Accounting for hidden energy dependency: The
impact of energy embodied in traded goods on
cross-country energy security assessments
Markus Bortolamedi∗†
Department of Business Administration, Economics and Law
Carl von Ossietzky University Oldenburg
March 20, 2015
Abstract
Energy security ranks high on the policy agenda of many countries.
Accordingly, to provide policy guidance, a large and growing body of
literature has proposed metrics to measure security of (primary) energy supply which are then applied in cross-country energy security
assessments. In general, the data used in these assessments are based
on production-oriented energy accounting frameworks. In doing this,
these studies neglect additional indirect foreign energy consumption
- i.e. consumption of energy embodied in traded goods. This paper
highlights this issue. It provides and applies a methodology that allows
including indirect foreign energy consumption into commonly used energy security indicators. In particular, it shows that the inclusion of
foreign primary energy embodied in traded goods does have an impact
on the values of energy security indicators as well as on the results of
cross-country energy security assessments - i.e. the indicator-specific
country rankings.
Keywords: energy security, energy security indicators, embodied energy, multi-regional input-output model
JEL Classification: D57, Q48
∗
Research funding is greatly acknowledged from the Lower Saxony Ministry for Science
and Culture (MWK). The ideas expressed here are remain those of the author who remains
solely responsible for errors and omissions.
†
E-Mail address: [email protected]. Tel.: +49 (0)4417984670
1
Introduction
Energy security ranks high on the policy agenda of many countries. Governments use the notion of energy security as a rationale for justifying massive
intervention into energy markets. As a prime example, the European Union
perceives its EU Energy Security and Solidarity Action Plan as an important
step to increase energy security for its member states (EU, 2008).
Accordingly, to provide policy guidance, a large and growing body of
literature proposes metrics to measure security of (primary) energy supply.
They measure either dependency levels on primary energy imports, primary
energy carriers and suppliers or levels of vulnerability, i.e. how much the
economy is exposed to events of supply disruptions essentially from an economic perspective (Bhattacharyya, 2009). These metrics are then frequently
applied in cross-country energy security assessments that rank countries according to the applied metrics at certain points in time. In general, they
are based on data coming from production-oriented accounting frameworks
for regional primary energy consumption - i.e. direct energy consumption
based on the territory principle - which are provided by various statistical
institutions, e.g. the International Energy Agency (IEA), the U.S. Energy
Information Agency (EIA), or Eurostat. Recent published studies providing
such cross-country assessments for European countries include, among others, Frondel and Schmidt (2014), Sovacool and Brown (2010), Löschel et al.
(2010), and Le Coq and Paltseva (2009).
However, energy is not only used directly in any given production and
final demand sector but also indirectly - i.e. in the production of imported
goods and services purchased for use in that sector (Battjes et al., 1998).
Since in a highly globalized world countries are heavily and increasingly involved in international trade, trade in non-energy goods and services can
substitute for trade and consumption of primary energy which in the end
masks energy security issues (Andrew et al., 2013). To put it more pointedly: If trading partners providing imports of non-energy goods and services
cannot secure their energy supply, how can energy security be provided without disconnecting from global trade of energy and non-energy goods (Tang
et al., 2013).
Since energy security assessments usually only focus on direct energy
use, they neglect this issue. This is of particular relevance for Europe. Because Europe is one of the dominant trading centers in terms of embodied
energy (Chen and Chen, 2011), commonly applied energy security assessments are based on only an incomplete picture of the European energy
situation. This implies that cross-country energy security assessments for
European countries might be inaccurate. Acknowledging indirect energy
consumption would require a switchover from production-oriented energy
accounting to more consumption-oriented approaches for energy accounting. This switchover in energy accounting would provide a novel perspective
1
on the European energy security problem. In the end, this would allow a
better-informed energy security debate.
So far only some rare studies address the implications of indirect energy
consumption for security of European primary energy supply. Bordigoni
et al. (2012) examine the role of these additional energy flows with a special focus on the European manufacturing industry. They apply a global
multi-region input-output (MRIO) model and find that in 2005 energy embodied in imported manufactured products used in European production
and final consumption comes close to energy consumed directly by the European industry and, therefore, has to be considered as a significant aspect
of the European energy situation. In a similar way, Tang et al. (2013) apply a single-region input-output (SRIO) model to quantify energy embodied
in international trade of the UK. They find that UK is a net-importer of
embodied fossil energy for every year within the time period 1997 to 2011.
Accordingly, since the gap between fossil energy consumption and domestic
fossil energy production is larger than commonly believed, they argue that
if net-imports are taken into consideration, the energy security problem is
greater than generally accepted. Therefore, they conclude that energy embodied in traded goods is another variable which should be considered in
the complex equation of energy security.
The principal objective of this paper is to further highlight the implications of the consideration of embodied energy in traded goods for commonly
conducted cross-country energy security assessments. First, a set of conventionally used energy security indicators is extended in such a way that
they also include indirect energy consumption. Second, a methodology is
provided that enables the quantification of indirect energy consumption.
Third, to quantify the impact of the switchover in the energy accounting
framework on indicator values and indicator-specific country rankings, the
methodology and the indicators are applied for 25 European countries for
the years 1995 to 2007.
The application shows that in many cases the incorporation of primary
energy embodied in traded goods has a substantial impact on the values
of the applied energy security indicators. Hereby, direction and magnitude
of these changes depend on the indicator as well as on the total amount
of regional net-imports of embodied energy, its composition as well as its
size relative to direct primary energy consumption. Since total and carrierspecific net-imports vary across countries, changes in the indicator values
are quite heterogeneous across countries for each indicator. Considering
indicator-specific country rankings, this heterogeneity is large enough to
cause substantial changes in the countries’ positions in the rankings. This is
particularly the case for primary energy carrier dependency. Therefore, the
question as to whether indirect energy consumption should be taken into
consideration is of particular relevance for comparing security of primary
energy supply across countries.
2
The remainder of the paper is organized as follows. Section 2 presents the
commonly used energy security indicators (section 2.1), the two alternative
energy accounting frameworks (section 2.2), the input-output model used
for the calculation of embodied energy in traded goods (section 2.3), and
the data used in the application (section 2.4). Section 3 presents the results
of the application. Section 4 concludes.
2
2.1
Methods
Energy security indicators
To quantify energy security, this study uses three indicators that are widely
used in the energy security literature (Löschel et al., 2010). The indicators capture dependency on (i) primary energy, (ii) primary foreign energy
supply, and (iii) primary energy carriers.
Primary energy dependency refers to the degree to which economic activities depend on primary energy as input. The larger the primary energy
input requirements of production and consumption activities are, the larger
the adjustment costs for the economy due to price shocks and physical supply
shortfalls are supposed to be (Kruyt et al., 2009). Primary energy intensity
(EI) of GDP is a commonly used indicator for primary energy dependency.
It is calculated as the ratio of total physical primary energy supply (TPES)
over gross domestic product (GDP).
T P ES
(1)
GDP
Dependency on external (foreign) primary energy supply provides insights about the domestic economy’s exposure to price and quantity risks
in global primary energy markets. Higher levels of primary energy imports
are considered as being more risky (Bhattacharyya, 2009). The crucial assumption behind this argument is that while governments can effectively
control domestic primary energy supply, they do not have control over external primary energy supply. The ratio of positive net-imports to total
primary energy consumption – referred to as net import dependency (NID)
– is commonly used as an indicator for primary energy import dependency.
The logic behind the use of positive net-imports rather than (gross) imports
is that shortfalls in energy imports can be compensated for by adjustments
in energy exports (Le Coq and Paltseva, 2009). NID is calculated as the
ratio of total positive net-imports of physical primary energy (PNI) over
total physical primary energy supply (TPES).
EI =
N ID =
3
PNI
T P ES
(2)
Dependency on primary energy carriers describes the reliance of economic activities on specific primary energy carriers (i.e. coal, natural gas,
crude oil, nuclear energy, and renewables). In the light of economy-wide
limited short-term substitution possibilities, high dependency on a single
primary energy carrier implies high exposure of the domestic economy to
price and quantity (supply) risks of a specific energy carrier – this situation
is generally considered as highly risky (Bhattacharyya, 2011). Concentration
or diversity indices are used to measure this dimension of dependency. A
wide-spread indicator here is the Herfindahl-Hirschman concentration index
applied to the primary energy mix of the economy’s total primary energy
consumption.1 Primary energy carrier dependency (PECD) is calculated as
the sum of the squared shares of total physical supply of primary energy
carrier pe (Spe ) in total physical primary energy supply (TPES). Larger
values of the indicator signal a more concentrated primary energy carrier
mix towards some particular primary energy carriers, which in turn implies
higher primary energy carrier dependency.
X Spe 2
P ECD =
T P ES
pe
(3)
In principle, the three indicators presented above in equation 1 to 3
do allow the quantification of energy security based on both, only direct
energy use and direct plus indirect energy use. However, it should be noted
that the calculation of total physical primary energy supply (TPES), total
positive net-imports of physical primary energy (PNI) as well as physical
supply of primary energy carrier pe (Spe ) depends on whether indirect energy
consumption is considered in the energy security assessment. Therefore,
the following section describes how to account for total and carrier-specific
energy consumption as well as for net-imports of primary energy within
the production-oriented and the consumption-oriented energy accounting
framework.
2.2
Production-oriented versus consumption-oriented energy
accounting
In analogy to the construction of regional greenhouse gas inventories, production-oriented and consumption-oriented accounting methods for domestic primary energy consumption differ in whether primary energy consumption is allocated to actors consuming primary energy directly for production
1
An alternative to the Herfindahl-Hirschman concentration index would be the
Shannon-Weaver index, which measures diversity rather than concentration. In contrast
to the latter, the Herfindahl-Hirschman concentration index puts relatively more weight
on the impact of larger shares of primary energy carriers in the fuel mix (Frondel and
Schmidt, 2014; Le Coq and Paltseva, 2009).
4
and final demand or to actors consuming final consumption goods which are
made with primary energy as input factor.2
In the production-oriented approach, total and carrier-specific energy
prod
consumption (T P ES prod and Spe
) include the amount of physical primary
energy which is directly consumed within the territory of a country (territory principle) by industries and households. Similarly, positive net-imports
of primary energy (P N I prod ) include only positive net-imports of physical
primary energy carriers (see equation 4 to 6).
T P ES prod = T P ES direct
(4)
prod
direct
Spe
= Spe
(5)
P N I prod = P N I direct
(6)
The switchover from production-oriented to consumption-oriented energy accounting requires including indirect primary energy consumption into
the calculation of total and carrier-specific primary energy consumption as
well as into the calculation of positive net-imports. In particular, total and
carrier-specific primary energy embodied in exports (EEE and EEEpe ) is
excluded and total and carrier-specific primary energy embodied in imports
(EEI and EEIpe ) is included (see equation 7 to 9).
T P ES cons = T P ES direct − EEE + EEI
(7)
cons
direct
Spe
= Spe
− EEEpe + EEIpe
(8)
P N I cons =P N I direct − EEE + EEI
(9)
This definition implies that primary energy required to produce a given
country’s exports is allocated to those countries that consume these exports.
Consequently, a part of domestic direct energy consumption is shifted to
other countries while a part of foreign direct primary energy consumption
is shifted to the domestic economy.
2.3
Multi-regional input-output model
Quantifying regional amounts of direct and indirect primary energy consumption requires analyzing intra-regional and inter-regional economic relationships among sectors of regional economies. For this, multi-regional
2
See Peters and Hertwich (2008) for an overview over different methodologies for constructing greenhouse gas inventories.
5
input-output (MRIO) models, that include the complete global production
chain of goods and services, are well established analytical frameworks.
The calculation of primary energy embodied in imports and exports
other than primary energy requires is based on a multi-regional input-output
(MRIO) model extended by sector- and carrier-specific primary energy consumption. The model includes r ∈ R regions, i ∈ I production sectors, or
equivalently, commodities, g ∈ G activities consisting of all production sectors plus public expenditure, investment and final consumption, and j ∈ J
international transport services. Ygr denotes the output in producing sectors and the level of public expenditure, investment and final consumption
in region r. Xisr are exports of commodity i from region s to region r, while
D (Z M ) denote domestic
Mir are imports of commodity i in region r. Zigr
igr
(imported) intermediate inputs of commodity i in activity g in region r.
Finally, Tjr stands for international transport service j produced in region
r.
The model includes five different primary energy carriers pe ∈ P E which
are the following: crude oil, coal, natural gas, nuclear energy and renewables.
The carrier-specific primary energy content of a good is composed of primary energy used in the production of the good itself as well as of primary
energy that is used to produce required intermediate inputs and associated
transport services. To calculate the full carrier-specific primary energy content (per USD of output) input-output accounting identities are used and
the associated linear system of equations is solved for the primary energy
content of production activities (pecYpe,gr ), the primary energy content of imports (pecM
pe,ir ), and the primary energy content of international transport
services (pecTj ).
Total carrier-specific primary energy embodied in output of activity g
in region r is equal to the sum of direct carrier-specific primary energy use
(evpe,gr ), carrier-specific embodied primary energy in domestic intermediates, and carrier-specific embodied primary energy in imported intermediates (see equation 10). Total carrier-specific primary energy embodied in
imports of commodity i in region r is equal to the sum of carrier-specific
embodied primary energy of all exports from region s to region r plus the
carrier-specific primary energy embodied in international transport services
(see equation 11). Carrier-specific primary energy embodied in international
transport services j is equal to the sum of carrier specific embodied energy in
the production of international transport services across countries (equation
12).
pecYpe,gr Ygr = evpe,gr +
| {z }
| {z }
energy embodied
in output
direct
energy use
X
M
pecM
pe,ir Zigr
i∈I
|
+
X
D
pecYpe,ir Zigr
i∈I
{z
}
energy embodied in
imported intermediates
6
|
{z
}
energy embodied in
domestic intermediates
(10)
X
X
Y
=
pec
X
+
pecTpe,j Tjisr
pecM
M
isr
ir
pe,is
pe,ir
| {z }
s∈R
j∈J
energy embodied
|
{z
} |
{z
}
in imports
energy embodied in
exports
pecTpe,j
X
Tpe,jr =
{z
embodied energy
energy embodied in
transport services
X
pecYpe,jr Tjr
(12)
r∈R
r∈R
|
(11)
|
}
{z
}
embodied energy in
in transport
inputs to transport
The MRIO model can be solved directly as a square system of equations
or solved recursively using a diagonalization algorithm. As conducted in a
similar way by Böhringer et al. (2011), this paper uses the latter approach.
The required data for the parameters are provided by the World InputOutput Database (WIOD) database and the Eurostat statistical database.
2.4
Data
The analysis is based on data taken from the World Input-Output Database
(WIOD). It consists of time series of detailed national and world inputoutput tables as well as of socio-economic and environmental accounts for
35 sectors in 41 countries for the years 1995 to 2009, whereby the environmental accounts include data on gross energy use by sector and 26 energy
commodities (Timmer, 2012). Considering regional aggregation, the data set
contains each of the EU27 countries as separate regions. For the analysis
the set of WIOD primary energy carriers is aggregated towards a set of five
different primary energy carriers pe ∈ P E as defined in section 2.3. Table
1 provides an overview on the aggregation of the different WIOD primary
energy carriers towards the aggregated set of primary energy carriers.
Since the WIOD data does not contain information on energy-commodity
specific import shares or consumption of commodity-specific energy coming
from abroad, satellite data on fuel-specific net-import shares obtained from
the Eurostat statistical database is used to split the gross energy use into
gross use of domestic and gross use of imported energy (evdpe,gr and evipe,gr
respectively). Hereby it is assumed that the share of carrier-specific consumption of imported energy in total energy consumption is identical in
each activity g.
7
pe ∈ P E
WIOD code
Crude oil
CRUDE
Coal
HCOAL, BCOAL, COKE
Natural gas
NATGAS, OTHGAS
Nuclear energy
NUCLEAR
Renewables
WASTE, BIOGASOL, BIODIESEL, BIOGAS,
GEOTHERM, HYDRO, SOLAR, WIND, OTHSOURC
Table 1: Aggregation of WIOD primary energy carriers
3
Results and discussion
This section shows the results of an application of the methodology presented above to a sample of 25 countries out of the 27 member states of the
European Union for the time period of 1995 to 2007.3 In doing so, annual
values for the three energy security indicators are calculated for each sample
country based on the production-oriented as well as on the consumptionoriented accounting framework for primary energy consumption as laid out
in section 2.2. The production-oriented indicator values are then taken as
a reference against which changes in the indicator values that occur due
to the switchover from production-oriented to consumption-oriented energy
accounting are quantified.
The analysis will be focused on three aspects which are of particular relevance for commonly conducted energy security assessments. First, section
3.1 assesses to what extent the impact of the switchover from productionoriented to consumption-oriented energy accounting varies across the three
applied energy security indicators. This will provide some hints on how sensible these indicators are to changes in the energy accounting framework as
well as on the general direction of the impact. Furthermore, it assesses to
what extent these changes are triggered by net-imports of primary energy
embodied in traded goods. Section 3.2 focuses on the heterogeneity of the
3
Cyprus is excluded from the sample because changes in energy intensity are comparatively large which would make a reasonable presentation of the heterogeneity in the
impact difficult. Malta is excluded because its energy mix almost exclusively comprises
imported refined oil. Hence, relative changes in the indicator values are either very large
or nonexistent when production-oriented indicators have a value of zero. Furthermore,
the years 2008 and 2009 are excluded from the analysis due to the impacts of the financial
crisis on global trade.
8
obtained changes along the regional dimension - i.e. between-country heterogeneity - for each of the years considered. This is of particular relevance
for cross-country energy security assessments that provide rankings based
on energy security indicators for specific points in time. If there is much
heterogeneity in the changes in the indicator values between the countries,
these rankings could provide only an inaccurate picture of cross-country
differences in energy security. By conducting a sequence of stylized crosscountry energy security assessments for each year in the time period 1995
to 2007, section 3.3 then assesses whether changes in the energy accounting
framework alter year and indicator-specific country rankings.
3.1
Heterogeneity across indicators
This section compares the variability in the changes in the obtained indicator
values across the three applied energy security indicators. Figure 1 provides
an overview of the variability in the obtained changes in the indicator values
(x-axis) for each of the applied energy security indicator (y-axis). Since the
sample includes 25 countries and 13 years, there are in total 325 observations
for each indicator.
Figure 1: Changes in energy security indicators due to inclusion of primary
energy embodied in traded goods.
It is apparent from figure 1 that direction and magnitude vary between
the applied indicators. On the one hand the indicator value increases in
the most cases - i.e. the indicator worsens - for energy intensity and netimport dependency. In particular, indicator values increase for roughly 75%
9
of all observations considered.4 On the other hand, indicator values for
primary energy carrier dependency decrease - i.e. the indicator improves
- for roughly 75% of the cases considered.5 Additionally, figure 1 shows
that the distribution of changes in the indicator values is asymmetric for
each indicator. Specifically, since there is a tail of high (extreme) outliers
stretching towards the right, changes in the indicator values are positively
skewed for energy intensity and net-import dependency. On the contrary,
since for primary energy carrier dependency the lower whisker is longer than
the upper whisker, changes in this indicator are negatively skewed.
Furthermore, it is also observable that the obtained positive and negative
changes in the indicator values are much more pronounced for energy intensity and net-import dependency than for primary energy carrier dependency.
Specifically, the variance of energy intensity and net-import dependency is
about 8 times the variance of primary energy carrier dependency. Normalization by the mean reduces differences in the spreading of changes in the
indicator values. In particular, the coefficient of variation for net-import
dependency is more than two times the coefficient of variation for primary
energy carrier dependency. The coefficient of variation for energy intensity
assumes an intermediate position. Table 3.1 shows summary statistics for
changes in energy intensity (EI), net-import dependency (NID) and primary
energy carrier dependency (PECD).
observations
minimum
maximum
median
mean
variance
coefficient of variation
EI
325
-48.0%
253.7%
15.8%
23.9%
2082.9
1.9
NID
325
-84.9%
216.9%
5.2%
15.3%
1925.8
2.9
PEDC
325
-65.6%
24.5%
-10.6%
-13.3%
259.1
1.2
Table 2: Descriptive statistics for changes of energy intensity (EI), netimport dependency (NID) and primary energy carrier dependency (PECD).
Because of the construction of the indicators, it is clear that changes in
the indicator values are triggered by net-imports of primary energy embodied
in traded goods. Therefore, to get some brief insights into the relationship
between net-imports of embodied energy and changes in the indicator values,
figure 2 to 4 plot changes in the indicator values against total net-imports
of primary energy embodied in traded goods (in petajoule).
Figure 2 shows a weak positive correlation between net-imports of em4
Specifically, the lower quartile is at -5.9% for energy intensity and at -5.1% for netimport dependency.
5
The upper quartile is at -2% for primary energy carrier dependency.
10
bodied energy and changes in energy intensity. In particular, the Pearson
correlation coefficient has a value of 0.293 and is significant at a level of 0.01
(two sided). This weak correlation is caused by the fact that while the direction of changes in energy intensity is determined by the sign of net-imports,
the magnitude of these changes is determined by the importance of embodied energy in total direct and indirect primary energy consumption - i.e.
the ratio of net-imports of embodied energy over total direct primary consumption - which varies widely across the observation within the sample.6
Larger absolute values of this ratio result in larger changes in energy intensity whereby their direction depends on whether net-imports are positive or
negative.
With regard to net-import dependency, figure 3 shows a positive correlation between net-imports of embodied energy and changes in the indicator
values which is somewhat weaker than between net-imports of embodied energy and energy intensity. In particular, the Pearson correlation coefficient
has a value of 0.288 and is significant at a level of 0.01 (two sided).
Figure 4 shows a positive correlation between net-imports of embodied
energy and changes in primary energy carrier dependency which is weaker
than for net-import dependency and energy intensity. Specifically, the Pearson correlation coefficient has a value of 0.187 and is significant at the 0.01
level (two sided). This is due to the fact that changes in primary energy
dependency do not depend on net-imports of embodied energy at the aggregated level but on carrier-specific net-imports, particularly on the amount
of carrier-specific net-imports of primary energy relative to carrier-specific
direct consumption of primary energy. Moreover, since in the most cases primary energy carrier dependency improves due to the switchover in the energy
accounting framework, positive net-imports of embodied energy mostly contain primary energy carriers that have low consumption shares in total direct
primary energy consumption while negative net-imports - i.e. net-exports mostly contain primary energy carriers that have high consumption shares
in total direct primary energy consumption.
From a policy perspective, this implies that if indirect energy consumption is included, the goal of lowering dependency on (foreign) primary energy
is more difficult to achieve. This is because it no more possible to simply
substitute direct energy consumption for indirect energy consumption. On
the other hand, achieving the goal of lowering dependency on primary energy
carriers becomes easier.
6
This deterministic relationship between the ratio of embodied energy to total direct
energy consumption and changes in energy intensity is due to the construction of the
indicator which measures changes in energy intensity. In particular, percentage changes
in energy intensity, (EI cons /EI prod )−1, can be rearranged to (EEI −EEE)/T P ES direct .
11
Figure 2: Correlation between net-imports of primary energy embodied in
traded good and changes in energy intensity.
Figure 3: Correlation between net-imports of primary energy embodied in
traded good and changes in net-import dependency.
12
Figure 4: Correlation between net-imports of primary energy embodied in
traded good and changes in primary energy dependency.
3.2
Heterogeneity between countries
This section assesses heterogeneity of changes in the indicator values between countries. In doing so, the data is arranged according to the time
dimension for each indicator. Figure 5 to 7 depict the variation of changes
in the indicator values between countries for each year and for each of the
three applied energy security indicators. Figure 5 considers energy intensity
(EI), figure 6 deals with net-import dependency (NID) and figure 7 looks at
primary energy carrier dependency (PECD).
Figure 5 shows that while the lower quartile is negative for each year
considered, the median is positive throughout. This implies that in each
year, the switchover from a production-oriented to a consumption-oriented
energy accounting framework increases energy intensity for at least half of
the countries considered. Furthermore, it is observable that over the years
changes in the indicator values become more positive. In particular, the
median in 2007 is more than twice the size of the median in 1995. Additionally, the lower quartile moves closer to zero. Although absolute spreading
- i.e. the variance - increases between 1995 and 2007, relative spreading i.e. the coefficient of variation - decreases. Table A1 in appendix A provides
descriptive statistics for changes in energy intensity for each year.
Turning to net-import dependency, figure 6 shows that, as it is the case
for energy intensity, in each year the obtained changes in the indicator values
increase for at least half of the countries considered. Hereby, the same logic
applies as above: while the lower quartile is negative for all years consid13
ered, the median is positive throughout. Furthermore, because the median
and the lower quartile increase from 1995 to 2007, changes in net-import
dependency also become more positive. As for energy intensity, absolute
spreading increased from 1995 to 2007 but relative spreading decreased. Table A2 in appendix A provides descriptive statistics for changes of net-import
dependency for each year.
With regard to primary energy carrier dependency, figure 7 shows that
in the years 1995 to 1999 obtained changes are negative for at least half of
the countries considered. After 1999, obtained changes in primary energy
carrier dependency are negative for at least three quarters of the countries
considered. Therefore, the distribution of observations shifts somewhat to
the left. In contrast to energy intensity and net-import dependency, absolute
and relative spreading both decrease for primary energy carrier dependency
from the year 1995 to 2007. Table A3 in appendix A provides descriptive
statistics for changes of primary energy carrier dependency for each year.
Considering country-specific changes in the indicator values for each
year, the countries can be allocated to one of the following four groups:
negative indicator changes for all years considered (group 1), negative indicator changes for at least half of the years considered (group 2), negative
indicator changes for less than half of the years considered (group 3), and
positive indicator changes for all years considered (group 4). Table 3 provides an overview over the allocation of each country in the sample to one
of these four groups. From this table it can be seen that while each indicator increases for Germany in each year considered, the opposite holds for
Belgium, Czech Republic, and the Netherlands.
Figure 5: Changes in energy intensity due to inclusion of primary energy
embodied in traded goods.
14
Figure 6: Changes in net-import dependency due to inclusion of primary
energy embodied in traded goods.
Figure 7: Changes in dependency of primary energy carriers due to inclusion
of primary energy embodied in traded goods.
15
Austria
Belgium
Bulgaria
Czech Republic
Denmark
Estonia
Finland
France
Germany
Great Britain
Greece
Hungary
Ireland
Italy
Latvia
Lithuania
Luxembourg
Netherlands
Poland
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden
EI
4
1
1
1
3
3
2
4
4
4
4
4
4
4
4
3
4
1
2
4
2
1
4
4
2
NID
4
1
1
1
4
3
2
4
4
4
4
4
4
4
4
3
4
1
2
4
4
1
4
4
2
PECD
1
1
2
1
3
1
2
1
4
1
1
1
3
1
1
2
1
1
1
1
1
3
2
1
1
Table 3: Country-specific changes in the values of energy intensity (EI), netimport dependency (NID) and primary energy carrier dependency (PECD)
for each year.
Key: 1 - negative changes for all years considered; 2 - negative changes for
at least half of the years considered; 3 - negative changes for less than half
of the years considered; 4 - positive changes for all years considered.
3.3
Production-oriented versus consumption-oriented country rankings
This section shows the implications of changes in the indicator values for
practical cross-country energy security assessments. In particular, it checks
whether a switchover from a production-oriented to a consumption-oriented
energy accounting framework alters indicator-specific country rankings for
given years. In doing this, it conducts a sequence of stylized cross-country
energy security assessment across the 25 sample countries for each of the
16
years within the time period 1995 to 2007. They are based on the three
energy security indicators as laid out in section 2.1 and calculated based
on the production-oriented as well as the consumption-oriented energy accounting framework as laid out in section 2.2. Each of these assessments results in an indicator-specific production-oriented and consumption-oriented
country ranking. To quantify differences between the production-oriented
and the consumption-oriented country ranking for a given indicator in a
given year, Spearman’s rank correlation coefficient - i.e. Spearman’s ρ - is
used. It is a non-parametric measure of statistical dependence between two
variables ranging from −1 (perfectly negative correlation) to 1 (perfectly
positive correlation). A value of 0 implies that the two variables are statistically independent. Hence, it is an inverse measure for the difference
between the country rankings, i.e. lower values imply higher differences
between the country rankings. In particular, while perfectly positive correlation implies that the production-oriented country ranking is identical
to the consumption-oriented country ranking, perfectly negative correlation
implies that the consumption oriented country ranking is reversed. Hence,
values smaller than one indicate that the production-oriented and the corresponding consumption-oriented country ranking are not identical. Table 4
provides values of Spearman’s rank correlation coefficient for energy intensity (second column), net-import dependency (third column), and primary
energy carrier dependency (fourth column) for each of the applied years
(first column).
Year
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
∗∗ correlation is
EI
NID
PECD
0.936∗∗
0.875∗∗
0.613∗∗
0.920∗∗
0.877∗∗
0.787∗∗
∗∗
∗∗
0.947
0.855
0.765∗∗
0.953∗∗
0.911∗∗
0.663∗∗
∗∗
∗∗
0.943
0.893
0.602∗∗
∗∗
∗∗
0.925
0.894
0.705∗∗
0.931∗∗
0.883∗∗
0.699∗∗
∗∗
∗∗
0.945
0.888
0.698∗∗
∗∗
∗∗
0.930
0.895
0.689∗∗
0.922∗∗
0.875∗∗
0.652∗∗
∗∗
∗∗
0.915
0.882
0.672∗∗
0.902∗∗
0.868∗∗
0.684∗∗
∗∗
∗∗
0.862
0.868
0.607∗∗
significant at a level of 0.01 (both sides).
Table 4: Differences between production-oriented and consumption oriented
country rankings for energy intensity (EI), net-import dependency (NID)
and primary energy carrier dependency (PECD)
17
As can be seen from the table, for each indicator there is a significant
positive correlation between production-oriented and consumption-oriented
indicator values for all years considered. This implies that countries having
a low (high) rank in the production-oriented country ranking tend to have
a low (high) rank in the consumption-oriented country ranking. However,
there is no case for which the two rankings are perfectly positive correlated.
Hence, the switchover from production-oriented to consumption-oriented energy accounting alters the country ranking for every indicator and for every
year considered. Interestingly, although variability of changes in primary energy dependency is the lowest, country rankings for primary energy carrier
dependency change the most. This is because region-specific productionoriented indicator values for primary energy carrier dependency lie closer
together than for the other two indicators. This makes it more likely that
changes in the indicator values are large enough such that they affect the
country ranking.
Since in 2007 the country rankings are the least positively correlated,
the remainder of this section takes a closer look at the changes in the
country rankings for the year 2007. Table 5 provides the consumptionoriented country ranking for energy intensity (EI), net-import dependency
(NID), and primary energy carrier dependency (PECD). Changes between
the production-oriented and consumption-oriented country ranking are included in brackets. In particular, upward (downward) arrows and the number behind the arrow show deviations from the countries’ position in the
production-oriented country ranking - i.e. how many positions a country
has moved upward (downward) in the country ranking due to the switchover
to the consumption-oriented energy accounting framework. Appendix B
provides further information about the total amount and composition of
regional direct and embodied energy consumption as well as of GDP for
the year 2007 on which the calculation of the indicator values is based. In
particular, table B1 provides information about the amount and the composition of regional total direct primary energy consumption, regional and
carrier-specific import shares as well as information about regional GDP.
Table B2 provides information about the amount and the composition of
regional total net-imports of embodied energy, as well as the importance
of total net-imports of embodied energy in regional imports of primary energy (the import imbalance ratio - IIR) and direct energy consumption (the
energy imbalance ratio - EIR).
With regard to energy intensity, table 5 shows that only seven countries
- i.e. Luxembourg, Ireland, Denmark, Italy, France, Hungary, and Poland maintain their position after the switchover in the energy accounting framework. Considering absolute changes, the Netherlands improves the most,
i.e. it moves up ten positions. On the contrary, Latvia worsens the most,
i.e. moves down nine positions.
Turning to net-import dependency, table 5 shows that the number of
18
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
EI
Luxembourg
Ireland
Denmark
Great Britain (↑ 1)
Netherlands (↑ 10)
Italy
Austria (↓ 3)
Germany (↑ 1)
Sweden (↑ 5)
Spain (↓ 3)
France
Portugal (↓ 2)
Belgium (↑ 3)
Finland (↑ 3)
Greece (↓ 2)
Slovenia (↓ 8)
Slovakia (↑ 6)
Hungary
Poland
Czech Republic (↑ 1)
Latvia (↓ 9)
Romania (↓ 2)
Estonia (↓ 1)
Bulgaria (↑ 1)
Lithuania (↓ 1)
NID
Estonia
Czech Republic (↑ 4)
Poland (↑ 2)
Bulgaria (↑ 8)
Romania (↑ 2)
Great Britain (↓ 4)
Sweden (↑ 1)
Netherlands (↑ 5)
Denmark (↓ 6)
Finland (↑ 1)
France (↓ 1)
Slovenia (↓ 8)
Germany (↑ 2)
Slovakia (↑ 3)
Hungary (↑ 1)
Austria (↓ 2)
Belgium (↑ 4)
Latvia (↓ 9)
Greece
Spain
Lithuania (↓ 3)
Portugal
Italy (↑ 1)
Ireland (↓ 1)
Luxembourg
PECD
Finland
Sweden (↑ 10)
Slovakia (↓ 1)
Slovenia (↑ 19)
Latvia (↑ 2)
France (↑ 7)
Germany (↓ 4)
Austria (↑ 1)
Romania (↓ 5)
Belgium (↑ 7)
Bulgaria (↓ 6)
Hungary (↓ 5)
Luxembourg (↑ 12)
Czech Republic (↓ 3)
Spain (↓ 5)
Great Britain (↓ 2)
Denmark (↓ 9)
Netherlands (↑ 2)
Portugal (↓ 1)
Ireland (↓ 5)
Italy (↓ 2)
Estonia (↑ 2)
Greece (↓ 1)
Poland (↓ 3)
Lithuania (↓ 9)
Table 5: Consumption-oriented country ranking for energy intensity (EI),
net-import dependency (NID) and primary energy carrier dependency
(PECD) for the year 2007. In brackets: Deviation from the countries’ position in the production-oriented country ranking - i.e. number of positions
a county has moved up (↑) or down (↓) due to the switchover in the energy
accounting framework.
19
countries that keep their position is smaller as for energy intensity. In particular, only Estonia, Greece, Spain, Portugal and Luxembourg each country
keep their positions in the country ranking. With regard to the countries
that change positions, Bulgaria improves the most, i.e. it moves up eight
positions, while again Latvia worsens the most, i.e. it moves down nine
positions.
In contrast to energy intensity and net-import dependency, figure 5 shows
that for primary energy carrier dependency only Finland keeps its position in the country ranking after the switchover in the energy accounting
framework. While Slovenia improves the most by moving up 19 positions,
Denmark and Lithuania worsen the most by moving down nine positions.
Against the background of these changes in the overall country rankings, it is obvious that there is also a large amount of changes in relative
positions within country pairs. Table 6 provides a matrix of all possible
country pairs and includes information on whether relative positions of the
two member countries of each country pair changes for the three energy security indicators. 1, 2, and 3 signal changes in relative positions for only
energy intensity, net-import dependency, and primary energy carrier dependency. 4 and 5 stand for changes in relative positions for energy intensity and
either net-import dependency or primary carrier dependency. 6 represents
changes in relative position for net-import dependency and primary energy
carrier dependency. Finally, 7 indicates that relative positions change for all
of the three indicators.
It can be seen from the matrix that changes in relative positions for
primary energy carrier dependency occur twice as frequent as changes in
relative positions for energy intensity and net-import dependency. Specifically, while for 27% of the country pairs turning from production-oriented
to consumption-oriented energy accounting changes relative positions of the
members of the country pair, it is only the case for 14% for energy intensity
and 15% for net-import dependency.
20
21
Austria
3
3
2
1
1
6
6
3
2
1
3
3
Belgium
1
6
7
6
3
5
3
3
3
3
3
Bulgaria
6
6
2
6
1
1
1
2
3
3
2
3
2
3
2
6
2
6
3
Czech Republic
2
3
3
Denmark
3
6
2
3
2
2
3
3
3
3
3
2
6
Estonia
1
3
3
3
3
Finland
5
2
5
1
2
2
France
5
3
1
1
3
5
6
3
3
2
6
3
3
3
Germany
1
3
1
6
3
2
Great Britain
2
2
3
3
1
3
2
2
3
Greece
1
1
6
3
1
3
3
1
5
Hungary
3
5
7
3
6
3
Ireland
3
3
3
2
3
3
Italy
3
3
3
2
1
Latvia
3
1
3
3
5
7
3
7
3
3
5
6
6
3
3
6
7
6
Lithuania
2
3
3
3
3
3
3
6
3
6
1
Netherlands
3
3
3
3
3
3
3
3
3
3
3
3
Poland
5
1
1
7
3
3
3
1
2
5
1
2
1
Portugal
1
2
3
3
2
3
2
2
3
Romania
1
1
3
3
3
3
1
Slovakia
3
1
6
3
2
3
2
1
2
3
Slovenia
3
1
1
5
5
1
1
2
7
5
2
1
6
1
5
5
1
6
2
Spain
3
2
3
1
3
1
3
3
6
1
3
3
7
1
1
1
3
1
3
6
3
6
3
Table 6: Change in relative positions of the countries after switchover in the energy accounting framework.
Key: 1 - change in relative position for energy intensity (EI); 2 - change relative position for net-import dependency (NID); 3
- change in relative position for primary energy carrier dependency (PECD); 4 - change in relative position for EI and NID;
5 - change in relative position for EI and PECD; 6 - change in relative position for NID and PECD; 7 - change in relative
position for EI, NID and PECD
Austria
Belgium
Bulgaria
Czech Republic
Denmark
Estonia
Finland
France
Germany
Great Britain
Greece
Hungary
Ireland
Italy
Latvia
Lithuania
Luxembourg
Netherlands
Poland
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden
Sweden
4
Conclusions and policy implications
Energy security ranks high on the policy agenda of many countries. Accordingly, to provide policy guidance, there is a large and growing body of literature that applies various metrics that measure security of primary energy
supply in cross-country energy security assessments. This paper argues that
country rankings that result out of these energy security assessments might
be inaccurate because indirect primary energy consumption - i.e. primary
energy embodied in traded goods - is generally neglected. Since Europe is
a major trading center in terms of embodied energy, this is of particular
relevance for the member countries of the European Union.
Accordingly, this paper has illustrated the implications of including indirect energy consumption into commonly conducted cross-country energy
security assessments. In doing so, first a methodology for the inclusion of
indirect energy consumption into commonly applied energy security indicators is provided. Second, to quantify region- and carrier-specific indirect
primary energy consumption, a multi-regional input-output model extended
by sectoral and carrier-specific primary energy consumption is provided and
applied for a sample of 25 European countries for the time period 1995
to 2007. Finally, changes in indicator values and indicator-specific country
rankings are quantified based on the input-output calculations.
It has been shown that changes in the indicator values are quite substantial for energy intensity and net-import dependency while they are comparatively small for primary energy dependency. Moreover, while energy
intensity and net-import dependency worsen if regional net-imports of primary energy embodied in traded goods are positive, primary energy carrier
dependency improves for most of the countries irrespective of the size of netimports. Hence, because one indicator improves while two others worsen,
being a net-importer of embodied energy does not necessarily imply that the
energy security problem is larger than commonly believed.7 This is because
not only the total amount of net-imports matters but also its composition.
Furthermore, it has been shown that changes in the indicator values are considerable heterogeneous across countries for given years. This is of particular
relevance for indicator-specific country rankings. In particular, it has been
shown that production-oriented and consumption-oriented country rankings
differ for each indicator and for each year. Particularly for the year 2007
absolute changes in the positions of some countries are quite large, especially
for primary energy carrier dependency. Furthermore, a direct comparison of
7
Since each indicator uses a different metric for energy security it is not possible to
tradeoff improvements of primary energy carrier dependency with decreases in energy
intensity and net-import dependency. In particular, while on the one hand the energy
situation becomes more risky - i.e. higher energy requirements, higher requirements of
foreign energy - it also becomes less risky - i.e. the primary energy mix is less concentrated
towards some specific primary energy carriers.
22
country pairs shows that changes in relative positions occur quite frequently
for primary energy carrier dependency and less frequently for energy intensity and net-import dependency.
The main purpose of this paper was to motivate future research by illustrating the sensitivity of cross-country energy security assessments to
changes in energy accounting. For this, it has quantified differences in indicator values and country rankings that result out of a complete switchover from
production-oriented to consumption-oriented energy accounting. Clearly,
this complete switchover is based on the implicit assumption that direct
and indirect energy consumption are equally important for energy security
- i.e. that disruption costs are completely passed through along the supply
chain of goods and services. However, while ignoring indirect energy consumption would be a mistake, equating it with direct energy consumption
would be equally wrong (Wagner, 2010). Hence, of particular relevance for
future research is the question of how important indirect energy consumption is for energy security. Its answer would provide a solid foundation for
the construction of an appropriate energy accounting framework that lies
somewhere in between the completely production-oriented and completely
consumption-oriented frameworks that were presented in this paper. Another issue is whether policies that are commonly considered for improving
energy security - e.g. increasing the use of renewables, increasing energy
efficiency - are equally effective if indirect energy consumption is taken into
account. In essence, the relevant question is whether these policies induce
sufficiently large substitution of direct energy consumption for indirect energy consumption such that, in the end, energy security is adversely affected.
References
Andrew, R. M., Davis, S. J., and Peters, G. P. (2013). Climate policy and dependence on traded carbon. Environmental Research Letters, 8(3):034011.
Battjes, J., Noorman, K., and Biesiot, W. (1998). Assessing the energy
intensities of imports. Energy Economics, 20(1):67–83.
Bhattacharyya, S. C. (2009). Fossil-fuel dependence and vulnerability of
electricity generation: Case of selected European countries. Energy Policy,
37(6):2411–2420.
Bhattacharyya, S. C. (2011). Energy Economics. Springer London, London.
Böhringer, C., Carbone, J. C., and Rutherford, T. F. (2011). Embodied Carbon Tariffs. NBER Working Paper 17376, National Bureau of Economic
Research, Inc.
23
Bordigoni, M., Hita, A., and Le Blanc, G. (2012). Role of embodied energy in
the European manufacturing industry: Application to short-term impacts
of a carbon tax. Energy Policy, 43:335–350.
Chen, Z. M. and Chen, G. Q. (2011). An overview of energy consumption
of the globalized world economy. Energy Policy, 39(10):5920–5928.
EU (2008). Second Strategic Energy Review - An EU Energy Security and
Solidarity Action Plan. Technical Report COM(2008) 781 final, European
Commission.
Frondel, M. and Schmidt, C. M. (2014). A measure of a nation’s physical
energy supply risk. The Quarterly Review of Economics and Finance,
54(2):208–215.
Kruyt, B., van Vuuren, D. P., de Vries, H. J. M., and Groenenberg, H.
(2009). Indicators for energy security. Energy Policy, 37(6):2166–2181.
Le Coq, C. and Paltseva, E. (2009). Measuring the security of external
energy supply in the European Union. Energy Policy, 37(11):4474–4481.
Löschel, A., Moslener, U., and Rübbelke, D. T. G. (2010). Indicators of
energy security in industrialised countries. Energy Policy, 38(4):1665–
1671.
Peters, G. P. and Hertwich, E. G. (2008). Post-Kyoto greenhouse gas inventories: production versus consumption. Climatic Change, 86(1-2):51–66.
Sovacool, B. K. and Brown, M. A. (2010). Competing Dimensions of Energy
Security: An International Perspective. Annual Review of Environment
and Resources, 35(1):77–108.
Tang, X., Snowden, S., and Höök, M. (2013). Analysis of energy embodied
in the international trade of UK. Energy Policy, 57:418–428.
Timmer, M. P. (2012). The World Input-Output Database (WIOD): Contents, Sources and Methods. WIOD Working Paper 10.
Wagner, G. (2010).
38(12):7710–7721.
Energy content of world trade.
24
Energy Policy,
A
Year
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
Descriptive statistics cross-country heterogeneity
N
25
25
25
25
25
25
25
25
25
25
25
25
25
Min
-32.2%
-40.1%
-48%
-32.8%
-23.1%
-28.5%
-36.5%
-32.3%
-31.6%
33.9%
-26.4%
-41.6%
-30.1%
Max
171.8%
181.8%
192.2%
200.8%
212.0%
253.7%
202.5%
129.8%
153.7%
136.7%
166.6%
151.1%
164.5%
Median
12.7%
10.6%
7.3%
10.9%
13.8%
16.5%
14.8%
16.1%
15.8%
21.3%
19.4%
19.6%
20.6%
Mean
19.8%
20.1%
17.8%
20.9%
27.0%
28.1%
22.6%
20.0%
23.2%
24.2%
26.7%
27.4%
33.4%
VAR
1906.0
2225.6
2132.5
2128.5
2387.4
3220.3
2245.4
1368.0
1682.7
1818.6
2122.6
2142.6
2497.0
CV
2.2
2.3
2.6
2.2
1.8
2.0
2.1
1.8
1.8
1.8
1.7
1.7
1.5
Table A1: Descriptive statistics cross-country heterogeneity for changes in
energy intensity.
Key: N - number of observations, Min - minimum, Max - maximum, VAR
- Variance, CV - Coefficient of variation
Year
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
N
25
25
25
25
25
25
25
25
25
25
25
25
25
Min
-84.9%
-53%
-84.2%
-47.8%
-38.3%
-48.1%
-57.6%
-50.7%
-48.7%
-49.2%
-35.5%
-67.3%
-35%
Max
88.5%
71.3%
66.1%
171.8%
216.9%
192.4%
129.8%
172%
153.9%
167.5%
162.6%
121.2%
129.1%
Median
5.2%
5.9%
3.9%
4.8%
4.7%
5.2%
5.7%
5.3%
5.0%
4.9%
4.4%
4.2%
9.6%
Mean
6.4%
5.6%
3.3%
15.2%
22.8%
21.1%
17.1%
18.1%
16.4%
18.6%
18.4%
15.7%
20.4%
VAR
1304.8
932.6
1041.0
2039.0
2945.4
2832.2
2127.1
2596.7
1958.7
2425.6
2131.8
1623.3
1553.8
CV
5.6
5.5
9.8
3.0
2.4
2.5
2.7
2.8
2.7
2.6
2.5
2.6
1.9
Table A2: Descriptive statistics cross-country heterogeneity for net-import
dependency.
Key: N - number of observations, Min - minimum, Max - maximum, VAR
- Variance, CV - Coefficient of variation
25
Year
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
N
25
25
25
25
25
25
25
25
25
25
25
25
25
Min
-55.6%
-54.3%
-56.9%
-59.9%
-60.8%
-55.7%
-57.5%
-64.7%
-65.5%
-65.6%
-64.6%
-64.2
-64.0%
Max
24.5%
14%
24.1%
3.9%
9.5%
8.7%
4.5%
2.9%
3.2%
3%
4.7%
4.9%
14.3%
Median
-8.6%
-9.1%
-10.4%
-10.7%
-12.1%
-8.7%
-13.4%
-11.9%
-12.3%
-11.6%
-9.9%
-9.8%
-10.4%
Mean
-10.2%
-10.6%
-11.0%
-12.6%
-12.4%
-12.9%
-13.7%
-14.5%
-15.1%
-15.1%
-15.2%
-15.4%
-14.8%
VAR
287.7
252.9
307.1
259.9
288.7
253.2
230.3
258.8
258.1
256.3
256.2
258.9
286.0
CV
1.7
1.5
1.6
1.3
1.4
1.2
1.1
1.1
1.1
1.7
0.9
1.4
1.1
Table A3: Descriptive statistics cross-country heterogeneity for primary energy carrier dependency.
Key: N - number of observations, Min - minimum, Max - maximum, VAR
- Variance, CV - Coefficient of variation
B
Regional data for energy security assessment
for 2007
26
27
Crude oil
370.51 (88%)
1421.49 (100%)
313.74 (100%)
323.35 (97%)
335.33 (0%)
2.75 (0%)
529.29 (100%)
3522.49 (98%)
4684.43 (98%)
3645.08 (7%)
918.18 (100%)
344.35 (83%)
145.15 (100%)
4262.47 (94%)
1.30 (0%)
246.30 (97%)
0.00 (0%)
2065.32 (100%)
929.20 (100%)
525.45 (100%)
566.63 (62%)
275.80 (100%)
0.00 (0%)
2525.16 (100%)
774.20 (100%)
Coal
122.18 (100%)
190.45 (92%)
347.37 (35%)
871.83 (0%)
195.90 (100%)
167.16 (2%)
279.67 (60%)
580.43 (92%)
3745.72 (35%)
1628.09 (69%)
372.30 (4%)
140.54 (49%)
96.97 (60%)
678.41 (99%)
4.34 (91%)
10.59 (84%)
3.57 (100%)
381.82 (100%)
2502.47 (0%)
120.47 (100%)
380.65 (26%)
169.81 (86%)
64.34 (19%)
885.52 (69%)
108.77 (94%)
Natural gas
335.02 (72%)
670.03 (94%)
124.78 (88%)
339.91 (80%)
172.63 (0%)
38.60 (87%)
169.01 (92%)
1661.19 (92%)
3257.49 (73%)
3456.20 (20%)
139.25 (99%)
453.52 (78%)
179.96 (92%)
2942.84 (85%)
56.57 (97%)
121.00 (100%)
48.07 (100%)
1670.73 (0%)
592.77 (57%)
158.00 (99%)
527.95 (29%)
235.20 (87%)
38.26 (100%)
1359.63 (97%)
54.32 (64%)
Nuclear energy
0.00 (0%)
526.21 (0%)
160.62 (0%)
286.57 (0%)
0.00 (0%)
0.00 (0%)
255.57 (0%)
4797.92 (0%)
1533.08 (0%)
687.70 (0%)
0.00 (0%)
160.68 (0%)
0.00 (0%)
0.00 (0%)
0.00 (0%)
109.22 (0%)
0.00 (0%)
45.83 (0%)
0.00 (0%)
0.00 (0%)
84.11 (0%)
169.34 (0%)
62.14 (0%)
601.23 (0%)
730.70 (0%)
Renewables
372.50 (0%)
89.87 (0%)
43.13 (0%)
96.47 (0%)
150.84 (0%)
25.18 (0%)
365.75 (0%)
776.39 (0%)
1120.51 (0%)
216.29 (0%)
72.64 (0%)
60.28 (0%)
19.46 (0%)
543.23 (0%)
59.33 (0%)
34.02 (0%)
5.42 (??%)
124.78 (0%)
224.71 (0%)
193.44 (0%)
197.66 (0%)
42.54 (0%)
31.32 (0%)
430.17 (0%)
657.12 (0%)
Total
1200.22 (57%)
2898.05 (77%)
989.64 (55%)
1918.13 (31%)
854.70 (23%)
233.69 (16%)
1599.29 (53%)
11338.42 (49%)
14341.23 (58%)
9633.36 (22%)
1502.36 (71%)
1159.37 (61%)
441.54 (83%)
8426.94 (85%)
121.54 (48%)
521.12 (71%)
57.06 (91%) 52.06
4288.48 (57%)
4249.15 (30%)
997.36 (80%)
1757.01 (35%)
892.70 (71%)
196.06 (26%)
5801.71 (77%)
2325.12 (39%)
794.18
422.99
234.39
170.16
77.13
46.84
1459.71
466.69
GDP
374.30
466.00
38.46
177.70
311.69
21.27
246.96
2603.38
3400.12
2886.28
309.08
137.16
262.52
2129.37
27.40
38.00
Table B1: Summary table direct primary energy consumption (in Petajoule), import shares (in brackets) and GDP (in millions
of US$) for 2007.
Austria
Belgium
Bulgaria
Czech Republic
Denmark
Estonia
Finland
France
Germany
Great Britain
Greece
Hungry
Ireland
Italy
Latvia
Lithuania
Luxembourg
Netherlands
Poland
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden
28
Crude oil
264.96
-360.19
-87.75
74.30
208.97
58.98
-41.28
930.85
630.97
524.12
108.21
53.92
308.35
-117.58
85.71
229.35
49.87
-859.52
103.22
53.29
35.03
-88.04
126.14
451.60
-102.19
Coal
268.99
326.76
-117.39
-172.05
133.74
-42.49
78.98
1240.80
1029.45
1323.03
221.71
122.23
169.00
1160.49
41.19
87.22
29.15
380.58
-297.02
129.00
62.55
5.65
23.99
828.28
230.24
Natural gas
63.36
-34.08
-22.80
29.17
60.00
6.54
77.44
704.25
648.79
434.61
198.05
24.13
76.85
293.35
25.68
22.78
-2.05
-453.08
109.53
77.23
6.52
-13.37
15.67
319.71
188.91
Nuclear energy
80.61
-109.61
-70.16
-48.54
44.28
6.37
-31.09
-638.61
7.85
199.06
50.58
-10.04
52.38
288.58
10.00
-18.88
13.25
104.93
78.92
43.68
16.20
-27.94
-11.68
76.14
-105.76
Renewables
-84.43
48.08
0.27
4.78
0.22
0.52
-137.10
129.24
95.34
287.80
33.20
14.57
23.61
107.98
-2.06
3.41
3.64
74.63
13.00
-11.20
4.30
-4.00
2.78
90.30
-187.69
Total
593.49
-129.04
-297.83
-112.35
447.22
29.92
-53.05
2366.52
2412.39
2768.63
611.75
204.82
630.20
1732.83
160.52
323.88
93.86
-752.46
7.64
292.00
124.60
-127.70
156.90
1766.03
23.50
IIR
86.03%
-5.80%
-54.59%
-19.16%
228.29%
80.53%
-6.21%
42.77%
29.00%
133.65%
57.10%
29.01%
171.43%
24.14%
273.58%
87.54%
181.75%
-30.75%
0.60%
36.41%
20.51%
-20.26%
312.40%
39.73%
2.58%
EIR
49.45%
-4.45%
-30.10%
-5.86%
52.33%
12.80%
-3.32%
20.87%
16.82%
28.74%
40.72%
17.67%
142.73%
20.56%
132.08%
62.15%
164.48%
-17.55%
0.18%
29.28%
7.09%
-14.30%
80.02%
30.44%
1.01%
Table B2: Summary table for carrier-specific and total net-imports of primary energy embodied in traded goods (in Petajoule),
energy imbalance ratio (EIR) and import imbalance ratio (IIR) for 2007. EIR - ratio of total net-imports of embodied energy
over total direct primary energy consumption; IIR - ratio of total net-imports of primary energy over direct net-imports of
primary energy.
Austria
Belgium
Bulgaria
Czech Republic
Denmark
Estonia
Finland
France
Germany
Great Britain
Greece
Hungry
Ireland
Italy
Latvia
Lithuania
Luxembourg
Netherlands
Poland
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden