand
Less Capacity in Relation to Demand
Base
More Capacity in Relation to Demand
Fares (airline costs)
Schedules
Delay Costs
Access Time/Cost
Airport Costs
Safety/Security
GRA, Incorporated
affect
$ Full Price
of Travel
June 13, 2006
9
Changes to Avionics Could Be
Major Cost Impact of NGATS
4 Many capabilities may require new equipage
4 Fleet has long turnover
4 GA fleet has a large number of old aircraft with low utilization
4 New deliveries cover market growth and retirements
4 About 50% of existing high performance fleet will still be
operated in 2017
4 Will equipage and retrofit strategies be designed to minimize
cost impact?
ƒ Related systems
ƒ Out of service costs
GRA, Incorporated
June 13, 2006 10
High Performance Fleet Forecast
and Retained Fleet Summary
45,000
40,000
35,000
30,000
Deliveries for Growth
25,000
Deliveries to Replace Retired A/C
20,000
15,000
10,000
5,000
0
2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Total Fleet Forecast
Total Retained Fleet
GRA, Incorporated
June 13, 2006 11
What Are NGATS Objectives and
Benefits?
4 Make service like today’s (in terms of transit times, reliability,
etc.) possible at fares like today’s
ƒ Also able to serve tomorrow’s demand level
ƒ Maintaining today’s service delivery quality seems pedestrian,
but requires challenging innovation to achieve
4 Make tomorrow’s service even better
ƒ Improving service qualities while also meeting future demand
levels is even more challenging
4 Most users of aviation system (passengers) value service
qualities like reliability of schedules and curb to curb times,
but are indifferent to the means used to achieve them
GRA, Incorporated
June 13, 2006 12
Economic Costs of a Shortfall
Increased Schedule Times
Embedded Delays
Delay Costs in
Relation to Schedule
Higher Fares
Lost Trips
Increased Travel Times and Increased
Unpredictability of Travel Times
Increased Out-of-Pocket Cost of Travel
Reduced Consumer Choice
Losses in
Consumer Surplus*
*Difference between what consumers would be willing to pay and what they have to pay for a
given quality level – measure of economic consequences used and recommended by OMB
GRA, Incorporated
June 13, 2006 13
Consumer Surplus
Consumer Surplus
Some consumers would be willing to pay
more, but are obliged to pay only the
market price (which allows them to use the
savings for other goods and services).
The difference between what some
consumers are willing to pay and the
market price they are obliged to pay
makes up “consumer surplus.”
Price ($)
P* = Market Price
Demand ~=~ Willingness to Pay
Quantity
Demanded at the
Market Price
GRA, Incorporated
Quantity (RPMs)
June 13, 2006 14
Balancing Supply and Demand
in Benefit-Cost Analysis
4 Currently the baseline method used by EAD analysts to
balance demand and capacity is to “eliminate” flights
4 In a market driven system, these flights will never occur
because passengers and airlines will adapt to the new
reality (higher delays, lack of capacity, etc.)
4 Added delays in the system over a long time affect the
aviation demand function
4 Former aviation passengers shift to competing modes
(auto, rail, etc.)
ƒ For example, drops in short-range commercial flights after
9-11 with added processing-slack times
4 EAD has models to study mode choice shift effects
GRA, Incorporated
June 13, 2006 15
From Unconstrained Demand
to Feasible Throughput
Unconstrained Flight
Schedule & Trajectories
START
Flights
Eliminated from
Forecasted
Schedule
Airport and Sector
Capacities
Demand and Capacity
Compared; Delays Calculated
NO
Delays
Tolerable?
YES
Feasible Flight
Schedule, Trajectories
END
4Future flight schedules would incur huge and unrealistic
delays if all demanded flights actually flew
4Excessive delay must be dealt with in the planning stage
4Capacity constraints will restrict the demand
4Solving the congestion problem
ƒ Alternative route
ƒ Alternative departure time
ƒ Flight elimination (some demand is left unsatisfied)
GRA, Incorporated
June 13, 2006 16
Economic Valuation
4 A “constrained” schedule is produced in
which not all of the demanded flights
actually materialize
Delays
Tolerable?
YES
Constrained Flight
Schedule, Trajectories
RPMs Flown
Calculated
END
Price Change for RPMs Flown,
Value of RPMS Lost/Gained,
Delay Cost
4 The flights that were eliminated have
economic value
4 We translate these lost flights into lost seats
(revenue passenger miles)
4 We estimate the yield increase necessary to
match demand with constrained supply of
RPMs
4 We value the lost RPMs using the concept
of “consumer surplus”
4 We also quantify the cost of the delay that
will exist due to capacity shortages
ƒ Airline variable operating costs
ƒ Passenger value of time
GRA, Incorporated
June 13, 2006 17
Illustration of Constraints Analysis
Yield and Consumer Surplus
836 881
20
yield (2005 cents)
19
946
1072
1220
1332
1625
19.3
Constrained by All
18
Noise Constrained
17
Emissions Constrained
18.3
Fares Higher Due to Scarcity
17.1
16
15
15.1
Terminal-Constrained
demand curve at 3X RPM
14
13
12
12.1
Enroute-Constrained
11
10
APO Forecast
9
10.6
2025
600
GRA, Incorporated
800
Extrapolated from APO
1000
1200
1400
RPMs (billions/fiscal year)
3X RPMs
1600
9.9
1800
June 13, 2006 18
Incremental Successes with OIs
Lead to Incremental Capacity Gains
NGATS will reduce or remove constraints over time
19
18
FAA Forecast (Mainline+Regional)
17
Extrapolated from APO
16
NGATS Capacities - Notional!
S
AT
NG
yield (2005 cents)
Constrained
15
14
13
(Actual values from current portfolio
to be evaluated this summer)
2004
12
2017
11
2025
2032
10 Elasticity = -1
600
GRA, Incorporated
The consumer
surplus benefits
will actually be
realized year by
year as OI
successes deliver
partial NGATS
capabilities
800
1000
1200
1400
RPMs (billions/fiscal year)
1600
1800
June 13, 2006 19
Illustration of Benefits
Note: Order likely to change relative magnitudes
20
Nothing solved
19
$8.6B
18
$11.0B
yield (2005 cents)
17
16
$18.0B
15
$2.1B
Emissions solved; +$8.6B consumer surplus
(const 2005$)
Noise solved; +$19.5B total
Noise, emissions solved; +$37.6B
Noise, emissions, enroute solved;
+$39.7B
demand curve at 3X RPM
14
$36.1B
13
Noise, emissions,
terminal solved;+$75.8B
12
$32.5B
11
All solved
+$108.3B
10
9
0
200
GRA, Incorporated
400
600
800
1000
1200
RPMs (billions/fiscal year)
1400
1600
1800
June 13, 2006 20
Wrap Up
4 Benefit-cost analysis will be a continuing need
4 EAD building “tool sets” using multiple models
ƒ NAS modeling/simulation
ƒ Environmental modeling
ƒ Security models
ƒ User impacts
4 Opportunity to extend and improve techniques
4 Need to retain linkages to agency investment analysis
standards
GRA, Incorporated
June 13, 2006 21
Back Up
Traditional Benefit-Cost Analysis
4 Used to justify NAS investments
ƒ Most appropriate tool
ƒ Meets OMB investment analysis standards
4 Should consider all benefits
ƒ
ƒ
ƒ
ƒ
Airline (or other user) costs and revenues
FAA investment and operations and maintenance costs
Value of passenger time (not always included but should be)
External effects
• Environment
• Congestion
• Safety
4 Compensation principle
ƒ Pareto improving: benefits exceed costs and winners could
compensate losers
GRA, Incorporated
June 13, 2006 23
Quantitative Analysis of Uncertainty
4 Quantitative analysis characterizing the probabilities of the
relevant outcomes and an assignment of economic value to
the projected outcomes
4 Balance throughness with the practical limits on your analytical
capabilities
4 Estimates cannot be more precise than their most uncertain
component
4 Disclose qualitatively the main uncertainties
4 Use a numerical sensitivity analysis
4 Apply a formal probabilistic analysis of the relevant
uncertainties
GRA, Incorporated
June 13, 2006 24
Rise in Average Yield to Model Impact of
Reduced Supply in Constrained World
Case 1
All of Price Rise Reflects
Pure Producer Rent
Airline Profits
Wages
Airports
ATC
Today
Transform
Social Surplus
Goes to
Consumers
Do Not Transform
No Real Δ Cost;
Capture of Rents by
Producers for Airlines or
Other Factors
Equivalent Amounts of Social Surplus
GRA, Incorporated
Case 2
Price Increases
Reflect Increased
Resource Costs
Airline Costs
Airports Costs
ATC Costs
Other Costs
Today
Transform
Reduced Costs
Increase
Consumer Surplus
Do Not Transform
Higher Costs
Increase
Average Fares
and Reduce
Consumer Surplus
Transformation Leads
to Real Resource Savings
June 13, 2006 25
What Drives NAS Demand/Capacity
Requirements?
4 Commercial passenger NAS users (airlines) compete in several
areas to satisfy passenger demand for air transportation services
ƒ
ƒ
ƒ
ƒ
ƒ
Fares
Flight frequency
Travel time
Expected delay time and reliability
Flight comfort and amenities
4 Demand growth is not uniform across all airports—unique
regional patterns of population and economic growth lead to
different levels of demand growth at individual airports
4 Airlines and other NAS users provide personal and commercial
services through flights
ƒ Flights are a common measure of demand/capacity, but, as shown
on the next slide, several related measures are also used
GRA, Incorporated
June 13, 2006 26
Quality Factors Affect Aviation Demand
4 Perceived safety/security
4 Schedule frequency / schedule delay (when can I travel,
relative to when I actually want to go)*
4 Delays/schedule reliability*
4 Back up choices (what happens if I miss flight)
4 Frequent flyer and other amenities
GRA, Incorporated
June 13, 2006 27
Demand Capacity Analysis Metrics
4 Our composite capacity metric is “feasible throughput” which
is measured in terms of number of flights
ƒ Flights eliminated based on delay tolerance
4 “Unconstrained demand” represents the public’s desire for air
transportation
ƒ The FAA’s Terminal Area Forecast does not consider whether future
NAS capacity will be sufficient to accommodate all the demand
ƒ Capacity constraints will force some of the demand to be left
unsatisfied
ƒ Confounded by lack of prices for infrastructure
4 Capacity is location specific as far as terminals/portals are
located (door-to-door)—different terminals have different levels
of service and are more or less appealing in meeting demand—
there is a lot of excess capacity at airports in terms of both
locational and time-of-day demand
GRA, Incorporated
June 13, 2006 28
NGATS Benefit as Consumer Surplus
Yield
Y++
Increased
Market
Clearing
Prices if
Capacity
Growth is
not
Realized
Baseline
NGATS benefit is the continuing
ability of consumers to derive
consumer surplus* by circumventing
upward pressure on yields (and other
travel costs) that would otherwise
follow from capacity shortfalls
Y+
OEP+
Notional NGATS
Y
D2025
D2006
*Consumer surplus is the
difference between consumer
willingness to pay for a good or a
service and the market price
they must pay. It is the measure
of benefits preferred by OMB
and other federal agencies.
D2014
RPMs
Capacity Growth
Enabled by NGATS
GRA, Incorporated
June 13, 2006 29
Overall Benefit-Cost Framework
JPDO
Research
NASA
FAA
DOD
TSA
Industry
Others
GRA, Incorporated
Infrastructure
Investment
Infrastructure
O&M
Unpriced
Benefits
and Costs
Vehicles
Investment
Aircraft
Operators
O&M
Private
Benefits
and Costs
Net
Benefits
of
NGATS
June 13, 2006 30