Transcript “Air travel demand and income: empirical investigation and future

Air Travel Demand and Income: Empirical
Investigations and Future Scenarios
Shoibal Chakravarty and Massimo Tavoni
Princeton Environmental Institute,
Princeton University
International Energy Workshop
Venice, Italy.
19 June 2009
Motivation
•
Meeting transportation demand sustainably is an important future challenge
•
Aviation:
– Small compared to road transport
– Growing fast (5.5%/yr over the past 20yrs), expected to grow at a rate twice that of road
mobility
– Concentrated in rich countries (OECD have ¾th share)
– Additional concerns (noise, safety, altitude emissions)
– Few alternatives (improve efficiency, alternative fuels, high speed rail)
*
* ~4%,including contrails
Typical Aviation Emissions
Source: Climate Change and Tourism – Responding to Global Challenges (UNWTO Report)
Return flight
emissions
Economy class
NYC-Paris
1.4 tCO2/cap
NYC-Tokyo
2.6 tCO2/cap
NYC-Chicago
0.4 tCO2/cap
Source: ICAO
Improvements in Aircraft Fuel Efficiency
Research Objective
•
Empirical characterization of air travel demand/income relation
– Travel demand literature points to high elasticity between air travel and income
– About 2 for developing countries, 1.5 for developed ones
– Other factors such as fares are important but are dominated by income
– Essentially all studies assume constant elasticity, have not estimated an exact relation
•
Use income distribution data to test hypothesis against panel of different countries
– The idea of using income distribution in travel demand analysis is relatively new and has
never been applied to aviation
– Storchmann, Energy Econ. 2005: uses Gini in determining car ownership
– Chamon et al., Economic Policy 2008: estimates an income threshold function for car
ownership
•
Generate long term BAU aviation demand forecasts
Surveys
• Sparse survey data on flying (NHTS, MTC, UK CAA)
– Flying is largely undertaken by those in richer households and
that most of the growth in flying is coming from people in these
households flying more often (CAA, 2006)
– NHTS: Suggest a non linear relation with marked increase in
elasticity after certain threshold
– MTC: No neat evidence of saturation.
NHTS 2001 Survey
NHTS 2001 Survey
Price Sensitivity
Proportion of Airfare in Travel Expenses
Data
• Income Distribution
– WDI 2007, WB PovCalNet, UN WIDER World Income Inequality Database (WIID
2b)
– Coverage: income/consumption shares in quintiles or deciles
– Beta2 distribution fit, mean anchored at GDP/cap MER
• Air travel
– ICAO: 1990-2004 passengers carried, passenger-km, (both national and
international)
– UNData (World Bank): 1970-2005, passengers carried
– Wide coverage of both developed and developing countries
Methodology
• S-shaped relation between income and flying, no country effects
Weibull, (Logistic), (Gompertz)
•Income distribution translates it into flights per capita
•Maximum Likelihood estimation on the S-shaped curve
•
Scenarios to 2050
–Use EIU/GS/IPCC GDP/cap projections, UN population projections
–Assume unchanged income distribution or use scenarios with higher /lower
inequalities.
–For given distance/trip and efficiency improvements, calculate energy and
emissions
The Methodology in One Cartoon
Two groups of countries:
1. Early Fliers: USA, CAN, AUS, SWE, FIN, NOR, GBR, NLD,
FRA, DEU, ITA, ISR, AUT, DNK, JPN
2. Catch up: BRA, CHL, CHN, CZE, HUN, KOR, MYS, ESP,
GRC, PRT, IND, IDN, PHL, THA etc.
1
2
GDP per capita
Projected Growth in Passengers
Revenue Passenger Kilometer Projections
Boeing
Airbus
WBCSD
Revenue Passenger Kilometer Projections
Projections assuming 0%, 0.5% and 1.0% growth in average
distance traveled per flight.
Projected Emissions
Aviation could account for 5%-12% of CO2 emissions in 2050.
Impact of change in inequality
Aviation is a luxury good with a very nonlinear elasticity profile.
Inequality could affect both rate of growth and possible saturation.
Case study:
Rapid increase in inequality in China in the last 25 years.
China (1992) 0.35
China (2005) 0.46
Consider 3 scenarios:
1) Constant inequality,
2) Decreases to 1992 levels and
3) Increases to Brazil (2005) 0.56.
Constant Inequality Projection for China
Red: China (2005) 0.46
Blue: Brazil (2005) 0.56
Green: China (1992) 0.35
Conclusions
• Evidence of a sigmoid relationship between income and flying.
• Need to know more about saturation at high incomes
(surveys), though it doesn’t seem to a significant factor yet
• OECD aviation demand will grow and the OECD will continue
to lead in per capita flying.
• High growth from middle level developing countries like Brazil,
China etc.
EXTRA SLIDES
PRICE ELASTICITIES
Note: Short series (3 years) except for the US.
Source: InterVISTAS Study for IATA (2007)
INCOME ELASTICITES
Source: InterVISTAS Study for IATA (2007)