Forecasting long-term trends in O-D passenger markets

Transcription

1 Forecasting long-term trends in O-D passenger markets Brian Pearce, Chief Economist IATA and Visiting Professor Cranfield th January 2015, Cranfield To represent, lead and serve the airline industry

2 My previous lecture was rather sceptical Source: The Daily Telegraph IATA Economics 2

3 % saar As short-term forecasts fail just when needed Jan 07 Actual and forecast US GDP growth - Fair Model Actual Oct 07 Oct 09 Jul 08 Apr Oct Source: Fair Model, Datastream IATA Economics 3

4 But decisions about the future still required New forecast service: 4,000 O-D country-pairs JV with IATA and Tourism Economics Global, country reports, web-tool For more information: David Oxley, IATA, David Goodger, Tourism Economics, IATA Economics 4

5 Indexed to equal 1 in 1950 So how should we model air travel demand? World scheduled air travel, freight and world real GDP Air cargo (FTK) 212x Air travel (RPK) 212x GDP (constant prices) 10x Source: IATA, ICAO, Haver IATA Economics 5

6 Is RPK=GDP 1.5 the best modelling approach? 7 6 Ratio of RPK growth to GDP growth Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 6

7 But real travel share does not rise with income Travel/gdp World Bank Passengers/GDP Boeing ASM/GDP World Bank Per Capita income (US$87) Source: This slide is borrowed from Bill Swan IATA Economics 7

8 And nominal share does not rise over time 1.2% Spending on air transport as a % of world nominal GDP 1.0% 0.8% 0.6% 0.4% 0.2% 0.0% Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 8

9 Bill Swan s conclusions Cross-sectional data: Travel Share not rising with incomes: Travel Share of GDP measured as ASK/GDP ratio Data shows small negative correlation with per-capita income No acceleration of travel share after joining middle class Time-series data confirmed pattern: Growth of Travel Share was independent of growth of GDP Based on Country-by-Country data Conclusion: Travel grows linearly with GDP growth Remaining 1/3 of travel growth is something else Useful question: What Else? Source: This slide is borrowed from Bill Swan IATA Economics 9

10 US$ in 2013 prices to fly a tonne kilometer US$ in 2013 prices per tonne kilometer Falling transport costs part of what else Unit cost and the price of air transport Boeing oil crisis US deregulation EU deregulation Unit cost (US$/ATK) 0.5 Price (US$/RTK) Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 10

11 Improving technology is a major cost driver Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 11

12 Proportion of countries in each range (%) And constraints on competition are also key 40% Distribution of countries by level of air market liberalization 35% 30% 25% 20% 15% 10% 5% Air markets are less liberal Air markets are more liberal Median WALI scores 2005: : % Weighted Air Liberalization Index (0 = completely regulated, 50 = completely liberalized) Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 12

13 Scope for lower fares on a number of markets Passenger yields, US$ per RPK 0.19 Passenger yields and average trip length Within Europe Within Africa Within South America Within East Asia Within ASEAN ASEAN- E Asia Middle Eastrest of world North-South America North Atlantic Pacific ASEAN-Africa ASEAN-Europe China-North America ASEAN-North America Europe- Australia 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 Average trip length (kms) Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 13

14 Travel costs could fall 30-50% on liberalization -60% Most similar to the biggest differences Estimated reductions in airfares from liberalisation Most similar to the biggest differences -50% -40% -30% -20% -10% 0% My estimation of what is possible using WTO report data (2006) Agenda for Freedom reports (combined liberalisation) Africa liberalisation report Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 14

15 And price elasticities vary significantly -2.0 David Gillen's air travel own-price demand elasticities (selected measures) All estimates Long haul Long haul business Long haul leisure Short/medium haul Short/medium haul business All short/medium haul leisure Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 15

16 Living standards matter (not aggregate GDP) Trips/ year/ person Months before next trip Low-income Middle-income High-income Below US$20K per person Above US$20K per person Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 16

17 Flights per capita* (2013, logarithmic scale) There is a relationship but not the usual one New Zealand Maldives Bahamas Singapore Hong Kong Norway Iceland Switzerland Seychelles Australia UK Luxembourg Brazil Sweden 1.00 Qatar USA Japan France 0.10 Russia China 0.01 Uganda GDP per capita (2013, US$ '000s) Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 17

18 Flights per capita* (2013, logarithmic scale) They are certainly key for emerging markets Belize Maldives Seychelles 1.00 Indonesia Malaysia Barbados Russia Uruguay 0.10 China Brazil India Gabon Slovakia Swaziland 0.01 Uganda 0.00 Middle-income High-income Low-income GDP per capita (2013, US$ '000s) Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 18

19 Countries becoming middle or high income Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 19

20 Demographics are important to get right % change % Demographic change, % 15% Population change 'Flying' population change 10% 5% 0% -5% -10% -15% -20% -25% India Indonesia Turkey US Brazil France China Germany Japan Russia Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 20

21 Emerging market population much younger Old-age dependency ratio, % (65+/15-64) Japan Germany US China Brazil Turkey Indonesia India Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 21

22 Working-age population makes more trips 2.5 Ratio of air passengers by age group to population by age group N/A* Under to to to 55 Over 55 LAX surveys (average 2006 and 2011) UK CAA (2012) Eurostat (average 2006 and 2011)* * Note: the Eurostat data shown in the 36 to 45 age group are actually for the wider age group Source: UK CAA IATA Economics 22

23 100+ Year 1990 Total population = 122.2m IATA Economics 23

24 100+ Year 1990 Total population = 122.2m IATA Economics 24

25 100+ Year 1991 Total population = 122.7m IATA Economics 25

26 100+ Year 1992 Total population = 123.2m IATA Economics 26

27 100+ Year 1993 Total population = 123.7m IATA Economics 27

28 100+ Year 1994 Total population = 124.1m IATA Economics 28

29 100+ Year 1995 Total population = 124.5m IATA Economics 29

30 100+ Year 1996 Total population = 124.8m IATA Economics 30

31 100+ Year 1997 Total population = 125m IATA Economics 31

32 100+ Year 1998 Total population = 125.3m IATA Economics 32

33 100+ Year 1999 Total population = 125.5m IATA Economics 33

34 100+ Year 2000 Total population = 125.7m IATA Economics 34

35 100+ Year 2001 Total population = 126m IATA Economics 35

36 100+ Year 2002 Total population = 126.2m IATA Economics 36

37 100+ Year 2003 Total population = 126.5m IATA Economics 37

38 100+ Year 2004 Total population = 126.8m IATA Economics 38

39 100+ Year 2005 Total population = 127m IATA Economics 39

40 100+ Year 2006 Total population = 127.1m IATA Economics 40

41 100+ Year 2007 Total population = 127.2m IATA Economics 41

42 100+ Year 2008 Total population = 127.3m IATA Economics 42

43 100+ Year 2009 Total population = 127.4m IATA Economics 43

44 100+ Year 2010 Total population = 127.4m IATA Economics 44

45 100+ Year 2011 Total population = 127.3m IATA Economics 45

46 100+ Year 2012 Total population = 127.2m IATA Economics 46

47 100+ Year 2013 Total population = 127.1m IATA Economics 47

48 100+ Year 2014 Total population = 127m IATA Economics 48

49 100+ Year 2015 Total population = 126.8m IATA Economics 49

50 100+ Year 2016 Total population = 126.6m IATA Economics 50

51 100+ Year 2017 Total population = 126.3m IATA Economics 51

52 100+ Year 2018 Total population = 126.1m IATA Economics 52

53 100+ Year 2019 Total population = 125.7m IATA Economics 53

54 100+ Year 2020 Total population = 125.4m IATA Economics 54

55 100+ Year 2021 Total population = 125m IATA Economics 55

56 100+ Year 2022 Total population = 124.6m IATA Economics 56

57 100+ Year 2023 Total population = 124.2m IATA Economics 57

58 100+ Year 2024 Total population = 123.7m IATA Economics 58

59 100+ Year 2025 Total population = 123.3m IATA Economics 59

60 100+ Year 2026 Total population = 122.8m IATA Economics 60

61 100+ Year 2027 Total population = 122.3m IATA Economics 61

62 100+ Year 2028 Total population = 121.7m IATA Economics 62

63 100+ Year 2029 Total population = 121.2m IATA Economics 63

64 100+ Year 2030 Total population = 120.6m IATA Economics 64

65 100+ Year 2031 Total population = 120.1m IATA Economics 65

66 100+ Year 2032 Total population = 119.5m IATA Economics 66

67 100+ Year 2033 Total population = 118.9m IATA Economics 67

68 100+ Year 2034 Total population = 118.3m IATA Economics 68

69 100+ Year 2035 Total population = 117.7m IATA Economics 69

70 Trade openness also matters for air travel Country The top 20 The bottom 20 Trade intensity Country Trade intensity Rank* Country Trade intensity Hong Kong Brazil Germany 85.8 Singapore United States Mexico 64.5 Luxembourg Pakistan Canada 64.1 Seychelles Central African Rep Turkey 60.6 Hungary Nigeria United Kingdom 60.3 Ireland Sudan Italy 58.3 United Arab Emirates Venezuela France 58.3 Slovak Republic Argentina India 55.5 Swaziland Colombia Russia 50.8 Estonia Japan Indonesia 50.5 Malta Myanmar China 48.6 Belgium Benin Japan 37.8 Maldives Cuba United States 30.2 Lithuania Martinique Brazil 27.9 Czech Republic Egypt 41.2 Vietnam Cameroon 42.0 Netherlands Australia 42.2 Slovenia Burundi 43.4 Malaysia Peru 45.4 Thailand Iran 47.5 The G7 and selected emerging markets Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 70

71 Natural and cultural assets matter too The highest 20 The bottom 20 Country Index level Country Index level Rank* Country Index level United States Haiti United States Switzerland Lesotho United Kingdom United Kingdom Burundi Canada Canada Sierra Leone Germany Australia Chad Japan Spain Swaziland France Germany Guinea Brazil Sweden Moldova China Austria Kuwait Italy Japan Mozambique Mexico France Burkina Faso India Brazil Yemen Turkey China Benin Indonesia Italy Madagascar Russia Mexico Bangladesh 81.1 Netherlands El Salvador 81.1 Malaysia Algeria 81.4 Belgium Mali 82.1 Portugal Paraguay 82.4 Korea, Republic Of Kazakhstan 82.6 The G7 and selected emerging markets Source: WEF Travel and Tourism Competitiveness Index human, cultural and natural resources sub-index IATA Economics 71

72 The TE-IATA econometric model Equations follow a standard Error Correction Mechanism (ECM) structure involving both short-run growth dynamics and convergence to longer-run level terms The levels relationships are crucial to the twenty-year outlook Short-run dynamics follow the current expected economic cycle and largely include economic drivers A primary driver of long-run levels is income and the share of population able to afford travel Other structural factors will also be accommodated within the long-run equation IATA Economics 72

73 Short-run equation Passengers between country AA and country BB D(LN(PAXAABB)) = x*d(ln(gdp,aa)) + (1-x)*D(LN(GDP,BB)) + y*( x*ln(up,aa/upave,bb) + (1-x)*LN(UP,BB/UPAVE,BB) ) + z*( x*d(ln(rxaabb))-(1-x)*d(ln(rxaabb)) ) + z1*d(ln(rxtour,aa)) + z2*d(ln(rxtour,bb)) + a* D(LN(PRAABB)) -0.5 * ( LN(PAXAABB(-1)) LN(LRAABB(-1)) ) GDP Unemployment Exchange rate Relative RX Price ECM Assumed GDP elasticity = 1 x share of travel from country AA on AABB flow (from TDM) y elasticity of unemployment (0.1) z elasticity of own nominal exchange rate (0.3) z1,z2 elasticity of cross exchange rate (nominal) ECM coefficient of 0.5 ensures rapid convergence to long-run (90% in 3 years) LRAABB = long-run relationship IATA Economics 73

74 Long-run equation components Air passenger volumes are modelled according to the following, split overall travel population factors and bilateral route factors Overall travel population factors GDP per capita / wealth effects Population and demographics Trade linkages Bilateral route factors Exchange rates Air fares (includes dynamics of greater access) Destination attractiveness (share of rising outbound) Visa facilitation measures IATA Economics 74

75 Adding up to differing outlooks for Drivers of additional passenger numbers, million Living standards Demographics Price and other Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 75

76 Resulting in much change over next 20 years Rank by size of O-D passenger flows in, out & within country US rank rank 0 1 China China 2 2 US UK 3 3 India Japan 4 4 UK Germany 5 5 Brazil Spain 6 6 Indonesia France 7 7 Spain Italy 8 8 Germany India 9 9 Japan Brazil France Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 76

77 Uncertainty can also be explored 9,000 Outlook for worldwide O-D passenger trips, million 8,000 7,000 6,000 5,000 4,000 Liberal policies scenario Current policies Closing borders scenario x growth billion additional O-D passenger trips 3,000 2, Source: IATA/Tourism Economics Air Passenger Forecasts IATA Economics 77

78 For more information. New forecast service: 4,000 O-D country-pairs JV with IATA and Tourism Economics Global, country reports, web-tool For more information: David Oxley, IATA, David Goodger, Tourism Economics, IATA Economics 78