A comparative analysis of trends in environmental externalities of road transport (1990-2010) in Belgium and the UK.

Abstract

Against a background of growing transport demand and congestion, EU transport policies have focussed on tighter emission standards for new vehicles. This paper evaluates the results of these policies by looking at the total air pollution externalities of road transport for the UK and Belgium. These results were obtained within recent the European ExternE project (1998-2000).
The national assessments are based on an aggregation of detailed bottom-up studies for a wide set of representative trajectories, traffic situations and technologies. These allow national level studies to take account of the location of emissions, specifically accounting for the higher external costs from vehicles in urban areas.
Calculations of costs per tonne of pollutant are based on the latest, thoroughly revised, ExternE-2000 accounting framework. Due to the dominance of local impacts for some pollutants (particles), fleet and mobility statistics were collected with a high level of spatial disaggregation. Trends have been assessed since 1990 and have been extrapolated to 2010 assuming transportation growth and fleet turnover.
For both countries, the current air pollution externalities of the use phase of road transport are estimated to be about 1% of national GDP. Of this, most arises from the passenger car fleet. The analysis of externalities over time (1990-2000) reveals how successful the introduction of tighter emissions standards across Europe has been, and importantly shows that country specific policies can lead to significant differences in the levels of national externalities. Between 1993 and 1998 externalities of the passenger car fleet have decreased by more than 1/3rd in the UK and by almost a quarter in Belgium. This decrease was driven by successive introductions of lower European emissions standards but counteracted by a growing fleet and increasing mileage.
The use of diesel fuel dominates external costs in both countries, as it leads to higher impacts (from particles) in cities and vehicles have higher average mileages. In both countries, urban emissions dominate overall road sector externalities (up to 70%). Motorway traffic accounts for 14% in the UK and up to 24% in Belgium.
For both countries, environmental external costs have been compared against current fuel duty levels. Low levels of excises on diesel in Belgium have promoted a shift from petrol to diesel fuelled cars which have higher impacts per vehicle km. This is the main reason for the lower reductions over time.
The analysis of future road sector externalities (to 2010) shows reductions will continue, as newer vehicles are introduced and older vehicles are removed from the fleet, even though vehicle km are estimated to rise in both countries. However, there are differences in the relative effects of different impact categories. The potential impacts of climate change become increasingly important in the overall road sector externalities - although they cannot be fully quantified.

Full text

Int Panis, L., Watkiss, P., De Nocker, L., Torfs, R., 2000. A comparative analysis of trends in environmental
externalities of road transport (1990- 2010) in Belgium and the UK. Proceedings of the TERA2K conference.
Milan, Italy. Scuola Enrico Mattei, ENI (S. Donato Milanes, Italy), Oct. 26-27, 2000.
A comparative analysis of trends in environmental externalities of
road transport (1990-2010) in Belgium and the UK.
Luc Int Panis1, Paul Watkiss2, Leo De Nocker1, Rudi Torfs1
1. Vito, Centre of Expertise IMS, Boeretang 200, B-2400 Mol,Belgium. E-mail: luc.intpanis@vito.be
2. AEAT Environment,Culham, OXON, UK-OX14 3ED, UK. E-mail: paul.watkiss@aeat.co.uk
Abstract
Against a background of growing transport demand and congestion, EU transport policies have focussed on
tighter emission standards for new vehicles. This paper evaluates the results of these policies by looking at the
total air pollution externalities of road transport for the UK and Belgium. These results were obtained within
recent the European ExternE project (1998-2000).
The national assessments are based on an aggregation of detailed bottom-up studies for a wide set of
representative trajectories, traffic situations and technologies. These allow national level studies to take account
of the location of emissions, specifically accounting for the higher external costs from vehicles in urban areas.
Calculations of costs per tonne of pollutant are based on the latest, thoroughly revised, ExternE-2000 accounting
framework. Due to the dominance of local impacts for some pollutants (particles), fleet and mobility statistics
were collected with a high level of spatial disaggregation. Trends have been assessed since 1990 and have been
extrapolated to 2010 assuming transportation growth and fleet turnover.
For both countries, the current air pollution externalities of the use phase of road transport are estimated to be
about 1% of national GDP. Of this, most arises from the passenger car fleet. The analysis of externalities over
time (1990-2000) reveals how successful the introduction of tighter emissions standards across Europe has been,
and importantly shows that country specific policies can lead to significant differences in the levels of national
externalities. Between 1993 and 1998 externalities of the passenger car fleet have decreased by more than 1/3rd
in the UK and by almost a quarter in Belgium. This decrease was driven by successive introductions of lower
European emissions standards but counteracted by a growing fleet and increasing mileage.
The use of diesel fuel dominates external costs in both countries, as it leads to higher impacts (from particles) in
cities and vehicles have higher average mileages. In both countries, urban emissions dominate overall road
sector externalities (up to 70%). Motorway traffic accounts for 14% in the UK and up to 24% in Belgium.
For both countries, environmental external costs have been compared against current fuel duty levels. Low levels
of excises on diesel in Belgium have promoted a shift from petrol to diesel fuelled cars which have higher
impacts per vehicle km. This is the main reason for the lower reductions over time.
The analysis of future road sector externalities (to 2010) shows reductions will continue, as newer vehicles are
introduced and older vehicles are removed from the fleet, even though vehicle km are estimated to rise in both
countries. However, there are differences in the relative effects of different impact categories. The potential
impacts of climate change become increasingly important in the overall road sector externalities - although they
cannot be fully quantified.
1 Introduction
Most papers on transport externalities give estimates of the marginal external costs (the costs caused by an
additional vehicle) for a specific country. This paper describes the issues and methodology associated with
aggregating this externality data to levels required by policy instruments and applies the approach to two
European countries, the United Kingdom and Belgium. The aggregation analysis has been undertaken to allow
analysis of the effects of successive emission standards (Euro types) and other factors in determining road sector
externalities.

2 Methodology
The methodology for analysis in this paper uses the updated ExternE methodology (CEC, [1]). Details of the
project and the methodology are detailed elsewhere in these proceedings. In summary the approach uses a
bottom-up or impact pathway approach to quantify the marginal external costs of transport. Impacts categories
include public health (morbidity and mortality), agriculture and materials.
2.1 Aggregation issues
Previous work under the ExternE project has developed and demonstrated aggregation of sector emissions, with
application to the electricity generation sector (CEC, [2]). These studies provide an important starting point for
transport sector aggregation. However, there are theoretical and practical issues that are different for the
transport sector and which necessitate a different approach.
The first issue is with respect to speed. Transport emissions vary significantly with speed and this variation is
non-linear. Moreover, for road transport, the speeds on particular road types (highway, main road, etc.) also vary
with the numbers of other vehicles on the road. As a result, even emission aggregation becomes an issue for
transport.
The second issue centres on the assessment of mobile emission sources. The emissions from vehicles vary
over time as well as by location. As ExternE uses linear dose-response functions, the issue of time dependency
is less important (all emissions from any one location for any time period can effectively be assessed as an
annual emission). In contrast, the issue of location is of major concern. Previous aggregation studies have
shown that the geographical transferability of global impacts and regional impacts (within one country) are
acceptable but the transferability of local impacts is poor. For power stations this is not a major issue; emissions
from the electricity generation sector are generally released from high stacks, away from major urban areas, and
local impacts are a small part of overall damages. In contrast, local impacts are often dominant for transport,
because emissions are released at ground level, often in areas of high population density. The implications of this
are clear. A much greater level of geographical resolution is required for the local level assessment for transport
aggregation. At the very least, an aggregation analysis should differentiate between urban and non-urban areas.
Moreover, the greater the level of data resolution within urban areas, the better the accuracy of the aggregation
results. Obviously as the size of the area under investigation increases (as in the national studies presented here),
the level of detail with respect to speed, traffic flow, meteorological conditions, etc. must be reduced to keep the
assessment manageable.
Work within ExternE has shown that there are small differences between total and marginal impacts.
Potentially greater differences exist between total and marginal costs, especially when considering large sectors
such as transport, though to date these have not been assessed with ExternE in detail. For the national
aggregation studies here, we have therefore assumed linearity.
2.2 Aggregation methodology
When aggregating at a national level, there are two main approaches that can be taken. The first aggregates
emissions and then applies area specific costs per tonne, whilst the second undertakes marginal cost assessment
in detail and then combines these with national vehicle km data. For the studies here, the UK analysis has used
the former methodology whilst the Belgian the latter.
For the UK, data were taken on the national vehicle fleet, specifically the number of vehicles, the fuel split
(gasoline and diesel) and the age distribution/technology standard (pre-Euro, Euro I, Euro II, Euro III), based on
year of first registration. Data were also derived for trip km and speed characteristics. The level of data with
respect to trip type varies considerable across Europe. The data in the UK are good, with information on the
kilometres traveled on different roads (e.g. motorway/highway, main road, minor road) in different areas (e.g. on
each road type in rural areas, in towns, in cities). This information allows emission factors to adjust for speed
characteristics for example using characteristic average speeds on different road types, and for location using
different unit costs for each area type.
Data are also available on journey lengths and this was used to add in cold start emissions. Similarly, data on
mileage distribution for different vehicles was taken to take into account that newer vehicles (especially diesels)
tend to have a higher annual mileage. Once the data was assembled, the unit pollution values for different
locations were applied to estimate damage costs. For this analysis, data were split into urban, motorway and
rural kms. The estimates of costs per tonne were derived from marginal cost runs using the EcoSense transport
model. The method for deriving UK data is summarized in the figure below.

Emission Factors
- hot exhaust (g/km)
by vehicle type
by Euro std
with speed
- cold start
Fleet Composition
- age distribution
- Euro standard
- Fuel split (% diesel)
Traffic Activity
- vehicle number
- speed
- number trips
- location of trips
Emissions
(tonnes/year)
by location
Unit pollution costs
- by location
(values in Euro per
tonne, derived from
marginal costs runs
in each location)
Aggregated Costs
Sum of costs
by location
Figure 1 : Methodology for Road Transport Aggregation by Emissions (UK)
For the Belgian aggregation, marginal costs per km for all vehicle types for different road or location types
were derived. Marginal costs runs (also using the EcoSense transport model) were undertaken for each
combination of emission standard (pre-Euro, Euro I, Euro II, Euro III), fuel (diesel, gasoline, LPG) for six drive
types. The six drive types (urban peak, urban non-peak, highway peak, highway non-peak, rural peak, and rural
non-peak) each assume representative average speeds. Effects from cold starts on emissions and vehicle age on
mileage were taken into account for passenger cars only. The analysis then uses these costs per km with the
national data on fleet distribution and national vehicle km to estimate total costs.
The question arises whether aggregation by emissions (UK) or by km (Belgium) is the better methodology to
use? Both are valid and for the calculations presented in this paper, there is little to choose between them,
though where more dis-aggregated data exists, the emission based approach is quicker to undertake.
Finally, cross comparison with other countries in Europe has found data availability varies enormously and is
likely to be the biggest hindrance to European wide studies at present. Many countries do not collate data dis-
aggregated by road type (and therefore speed) or location. This issue may be the largest barrier to the
application of such an approach for different countries.
3 Results
3.1 Estimates of air pollution externalities at the national level
For both countries, the analysis quantified the national air pollution externalities from the road sector. In both
cases, externalities were found to be very significant. Annual transport sector externalities in 1997-8 from
tailpipe emissions were estimated at 8.7 billion Euro in the UK and of 2.4 billion Euro for Belgium. In both
countries, this is around 1% of GDP. The results do not include upstream emissions from transport (evaporative
emissions, the construction of vehicles, fuel production and infrastructure), noise externalities and ecological
damage. Total external costs from the sector may therefore have been underestimated. The breakdown of these
total externalities, and the comparison with vehicle km, are shown in Table 1 below.
Table 1: Breakdown of mileage and external costs by vehicle type
% vehicle km
% damage
UK
B
UK
B
Passenger cars
90.9
91.7
49.5
66.3
Heavy Duty (Trucks)
7.9
7.6*
39.9
26.6*
Bus
1.2
0.7
10.6
7.2
* excluding foreign trucks

Although cars have a very high share of the vehicle km driven (>90%) they only account for around 50% -
66% of the damage. The bulk of remaining externalities is attributable to heavy-duty vehicles. Although these
vehicles only represent a small proportion of the national vehicle km, they have a dis-proportionately high
damage per km traveled and are therefore extremely important at a national level. The analysis of externalities
by location also provides some key conclusions. Table 2 below shows the breakdown of road sector externalities
by road/area type. As the table shows, total road sector externalities are dominated by urban emissions for both
countries.
Table 2: Breakdown of external costs by location (1997)
% of Externalities Belgium
% of Externalities UK
Urban
62
65
Rural
14
21
M-way
24
14
In both countries, an evaluation has been made of how environmental costs compare against vehicle and fuel
taxation. The marginal cost assessments that provide the information for the aggregation analysis show that in
general fuel duty exceeds environmental costs for gasoline vehicles and for diesel in rural areas. However, for
diesel vehicles in urban areas, environmental costs are greater than fuel duty levels. The effect is more
pronounced for Belgium, which has lower duty levels on diesel (in the UK prices of the two fuels are broadly
equivalent).
3.2 Comparison of passenger car externalities
The total amount of air pollution externalities from passenger cars in the UK and Belgium is shown in Figure 2.
In 1997 the impact was about twice as high in the UK as compared to Belgium. The average cost per kilometer
must therefore be much higher in Belgium given that the UK’s fleet of passenger cars is approximately 5 times
as large.
Although there are a number of differences in the approach used (including whether catalyst performance is
included and assumptions on cold start and mileage distribution), these do not hide the fact that externalities in
the Belgian fleet are much greater than for the UK. There are two possible explanations for this. Firstly,
exposure in Belgium could be higher because of higher population density and secondly Belgian cars could have
higher emissions of dominant pollutants. In examining trends between the countries it appears both of these are
important factors. PM-emissions are relatively high in Belgium because of the high proportion of diesel cars,
though it is hard to disentangle this from the greater exposure. Local impacts are higher in Belgium for all
pollutants in rural areas, but generally lower in urban areas because of the modest size of Belgian cities.
Regional impacts are always higher in Belgium because of its high rural population density and proximity to
other highly populated regions.
The analysis of the total externalities over time does, however, show similar trends between the two
countries. Total externalities have declined markedly over the past five years, mostly as a result of the
introduction of successive European emission standards. Interestingly, the relative decline in externalities does
differ between the countries – over the period 1993-1997, air pollution externalities declined by 14% in Belgium
and 21% in the UK. Int Panis et al. [3] have suggested 4 reasons for the slow reduction in Belgium.
1. Fleet growth
2. Increasing average mileage per car
3. The large and increasing proportion of diesel cars in the fleet
4. Slow fleet turnover

Passenger cars
Total environmental externalities
0
1
2
3
4
5
6
1989 1991 1993 1995 1997 1999
Euro (billions)
United Kingdom
Belgium
Figure 2: Total environmental externalities of
passenger cars
Passenger car fleet
relative mileage growth
95
100
105
110
115
1990 1991 1992 1993 1994 1995 1996 1997 1998
Total mileage
1993=100%
Belgium
United Kingdom
Figure 3: Mileage increase of the entire fleet of
passenger cars is very similar in the
UK and Belgium
Although the first two factors clearly reduce the effect of lower emission standards, they do not explain the
difference between both countries. Fleet mileage trends in both countries are very similar (Figure 3). Instead the
difference occurs because of the higher proportion of diesel vehicles in Belgium, and the fact that most UK
diesel vehicles are newer (due to greater growth over recent years). Most UK-diesels therefore comply with
Euro1 and Euro2 emission standards, whilst diesel vehicles in Belgium make up 40% of the fleet, account for
more than 50% of the mileage and are responsible for around 85% of the externalities from the sector because
the greater proportion of the diesel fleet is uncontrolled.
There is, however, one other factor that is important – with respect to fleet turnover. Despite the fact that
new car sales were very similar between 1996 and 1999 in the two countries, penetration of new Euro standards
was achieved slightly faster in the UK and, perhaps more importantly, resulted in more uncontrolled cars being
scrapped (Figure 4).
The results demonstrate that European and national level policies can influence national level externalities.
The studies show the reductions in national air pollution externalities from the road sector that have occurred
because of progressive Euro emissions standards, even against increasing fleet growth and mileage in both
countries. They also demonstrate that policies with respect to fuels or vehicles can have a major influence - in
Belgium, the greater take-up of diesel cars over time (which historically were perceived to be better for the
environment because of fuel efficiency) has led to large externalities.
0%
20%
40%
60%
80%
100%
United Kingdom
Belgium 1996
United Kingdom
Belgium 1997
United Kingdom
Belgium 1998
Euro II
Euro I
Euro0
Figure 4: relative shares of Euro types in the fleet of petrol cars (1996-1998)

3.3 Comparison of heavy duty vehicles for goods transport
As shown in Figure 5, the difference between the UK and Belgium is even more pronounced for trucks.
Externalities have fallen consistently since 1990 in the UK and fallen by 35% between 1993 and 1998. In
contrast, externalities have risen in Belgium. A small part of this difference can be attributed to the way that
traffic growth is distributed across urban, rural and motor-way locations, but the fact remains that externalities in
Belgium have remained relatively constant at best. When looked at within Belgium, the obvious conclusion is
that this arises because of the very large growth in total HDV kms (Int Panis [3]). However, as Figure 6 shows,
total mileage was very similar in both countries and UK-mileage growth was only marginally lower than in
Belgium. The answer to the differences must therefore lie in differences in the fleet composition and evolution.
Comparison of total HDV externalities
60%
70%
80%
90%
100%
110%
120%
1990 1991 1992 1993 1994 1995 1996 1997 1998
External costs
1993=100%
Belgium
United
Kingdom
Figure 5: relative comparison of HDV
externalities in the UK and Belgium
Comparison of HDV mileage growth
95%
100%
105%
110%
115%
120%
1990 1991 1992 1993 1994 1995 1996 1997 1998
Fleet mileage
1993=100%
Belgium
United Kingdom
Figure 6: relative comparison of vehicle.km
evolution for HD vehicles
Truck fleet
comparison of Articulates (UK) and 16-40 tonne (B)
by Euro standard
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
United Kingdom
Belgium 1996
United Kingdom
Belgium 1997
United Kingdom
Belgium 1998
United Kingdom
Belgium 1999
United Kingdom
Belgium 2000
Euro II
Euro I
Euro0
Figure 7: Introduction of new (Euro1 & Euro2) heavy trucks in the UK fleet as compared to Belgium
(1996-2000)

In both countries there has been a move to heavier goods vehicles in recent years due to major changes in the
haulage industry. This has led to differences in the environmental performance between different weight classes
of goods vehicles. Most recent sales of goods vehicles have been towards larger articulated vehicles. A greater
proportion of these heavier vehicles therefore complies with Euro 1 or Euro 2 emissions standards. However,
the fleet turnover and the degree of switch to heavier vehicles has been greater in the UK. This can be seen in
Figure 7 and Figure 8. Newer types of trucks are introduced much faster in the UK. By extrapolation it is found
that over 50% of all UK articulate trucks will comply with Euro2 emission standards in 2000. Moreover, older,
uncontrolled, trucks (Euro0) are scrapped much faster in the UK and as these vehicles account for the bulk of
externalities, it is their fraction that is most important for the decrease. These differences result in the
differences in total externalities over time. Again, emissions legislation has been important in driving the
reductions in externalities in the sector in the UK, though this has been driven by the economics of operating
heavier vehicles in this country, rather than leading to externalities reductions per se, as the Belgian fleet shows.
Truck fleet
comparison of Rigids (UK) and 3,5-16 tonne (B)
by Euro standard
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
United Kingdom
Belgium 1996
United Kingdom
Belgium 1997
United Kingdom
Belgium 1998
United Kingdom
Belgium 1999
United Kingdom
Belgium 2000
Euro II
Euro I
Euro0
Figure 8 Introduction of new (Euro1 & Euro2) light trucks in the UK fleet as compared to Belgium
4 Extrapolations up to 2010
4.1 Methodology and uncertainty
Although there are issues in forecasting future trends in road sector externalities, not least data availability, such
an exercise can provide valuable information for policy makers. For example, will future increases in vehicle
km and fleets offset the benefits achieved in externalities reductions seen over recent years? Will future
emissions legislation be as effective as historical initiatives? Projections on the future road sector externalities
for the UK and Belgium have been undertaken. The data for Belgium has been taken from a new model
developed by VITO (TEMAT). In the UK, AEAT have used estimates compiled under the National Atmospheric
Emissions Inventory.
In preparing the estimates of future road sector externalities, we have taken account of proposed legislation,
including new emission standards (Euro3 in 2000 and Euro4 in 2005) and planned reductions of the sulphur
content of petrol and diesel fuel (50ppb in 2005). These have been combined with predictions on future fleet
growth and km to obtain the estimates for passenger cars shown in Figure 9 and Figure 10.

The results show that as stricter emission control technologies are introduced, there is a continued decrease in
the associated external costs. However, analysis of the data has shown that this is more a consequence of the
scrapping of old cars rather than of the introduction of sophisticated Euro3 and Euro4 vehicles – the benefits of
these newer vehicles is much less than for previous standards (Euro 1 and 2). In consequence, it may be that
greater reductions in externalities, or at least much quicker improvements, could be achieved through higher fleet
turnover (i.e. targeting the removal older vehicles) than through emissions legislation.
As part of the studies, analysis of the key sensitivities has been undertaken. The key driver for the Belgian
fleet is the proportion of diesel vehicles. It seems very unlikely that the share of diesel cars will remain constant
at the 1998 level (38%). Indeed, if recent trends continue Belgium could end up with a fleet in which more than
1 car in 2 is a diesel. To study the relative importance of a whole range of uncertain factors, Vito has started a
Monte Carlo analysis. The importance of the results to this assumption is shown in Figure 9. Such an analysis is
important to identify those parameters that are most strongly correlated with trends and should therefore become
a focus of national transport policy.
Despite increasing levels of km in the fleets of both countries, there are still reductions. Over the time frame
considered, externalities are expected to decrease faster in Belgium then in the UK (71 vs. 42%). Within both
countries, the rate of decrease seems to slow down from 2005 onwards. However, in both cases the proportion
of diesel vehicles in the fleet is the key in determining these reductions. Any duty differentials or policies to
encourage continued growth in diesel car use are likely to offset potential reductions.
In contrast with findings for passenger cars, there is a major difference between the UK and Belgium for
heavy-duty vehicles for goods transport (Figure 11). While expected costs fall to about one third by 2010 in the
UK, there is only a limited reduction expected under the Belgian scenario (-16%). At first glance the picture
looks similar to the results in Figure 5, however, there is another reason. During the 1990-ties, the growth
(expressed as vkm) was very similar in both countries (Figure 6) and the discrepancy was entirely explained as a
result of dynamic changes in the composition of the fleet. Forecast mileages for the haulage industry in both
countries however are very different. Semi-official Belgian scenarios assume a 3.4% annual growth which is
achieved by an increasing average mileage per vehicle (2.3% up to 2005 and 1.8% afterwards) complemented
with additional (new) vehicles to meet the rest of the demand. Total fleet mileage is therefore estimated at 140%
in 2010. Actually this is a pure extrapolation of the fast growth that was observed in the late 1990-ties.
The assumption for the growth rate of the UK haulage industry is much lower (~1.5% annually) than the
observed growth of the late 1990-ties and the Belgium assumptions. The 1.5 % annual mileage growth
corresponds approximately to the growth that the TEMAT scenario uses for the Belgian fleet growth (1.4%).
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Euro (billions)
Total BAU
Euro0
Euro1
Euro2
Euro3
Euro4
Uncertainty
Figure 9: Extrapolated evolution of air pollution externalities
(Belgian passenger cars, TEMAT BAU-scenario)

0
0.5
1
1.5
2
2.5
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Billion Euro
United Kingdom
Belgium BAU
Belgium BAU'
Figure 10: Air pollution externalities from passenger cars. BAU is the ‘official’ Belgian
business as usual scenario (fixed diesel/petrol ratio in the fleet). BAU’ is a modified
version in line with the UK assumption (fixed diesel/petrol ratio in new cars).
0
0.5
1
1.5
2
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Billion Euro
UK-lorries
B-lorries
B'-lorries
Figure 11: External costs of air pollution from traffic within the UK and Belgium with HDV.
(Belgian data do not include trucks with foreign license plates)
When we bring both scenarios in line (B’ – lorries in Figure 11) there is still a large difference to be
explained. It is likely that this can be explained by the faster turnover of vehicles in the UK fleet of trucks. The
Belgian fleet is expected to hold 19% of uncontrolled trucks in 2010 that account for 38% of the costs. This may
prove to be a questionable assumption, especially because no downscaling of the mileage for age was undertaken
in HDVs. Phasing out uncontrolled trucks in Belgium would not eliminate the difference completely. It therefore
remains unclear if Figure 11 reflects any true differences between countries. Possible differences with respect to
the types of lorries (especially capacity) cannot be adequately studied. The criteria along which the fleet is
subdivided are different in both countries. In addition the current version of the Belgian TEMAT model cannot
model shifts within subcategories. The UK model assumes that the shift from smaller (rigid) towards larger
(articulate) trucks will continue.

4.2 Breakdown by category
There is one final area that warrants analysis with the future predictions – the relative importance of different
impact categories. The analysis by impact category shows that the reductions in road sector externalities occur
as a result of reductions in regulated pollutants (i.e. NOx, CO, SO2, PM10, VOCs) because of stricter Euro
emission standards and fleet turnover. However, successive measures have made little difference in controlling
greenhouse gas emissions. As a result, the potential costs of climate change have increased as a proportion of
overall air quality externalities, due to traffic growth. Moreover analysis of future externalities shows that this
trend is likely to continue. The figures below show expected costs from climate change for Belgian HDVs and
UK passenger cars.
These costs are estimated using the cost values per tonne from ExternE (2000), which are low compared to
many (older) estimates in the literature. However, all estimates of global warming impacts are very uncertain and
many do not include major potential impacts, especially on ecosystems. The use of the higher value from
ExternE (2000) of 16.4 Euro/tonne CO2 (a value closer to many estimates of mitigation costs), shows that
climate change becomes increasingly important as a proportion of all air pollution externalities. The use of the
central and upper values are shown below for the UK passenger car fleet. By 2010, under the high scenario, it is
estimated that the proportion of externalities from CO2 will comprise almost 20% (compared to just over 10 %
in 2000).
0%
20%
40%
60%
80%
100%
120%
140%
160%
180%
200%
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Euro5
Euro4
Euro3
Euro2
Euro1
Euro0
Figure 12 : Potential global warming costs from Belgian trucks (1993-2010)
0
2
4
6
8
10
12
14
16
18
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Clima te Change as % of Total Externalitie s
Ex te rn e 20 00 c en tr al
Ex te rn E 20 00 h igh
Figure 13 Potential global warming impacts for UK passenger cars (2000-2010)

5 Conclusions
This paper reports on aggregation studies for road sector externalities for Belgium and the UK. Work on the
underlying issues when moving to national level aggregation shows that the key factor in the analysis is likely to
be the level of dis-aggregation, especially with respect to urban emissions and their local impact. The level of
data across Europe is highlighted as a potential barrier to the application of such an approach more widely and it
seriously hampers attempts to make inter-national comparisons.
The studies here show that road sector externalities are extremely significant. Annual transport sector
externalities in 1997-8 from tailpipe emissions were estimated at 8.7 billion Euro in the UK and of 2.4 billion
Euro for Belgium. Passenger cars and heavy-duty vehicles are responsible for most of the damage. In both
countries, urban traffic is responsible for the majority of total sector externalities.
In both countries, the introduction of progressively tighter emissions standards has led to very significant
reductions in sector externalities, demonstrating the benefits of the Euro legislation programme. Looking at the
estimates for 2010, and assuming no thresholds for the exposure-response functions, it seems that we are coping
with regulated pollutants from passenger cars in both countries. Emissions from the haulage industry seem to be
under control in the UK only.
As better emission control technologies are introduced, there is a continued decrease in the associated
external costs. However, analysis of the data has shown that this is more a consequence of the scrapping of old
cars. In consequence, it may be that greater reductions in externalities, or at least much quicker improvements,
could be achieved through higher fleet turnover (i.e. targeting the removal older vehicles or specific vehicle
categories) than through stricter emissions legislation.
The analysis also shows that the relative effects of climate change are likely to become more important in
driving the sector externalities in the future as a result of continued fleet km growth combined with reductions in
other pollutants. Some estimates show that by 2010, potential climate change effects may comprise a quarter of
total air pollution externalities.
A comparison between the two countries shows some extremely interesting differences. The average cost per
kilometer is much higher in Belgium than in the UK. This is due to the higher proportion of diesel vehicles in
Belgium, and the fact that most UK diesel vehicles are newer. These differences remain due to the large
differential in diesel fuel duty between the two countries. Other reasons do exist: penetration of new Euro
standards was achieved slightly faster in the UK and resulted in more uncontrolled cars being scrapped.
The comparison of goods vehicles is different. Externalities have fallen consistently since 1990 in the UK
but in contrast have risen in Belgium. In both countries there has been a move to heavier goods vehicles in
recent years due to major changes in the haulage industry. However, the fleet turnover and the degree of switch
to heavier vehicles has been greater in the UK and therefore national fleet km are dominated by newer, cleaner
vehicles (albeit heavier ones). Again, emissions legislation has been important in driving the reductions in
externalities in the sector in the UK. However, as the comparison between countries shows, this is due to the
operating environment within the specific country, rather than European wide policies alone. However it was
shown that both European and national policies can have a large influence on national level externalities.
Acknowledgements
This paper is based on work co-financed by the Belgian Federal Office for Scientific, Technical and Cultural
Affairs (OSTC) and by the JOULE III program of the European Commission. The authors acknowledge the help
of Ina De Vlieger (Vito) and Tim Murrells (AEAT Environment) in collating the emissions and vehicle data.
References
[1] CEC, European Commission, DGXII, Science, Research and Development, JOULE. External costs of
Transport. Final Report, 2000 (in press)
[2] CEC, European Commission, DGXII, Science, Research and Development, JOULE. Externalities of Fuel
Cycles ‘ExternE’ Project. Report numbers 16520 EN to EUR 16525 EN.CEC, 1995.
[3] Int Panis, L., De Nocker, L., De Vlieger, I., Torfs, R.. Trends and uncertainty in air pollution impacts and
external costs of Belgian passenger car traffic. In: Actes du 9ème colloque Transport et pollution de l’air.
INRETS, actes n° 70 (ed. Robert Joumard), 1: 207-214, 2000.