IATA Economics Briefing Impact of Recession Dec08

IATA ECONOMIC BRIEFING DECEMBER 2008 THE IMPACT OF RECESSION ON AIR TRAFFIC VOLUMES

Recession is now forecast for North America, Europe and Japan late this year and into 2009. The last major downturn in air traffic, driven by recession rather than terrorist attack, was in 1991 when global passenger traffic (RPKs) fell 2.6%. We now forecast that global passenger traffic will fall by 3% in 2009.

Weakness in travel markets has lasted three years in previous recessions. We do not expect a return to traffic growth above 4% until 2011. Economic forecasts imply that airline traffic will remain below the previous trend over the medium-term, with passenger travel forecast to be 9% lower by 2016 than pre-crisis industry forecasts.

Substantial downward revision of forecast air travel post-financial sector crisis

Global passenger kilometers flown

1000

2000

3000

4000

5000

6000

7000

1980 1984 1988 1992 1996 2000 2004 2008 2012 2016

Bill

ions

9% lower

Pre-crisis industry forecasts (ICAO FESG)

Revised IATA forecast

1980-82 1991-93 2001-03 2008-10

12% lower

Source: ICAO, IATA

Air freight volumes are also expected to shrink in both 2008 and 2009. However, a more rapid recovery in line with world trade is anticipated with a return to 6%+ growth by 2010. IATA’s new forecasts are shown below:

IATA’s revised forecasts for global air transport volumes

2007 2008 2009 2010 2011 2012 2013

Passenger departures, millions 2,260 2,305 2,236 2,256 2,373 2,535 2,698 % change 6.4% 2.0% -3.0% 0.9% 5.2% 6.8% 6.4% RPKs, billions 4,201 4,285 4,157 4,194 4,412 4,713 5,015 % change 6.7% 2.0% -3.0% 0.9% 5.2% 6.8% 6.4% Freight tonnes, millions 41.6 41.0 38.9 40.9 43.7 46.4 49.6 % change 4.5% -1.5% -5.0% 5.1% 6.9% 6.2% 6.7% FTKs, billions 158 156 148 156 167 177 188.7 % change 4.8% -1.5% -5.0% 5.1% 6.9% 6.2% 6.7%

Passenger departures, millions 828 854 829 836 885 953 1017 % change 8.8% 3.2% -3.0% 0.8% 5.8% 7.7% 6.7% RPKs, billions 2,545 2,626 2,548 2,570 2,719 2,929 3126 % change 7.6% 3.2% -3.0% 0.9% 5.8% 7.7% 6.7% Freight tonnes, millions 25.3 24.9 23.7 24.9 26.6 28.2 30.1 % change 5.9% -1.5% -5.0% 5.1% 6.9% 6.2% 6.7% FTKs, billions 132 130 124 130 139 148 158 % change 5.3% -1.5% -5.0% 5.1% 6.9% 6.2% 6.7%Source: ICAO data to 2007, IATA forecasts 2008-2013

Domestic and international scheduled traffic

International scheduled traffic only

IATA Economics: www.iata.org/economics

IATA Economic BriefingDecember 2008

Comparing IATA’s forecasts with its recent survey of airline industry forecasts 2007 2008 2009 2010 2011 2012

Passenger departures IATA forecast, millions 2,260 2,305 2,236 2,256 2,373 2,535 % change 6.4% 2.0% -3.0% 0.9% 5.2% 6.8% Industry survey, millions 2,260 2,323 2,386 2,493 2,621 2,749 % change 2.8% 2.7% 4.5% 5.1% 4.9%

Passenger departures IATA forecast, millions 828 854 829 836 885 953 % change 8.8% 3.2% -3.0% 0.8% 5.8% 7.7% Industry survey, millions 828 869 909 952 1000 1049 % change 4.9% 4.7% 4.7% 5.0% 4.9% Freight tonnes IATA forecast, millions 25.3 24.9 23.7 24.9 26.6 28.2 % change 5.9% -1.5% -5.0% 5.1% 6.9% 6.2% Industry survey, millions 25.3 26.4 27.6 28.8 30.2 31.6 % change 4.3% 4.5% 4.5% 4.7% 4.7%Source: Industry survey - IATA Passenger Forecasts 2008-2012, IATA Freight Forecasts 2008-2012http://www.iata.org/ps/intelligence_statistics/

Domestic and international scheduled traffic

International scheduled traffic only

Each year IATA surveys the airline industry to estimate the ‘industry consensus’ of traffic forecasts on the majority of country pairs. The results of the latest survey, undertaken in the summer, have just been published in two reports ‘IATA Passenger Forecast 2008-2012’ and ‘IATA Freight Forecast 2008-2012’. These reports can be accessed from www.iata.org/ps/intelligence_statistics/.

The table above compares our own forecasts with the ‘industry consensus’ identified by the survey. It is clear that expectations prevailing during the summer for the growth of passenger and freight markets will be disappointed both this year and next. By 2010 we forecasts that the level of global passenger departures will be 10% lower than expected by the industry before the recent financial crisis. International freight tonnes are likely to be 13% lower than expected by 2010.

Once recovery is solidly established by 2011 we forecast growth rates that are much closer to industry expectations. The two reports of the survey contain much valuable information about medium-term industry expectations for country-pair passenger and freight market developments. These medium-term expectations may not have been significantly changed by the short-term fluctuations due to the recession.

Global growth in passenger traffic and GDP

-4%

-2%

0%

2%

4%

6%

8%

10%

12%

14%

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

0

1

2

3

4

5

6

7

8

% c

hang

e ov

er y

ear

1980-82 1991-93 2001-03 2008-10

Global RPKs (left scale)

Global GDP (right scale)

Source: ICAO, EIU IATA

Global revenue passenger kilometers (RPKs) have fallen twice in the past 35 years, in 1991 due to recession and in 2001 due to the 9-11 terrorist attack on top of a sharp economic slowdown. There was also a sharp slowdown during the 1980-82 recession but not an absolute fall in travel.


http://www.iata.org/ps/intelligence_statistics/


It is clear from the chart above that air travel is driven primarily by economic growth and that it is highly leveraged to the economic cycle, expanding and contracting at roughly twice the rate of the overall economy. The latest economic forecasts show global GDP growth slowing to less than 1%, with the US, Japan and Europe in recession. This is consistent with global RPKs falling by 2.2% in 2009 and not rising above 4% until 2011.

Global passenger kilometers flown

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

-5 ye

ars

-4 ye

ars

-3 ye

ars

-2 ye

ars

-1 ye

ar

Cycle

peak

+1 ye

ar

+2 ye

ars

+3 ye

ars

+4 ye

ars

+5 ye

ars

+6 ye

ars

+7 ye

ars

+8 ye

ars

+9 ye

ars

Billio

n

1979 peak

1990 peak

2000 peak

2008 peak

3 years of low growth

9.7% pa 1970s trend

5.4% pa 1980s trend

4.8% pa 1990s trend

4.7% pa 2000s trend

Forecast

Source: ICAO, IATA

The duration of weakness in travel markets has typically been around three years, with negative, zero, or very low growth of global RPKs.

Moreover, the historical experience has been that travel volumes rarely return to the previous trend or peak-to-peak growth. The post-2000 recovery was the exception but that was driven by a credit boom that is unlikely to be repeated. Average growth from 2008 to the next peak is forecast to slow to an annual rate of 4.5%.

Global passenger kilometers flown relative to trend

-30%

-25%

-20%

-15%

-10%

-5%

0%

Cycle peak +1 year +2 years +3 years +4 years +5 years +6 years +7 years

% tr

end

pass

enge

r kilo

met

ers

1979 peak 1990 peak 2000 peak 2008 peak

These bars show the % gap between the lines in the previous chart

Source: ICAO, IATA

The last two downturns (1991-93 and 2001-03) saw global RPKs continuing to decline relative to the previously established trend for three years, before starting to narrow the gap. After the 1980 downturn the market never return to its previous trend.

This was partly a reflection of a structural fall in economic growth from 3.8% a year to around 3.3% a year, following the oil shocks of the 1970s. It was also a reflection of maturing travel markets in the developed economies.



Growth in global passenger kilometers flown

-4.0

-2.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

% c

hang

e

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

4.0

Rat

io o

f tra

ffic

to G

DP

grow

th

1980-82 1991-93 2001-032008-2010Ratio RPK to

GDP growth (right scale)

Global RPKs

Source: ICAO, EIU, IATA

One enduring industry ‘fact’ is that traffic grows twice as fast as GDP. As the chart above shows this is only true on average for the good years of the economic cycle: 1971-1979, 1984-1990, 1993-2000, 2004-2007.

It is not true during downturns when the ratio has turned negative, with traffic falling faster than GDP in 1991 and in 2001. On average, throughout the cycles of the past 35 years, global RPKs have risen by 1.6 times GDP.

This ratio is forecast to turn negative once more in 2009 as global RPKs shrink, before recovering toward 2 by 2011, producing strong growth in 2011-13.

Global GDP and airline real yields

0

1

2

3

4

5

6

1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

60

80

100

120

140

160

180

US

cent

s/A

TK in

con

stan

t pric

es

Global GDP growth (left scale)

Airline real yields (right scale)

1980-82 1991-93 2001-03 2008-10


Since air transport market liberalisation begain in the late-1970s real yields have fallen by almost 50%. Air travel

and air freight has become considerably cheaper and clearly this has boosted volumes. However, this impact has been a steady trend. Fluctuations in real yields from year to year have not seemed to be a major driver of travel. Economic growth, GDP, has been the principle driver of changes in global RPKs.



This observation from the charts above has been substantiated by econometric research (see Air Travel Demand, IATA Economics Briefing No 9) which found demand at the national or regional level to be relatively inelastic with respect to price, with elasticities in the range of 0.5-0.6, but very responsive to GDP, with elasticities of 1.5-2.

Global GDP

10

20

30

40

50

60

70

-5 ye

ars

-4 ye

ars

-3 ye

ars

-2 ye

ars

-1 ye

ar

Cycle

peak

+1 ye

ar

+2 ye

ars

+3 ye

ars

+4 yea

rs

+5 ye

ars

+6 ye

ars

+7 yea

rs

+8 ye

ars

+9 ye

ars

US$

trilli

on, 2

007

pric

es

1979 peak

1990 peak

2000 peak

2007 peak

3.8% pa 1970s trend

3.3% pa 1980s trend

3.2% pa 1990s trend

3.3% pa 2000s trend

Source: EIU, IATA

Gross domestic product (GDP) measures economic activity and its cyclical pattern is clearly responsible for the majority of the fluctuations in air travel volumes.

Peak-to-peak trends in economic growth have declined since the 1970s, as has air travel growth, and GDP has rarely returned to its previous trend after a recession shock, with the exception of the most recent cycle. It has already been noted that this was driven by a credit boom which is unlikely to be repeated.

Global GDP relative to trend

-7%

-6%

-5%

-4%

-3%

-2%

-1%

0%

1%

2%


% tr

end

GD

P


Source: EIU, IATA

If the current economic forecasts are right then this economic downturn is shaping up to be worse than 2001 and

closer to the business environment experienced during the 1991 recession, when GDP fall to 3% below trend before starting to recover four years after the previous peak.

Even seven years after the 1991 recession began GDP was still over 1% below its previous trend. This is the major reason why air travel markets rarely return to their previous trend levels following major shocks to demand.



Global airline passenger traffic and capacity

-5

-3

-1

1

3

5

7

9

11

13

15

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e

1980-82 1991-93 2001-03 2008-10

Traffic growth

Capacity growthTraffic - capacity

growth differential

Source: ICAO, IATA

Of course airlines will respond to demand shocks by trying to bring capacity into line. During downturns, as the chart above shows, demand usually falls faster than airlines are able to reduce capacity in the first year of recession. But then the industry usually reduces capacity in line with demand.

This cycle is a little different in that airlines, particularly in the US, started to slash capacity in response to earlier high fuel prices. Outside the US cuts have been less. Nonetheless, capacity is moving more in line with demand than in previous cycles, which should help limit the damage of the recession on airline profitability.

Global airlines unit costs and yields

-10%

-5%

0%

5%

10%

15%

20%

25%

1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

Unit costs

Yields

1980-82 1991-93 2001-03 2008-10

Source: ICAO, IATA

Many analysts in the US are suggesting that planned capacity cuts will tighten supply-demand conditions for US airlines in 2009 to the extent that they could raise yields and make profits, despite the recession.

However, the chart above shows that, since deregulation in the late-1970s yields follow unit costs almost exactly. There is a correlation coefficient of 0.99 between the two series. This is consistent with other evidence that the airline industry is highly competitive and unable to sustain fares significantly above unit costs.

Yields rose sharply last year and in 2008 because of fuel prices. During 2009 unit costs will fall sharply because of the fall of oil and jet fuel prices. Yields are also likely to decline as a result.



Global airline yields and load factors

-10%

-5%

0%

5%

10%

15%

20%

25%

1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

50

55

60

65

70

75

80

% A

SKs

Passenger load factor (right scale)

Yield growth (left scale)

1980-82 1991-93 2001-03 2008-10

Source: ICAO, IATA

Load factors are likely to hold up much better in this downturn than in the past, because of the pre-emptive cuts in capacity. Load factors are likely to rise for US airlines on domestic markets. However, the chart above confirms the earlier assessment that there is little evidence that rising load factors necessarily lead to better yields.

Yields have typically fallen during recessions but this seems to have been driven more by falling costs than falling load factors; competition keeps fares and freight rates close to costs at all times during the cycle.

Average distance flown and GDP growth

-2%

-1%

0%

1%

2%

3%

4%

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

0

1

2

3

4

5

6

7

8%

cha

nge

over

yea

r

1980-82 1991-93 2001-03 2008-10

Growth in average distance flown (left scale)

Global GDP growth (right scale)


The average distance flown has risen in the past 35 years as long-haul city-pairs and frequencies have increased at

a faster rate than domestic services. However, there is a cyclical pattern to distance flown, presumably as passengers switch from more expensive long-haul trips to short-haul.

In our forecasts we have assumed that average distances stop growing during the downturn.



Global scheduled passenger departures

500

1000

1500

2000

2500

3000

1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011

Milli

on

1980-82 1991-93 2001-03 2008-10

IATA survey of airline forecasts


8% lower

Source: ICAO, IATA

As a result of average distances stabilising we forecast similar growth of passenger departure numbers as for global RPKs.

IATA has just published its latest annual survey of airline traffic forecasts (IATA Passenger Forecast 2008-12) which surveyed airlines in the middle of the year. As the chart above shows the recession shock will disappoint expectations and planning contingent on them, with passenger numbers forecast to be 8% lower than expected by 2012.

Global airline passenger departures and GDP growth

-4

-2

0

2

4

6

8

10

12

14

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

0

1

2

3

4

5

6

7

8%

cha

nge

over

yea

r

1980-82 1991-93 2001-03 2008-10


Global passenger departures (left scale)


It is little surprise that growth in global passenger departures mirrors the pattern for global RPKs, and its

relationship with GDP growth. Following the sharp decline in passenger numbers already seen in the second half of 2008, a fall of 3% is forecast

for 2009.



Global air passenger departures

0

500

1000

1500

2000

2500

3000

3500

-5 ye

ars

-4 ye

ars

-3 ye

ars

-2 ye

ars

-1 ye

ar

Cycle

peak

+1 ye

ar

+2 ye

ars

+3 ye

ars

+4 ye

ars

+5 ye

ars

+6 ye

ars

+7 ye

ars

+8 ye

ars

+9 ye

ars

Milli

on

1979 peak

1990 peak

2000 peak

2008 peak

3 years of little growth

4.4% pa 2000s trend

3.7% pa 1990s trend 4% pa

1980s trend

7.8% pa 1970s trend

Source: ICAO, IATA

Until recently the trend in passenger numbers was considerably slower than growth in RPKs, because of the growth in long-haul and its impact on average distances flown. In the 1970s passenger numbers grew by 2% points less than RPKs, in the 1980s by 1.4% points, in the 1990s by 1% point. In recent years that difference slowed to just 0.3% points and has now almost halted.

Global passenger departures relative to trend

-30%

-25%

-20%

-15%

-10%

-5%

0%

5%

10%


% o

f tre

nd p

asse

nger

dep

artu

res


Source: ICAO, IATA

Again the pattern of passenger numbers, relative to trend following a demand shock, is similar to both global RPKs

and GDP. The deterioration in passenge markets relative to the previously established trend continues for three years before any improvement is felt.



International passenger kilometers flown and global GDP

-5

0

5

10

15

20

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013

% c

hang

e ov

er y

ear

0

1

2

3

4

5

6

7

% c

hang

e ov

er y

ear

1980-82 1991-93 2001-03 2008-10

International RPKs (left scale)



The rapid growth in long-haul in the past 35 years has meant that international RPKs have risen at a faster rate

than domestic and therefore overall global RPKs. International travel, with a higher GDP elasticity than short-haul, is also more highly geared to the economic cycle. It rises faster in an upturn and falls further in downturns.

In 2009 international RPKs are forecast to shrink by 3% and grow only sluggishly in 2010, before a strong recovery in growth above 7% by 2012.

International passenger kilometers flown

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

-5 ye

ars

-4 ye

ars

-3 ye

ars

-2 ye

ars

-1 ye

ar

Cycle

peak

+1 ye

ar

+2 ye

ars

+3 ye

ars

+4 yea

rs

+5 yea

rs

+6 ye

ars

+7 ye

ars

+8 yea

rs

+9 yea

rs

Bill

ion

1981 peak

1990 peak

2000 peak

2008 peak

4.9% pa 2000s trend

7.2% pa 1990s trend

6.8% pa 1980s trend

10.7% pa 1970s trend

Forecast

Source: ICAO, IATA

Throughout the 1970s, 1980s and 1990s international RPKs have grown at a trend between 1-2.4% points faster than overall global RPKs. However, the trend since the peak in 2000 has been barely 0.2% points faster. This is likely to reflect the impact of liberalising intra-EU travel markets, with entry in domestic EU markets from fast growing LCCs, and domestic markets in India and China which also have expanded very rapidly.

Although domestic market growth has accelerated due to liberalisation in recent years, there has clearly been a sharp slowdown of international RPK growth during the 2000s. This seems to have been largely due to growth on shorter-haul international markets. Growth in international passenger departures has been much more stable with trend growth of 6.8% during the 1990s and 6.2% during the 2000s.



International freight tonnes flown

0

5

10

15

20

25

30

35

1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011

Mill

ion

tonn

es

IATA survey of industry forecasts


1980-82 1991-93 2001-03 2008-10

Source: ICAO, IATA

Air freight volumes have declined much earlier and to a greater degree than passenger markets. A fall of 1.5% is forecast for international freight tonnes in 2008, and a further decline of 5% is forecast for 2009.

Compared to the growth in air freight expected by the industry, when surveyed in the middle of the year, air freight volumes are forecast to be 10% lower than expected by 2012.

Global air freight tonne kilometers flown and world trade

-10

-5

0

5

10

15

20

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% c

hang

e ov

er y

ear

World trade in manufactures

Air freight tonne km flown

1980-82 1991-93 2001-03 2008-10


Global air freight volumes have typically grown quite closely in line with world trade volumes of manufactured

goods, for obvious reasons. Air freight is slightly more volatile than overall world trade, because it is the most costly transport mode and so is substituted away from during downturns.

However, in the past few years the growth of air freight had been rather slower than expected given the expansion of world trade. World trade is expected to decline in 2009 producing a 5% decline in global FTKs.



Air freight tonne kilometers share in world trade volumes

0.30

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010

% w

orld

trad

e vo

lum

e

1980-82 1991-93 2001-03 2008-10


The slower growth of FTKs experienced during the 2000s is consistent with a change in the trend of market share.

The chart above shows there was apparently a sharp change in trend after air freight’s share of world trade volumes peaked at 0.66% in 1997 (not the share of world trade value in current US$, which is around 35%). This was before the rise in fuel costs and so may reflect a shift in technology, such as the reported improvement in the time performance of the container shipping industry. Whatever the underlying cause of this shift we have assumed this down-trend in market share will continue in our forecasts.

Growth in average distance flown by air freight

-6%

-4%

-2%

0%

2%

4%

6%

8%

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007

% c

hang

e ov

er y

ear

1980-82 1991-93 2001-03

Source: ICAO, IATA

Air freight does not behave in the same way as passengers during recessions. There is no indication of any

cyclicality in average distances flown by air freight. In fact in the past 15 years there has been no trend increase in average distance flown. We have assumed this to continue in our forecasts.



Global air freight tonnes

0

10

20

30

40

50

60

70

-5 ye

ars

-4 ye

ars

-3 ye

ars

-2 ye

ars

-1 ye

ar

Cycle

peak

+1 ye

ar

+2 ye

ars

+3 ye

ars

+4 ye

ars

+5 ye

ars

+6 ye

ars

+7 ye

ars

+8 ye

ars

+9 ye

ars

mill

ion

tonn

es

1980 peak

1990 peak

2000 peak

2007 peak

6.2% pa 1970s trend

5.2% pa 1980s trend

5.1% pa 1990s trend

4.6% pa 2000s trend

Source: ICAO, IATA

Growth in air freight tonnes have slowed since the 1970s but at a much reduced rate than passenger travel. In the 1970s growth in the tonnes of freight shipped by air averaged 6.2%. By the 2000s that trend had slowed to 4.6%.

There also seems to be less of a sustained divergence from trend with air freight volumes. There is less evidence that air freight markets are ‘maturing’ and therefore slowing in the same way as passenger markets.

Global freight tonnes relative to trend

-18%

-16%

-14%

-12%

-10%

-8%

-6%

-4%

-2%

0%

2%


% o

f tre

nd fr

eigh

t ton

nes


Source: ICAO, IATA

Air freight volumes also typically recover from a demand shock a little more quickly than passenger markets. There

is no hard and fast rule but the most recent cycle saw a recovery after two years rather than the three years common to passenger markets.

IATA Economics 1st December 2008


Documents

IATA Economics Briefing Impact of Recession Dec08