Air Traffic Control:
encouraging technology adoption &
overcoming weak regulation
Nicole Adler
Stef Proost
Madrid, 7th of March 2017
Adler & Proost, Deliverable 4.1, COMPAIR
2
Motivation
 present ATC system in EU is composed of 37 national
providers
 compared to FAA, EU system is 34% more costly (2011)
 barriers to cost efficiency:
 ownership form: governmental organizations
 fragmentation: missing economies of scale
 protectionism: power of Labor Unions & member states
 weak regulation: failure to implement FABs or strict pricecaps
 barriers to increasing capacity:
 opposition to change
 fear of technology
 relatively low congestion currently
Adler & Proost, Deliverable 4.1, COMPAIR
3
how could cost efficiency and technology
adoption be encouraged simultaneously?
 changes in ownership form
 horizontal integration
 vertical integration
 privatization
 changes in pricing regulation
 strict individual price-caps
 peak / off peak charges
 no regulation
 changes in capacity
 SESAR technologies
Adler & Proost, Deliverable 4.1, COMPAIR
4
Outline of talk
•Methodology to analyse aviation market
 2-stage game
•Case study
 Western Europe
•Conclusions & Future Research
Adler & Proost, Deliverable 4.1, COMPAIR
5
2 Stage Game
 Stage 1: Air Traffic Control Providers set
prices, labor & technology levels
 ANSPs set peak/off-peak charges
 may be price capped according to regulatory rules
 Capacity = f(labor levels, technology investment)
 ATC terminal limits flights in peak
 form of slot allocation
 Auction:
 sealed bid, lexicographic
 1st peak price; 2nd off-peak price; 3rd home bias; 4th capacity
 complete information
 combinatorial with inter-dependent valuations
Adler & Proost, Deliverable 4.1, COMPAIR
6
2 Stage Game
 Stage 2: Airlines choose flight paths given schedules
 3 cost components: operational, congestion & ATC charges
 All cost components impacted by ATC provider decisions in 1st stage
 Revenue loss: flying off-peak lowers airfares
 Option to ‘not fly’ necessary for demand elasticity
 Note:
 Congestion is non-linear
 Closer to capacity: the higher the delays
Adler & Proost, Deliverable 4.1, COMPAIR
7
3 scenarios: with(out) tenders
 With tenders: limit on number of auctions permitted to participate
 Business as Usual




government organization
set charges according to price caps
maximize labor
examples: DFS (Germany) and DSNA (France)
 Non-profit public companies




set charges to cover costs
maximize capacity according to company charter
airlines on Board of public company
example: NavCanada
 For-profit private companies
 maximize profits
 example: NATS (public-private partnership in UK)
Adler & Proost, Deliverable 4.1, COMPAIR
8
ATC en-route production function
• Estimate stochastic production function
using data from PRB
• 37 air navigation service providers
• 11 years of data; 399 observations
• Output: IFR km controlled
• Inputs: ATCOs, airspace
• Explaining inefficiency: complexity
Adler & Proost, Deliverable 4.1, COMPAIR
9
En-route Stochastic Production Function
Estimates
ACC1
ACC2
ACC3
ACC4
Total IFR km
controlled
Total IFR km
controlled
En-route
Input
Output
Labor in ACC
Total IFR km
controlled
0.86
Total IFR km
controlled
40.45
Size of airspace
En-route sectors
constant
0.00
2.78
14.65
117.46
0.87
37.61
0.85
38.24
0.83
40.86
0.05
2.45
0.07
3.30
0.06
3.41
14.77
73.36
14.44
102.45
14.54
118.48
-1.29
-4.33
-1.30
-7.29
3.85
5.13
Z - Variables explaining the mean of the inefficiency (Mu)
Complexity
-1.31
-4.20
Variability
Log Likelihood
Lambda
-249.70
0.73
T1
-283.17
3.54
4.59
-248.23
10.82
T2
Terminal
Adler & Proost, Deliverable 4.1, COMPAIR
0.89
-221.99
4.50
0.73
8.52
T3
10
Outline of talk
•Methodology to analyse aviation market
 2-stage game
•Case study
Western Europe
•Conclusions & Future Research
Adler & Proost, Deliverable 4.1, COMPAIR
11
Case Study of Western Europe
7
33
23
34
28
27
1
15
30
13
16
17
31
5
8
2
10
12
4
11
29
14
3
18
25
19
24
20
21
22
32
6
35
Adler & Proost, Deliverable 4.1, COMPAIR
9
26
12
Players
 Up to 12 potential en-route air navigation service providers:
 serve ~50% of EU traffic
 two from each country
 U.K.
 Netherlands
 Germany
 Spain
 Belgium
 France
 5 Airlines:
 3 alliances:
 Star (Lufthansa)
 Oneworld (BA)
 SkyTeam (AF-KLM)
 Low cost carrier (EasyJet)
 Unaligned carrier (Emirates)
Adler & Proost, Deliverable 4.1, COMPAIR
13
Preliminary Results: no tender (1/5)
 Labor rent seekers
 Labor levels higher than the 5,000 currently employed; std. technology level
 Today profits are approx. 20%, not allowed in model
Adler & Proost, Deliverable 4.1, COMPAIR
14
Preliminary Results: with tender (2/5)
For Profit ATC providers
• Only 3 companies remain
 Germany/Holland; UK/Belgium; France/Spain
• SESAR technologies adopted in full
• Revenues halved compared to current equilibria outcome
• Profits around 5%
Adler & Proost, Deliverable 4.1, COMPAIR
15
Preliminary Results: with tender (3/5)
For Profit ATC providers with Increased Demand
Adler & Proost, Deliverable 4.1, COMPAIR
16
Preliminary Results: with tender (4/5)
Impact on Airlines
Adler & Proost, Deliverable 4.1, COMPAIR
17
Preliminary Results: with tender (5/5)
Non-profit ATC providers
•Only 3 companies remain as with for-profit
• Germany/Holland; UK/Belgium; France/Spain
•Charges different than for-profit but overall revenues lower
•Employ SESAR technology & more ATCOs than for-profit
Adler & Proost, Deliverable 4.1, COMPAIR
18
Conclusions
•Modeling air traffic control via 2-stage game enables
cost-benefit analysis including distributional effects
across stakeholders
•New technology (capacity)
 Given cost of technologies according to 2012 Eurocontrol
Masterplan, likely to be adopted ONLY if ownership form
changes
Adler & Proost, Deliverable 4.1, COMPAIR
19
Conclusions: Ownership form
 Without tenders:
 Non-profits provide highest capacities
 utilize technologies & high labor levels
 prices close to price caps
 For-profits create capacities similar to current levels
 utilize technology with lower labor levels
 prices = price caps; profits of 25%
 With tenders:
 Very important: multiple bidders
 Single participant in each airspace leads to no equilibria
outcome
 Need to set minimum capacities within tender
 Leads to 3 companies serving 6 airspaces in case study
 Permits defragmentation of European airspace
 Prices halved
 For-profits set higher prices & lower capacities than non-profits
Adler & Proost, Deliverable 4.1, COMPAIR
20
Air traffic control:
encouraging technology adoption & overcoming weak regulation
Thank you very much
for your attention!
This project has received funding from the SESAR Joint Undertaking
under the European Union’s Horizon 2020 research and innovation
programme under grant agreement No 699249