247      ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015  |  247-277
ANALYSIS OF THE EFFECTS OF 
INTRODUCTION OF AN ADDITIONAL 
CARBON TAX ON THE SLOVENIAN 
ECONOMY CONSIDERING DIFFERENT 
FORMS OF RECYCLING
1
ALEKSANDAR KEŠELJEVIĆ
2
  Received: 29 April 2013
MATJAŽ KOMAN
3
 Accepted: 14 November 2014
ABSTRACT: This paper outlines some of the environmental and economic implications of an 
additional CO2 tax of EUR 15/tCO2 in Slovenia in the period 2012-2030 in order to deter-
mine whether it yield a double dividend. Authors analyze (using E3ME  model) different 
forms of revenue recycling by reducing the social security contributions of either the employ-
ers or the employees or by reducing the public deficit, in order to identify the optimal fiscal 
instrument for improving the environmental and economic welfare (double dividend). In this 
policy orientated paper authors argue that a reduction of employee social security contribu-
tions has more favourable effect than a reduction in employers' social security contributions. 
Keywords: green tax, environmental tax reform, double dividend, carbon tax, recycling, E3ME model
JEL Classification: E17, H23, Q50
1. INTRODUCTION – GREEN TAXES AND ENVIRONMENTAL TAX REFORM (ETR) 
The idea of a green tax dates back to Arthur C. Pigou (1920); hence, green tax is also 
referred to as a Pigouvian tax. It is based upon a fundamental principle that the polluters 
should pay a tax in the amount equal to the damages resulting from their impact on the 
environment (i.e. negative externalities). The costs are namely not incurred only by the 
company whose emissions pollute the environment; rather, the costs are sustained by the 
entire society. It is then the task of the government to impose the green tax to internalize 
the pollution costs as much as possible. In such case, the polluting industrial activity is 
reduced to a socially desirable level (Turner, 1994). 
Introduction of the green tax represents also an important development in the public 
finance reform since it involves also a reconsideration of the present tax system, aimed 
1 ACKNOWLEDGEMENTS: Kešeljević's and Koman's research in this paper was  supported by grant No V5-
1004 of the Slovenian Research Agency  (ARRS) and  the Institute of Macroeconomic Analysis and Development 
(UMAR). The authors would like to thank  Katarina Ivas from Insitute of Macroeconomic Analysis and Develop-
ment and people from Cambridge Econometrics, specially Eva Alexandri, for helpful comments  and suggestions. 
2 University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: saso.keseljevic@ef.uni-lj.si
3 University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: matjaz.koman@ef.uni-lj.si

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 248
predominantly at taxing labour and capital. The environmental tax reform (ETR) argues 
in favour of green taxes in a revenue-neutral fashion to reduce other distortionary levies. 
Instead of taxing “good things” like labour, income and capital, the government should 
start taxing “bad things” like pollution, use of natural resources etc. (Bousqet, 2000; Pat-
uelli et al., 2005). The main goal of an environmental tax reform is therefore an improve-
ment in both environmental (first dividend) and economical aspects (second dividend). 
Environmental dividend involves reduction in emissions and economic dividend stems 
from lower costs, improved competitiveness, and higher employment. Therefore, the term 
“double dividend” is increasingly used to describe the environmental tax reform (Glomm 
et al., 2008; Ekins, 2009). 
Experience from European countries has shown, that effects of a comprehensive ETR have 
been positive in most cases (Sweden, Denmark, Netherlands, UK, Finland, Norway, Ger-
many). Therefore, the environmental tax reforms (ETR) have become a relevant instru-
ment in the economic policies of the developed world in recent years. 
Our primary goal is to determine the effect of an additional carbon tax (EUR 15 per ton of 
CO2 i.e. EUR 55 per ton of carbon) in the period 2012–2030 on the Slovenian economy, 
in order to determine whether an additional carbon tax would indeed yield a double divi-
dend. We shall examine the possibilities of different recycling options either through re-
duction of budget deficit or reduction of employer/employee social security contributions, 
in the form of different scenarios (using E3ME model) in order to identify the optimal 
fiscal instrument for improving the environmental (first dividend) and economic welfare 
(second dividend). 
The article is structured as follows. In section two the concept of double dividend is intro-
duced. In section three we present the E3ME model and the impact of green taxes within 
the model.  Results regarding the environmental and economic implications of an envi-
ronmental tax reform are presented in section four. Finally, the last section deals with the 
conclusions and policy implications derived from the contents of the paper.
2. A DOUBLE DIVIDEND 
The two central dilemmas regarding the green tax have to do with regressiveness and loss 
of competitiveness. Many authors have argued that incidence of green taxes falls largely 
on the low-income class (Roed, 2006; West, Williams, 2004; Labandeira, Labeaga, 1999; 
Tiezzi, 2001; Clinch et al., 2006). Negative effect on cost competitiveness of the economy 
will be greater when (1) elasticity of demand for a certain good is relatively high; (2) there 
is strong competition in the industry; (3) a particular sector is highly energy-intensive; 
(4) ecotax is introduced in a small number of  countries; and (5) there is no option to 
substitute the polluting activity with an environmentally friendlier technology (Kosonen, 
Nicodème, 2009; Clinch et al., 2006; Patuelli et al., 2005; Baron, 1997; Envoldsen et al., 
2009). Thus, if the government introduces ETR without recycling the tax revenue within 
the system, an economic downturn would likely occur.

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 249
Recycling in this case refers to targeted use of green tax revenue, especially for reducing the 
taxation of labour and social security contributions. Besides a reduction of social security 
contributions or personal income taxes, other forms of financial recycling are also possible 
by transfers to households/industries for greater energy efficiency
4
 or interventions in cor-
porate income taxes and value added tax. In case of total recycling, the total tax burden re-
mains unchanged (fiscal neutrality) (Speck, Jilkova, 2009; Ludewig et al., 2010; OECD, 2007; 
Hoerner, Bosquet, 2001; Clinch et al., 2006; Patuelli et al., 2005; Hansen, Holger, 2000).
We expect an environmental tax reform to lead to an improvement from environmen-
tal aspects, e.g. owing to lower carbon dioxide emissions, as well as to improve the cost 
competitiveness of the economy as a result of lower labour costs and higher technological 
efficiency of businesses. Hence, economic growth and employment will actually increase 
(Benoit, 2000; Hoerner et al., 2001; Patuelli et al., 2005; Tuladhar, Wilcoxen, 1999). Not 
surprisingly, the European countries with the highest tax on labour were the first to im-
plement the environmental tax reform and look for double dividend (Finland, Sweden, 
Denmark, Netherlands, Germany, and Norway). 
The first (environmental) dividend of the double dividend hypothesis is widely accepted. 
Johansson (2000) argues that in Sweden the CO2 emissions were 15% lower than they 
would have been in the absence of the green taxes. Berkhout and Linderhof (2001) point 
out that in the Netherlands, the price of electricity and fuel for domestic use rose dramati-
cally as a result of the green tax and ex-post studies show that consumers now use 15% 
less electricity and 5–10% less fuel. Baron (1997) pointed out that in Denmark recycling 
of tax revenues through investment in energy efficiency has led to about 4.7% reduction 
in CO2 emissions. Labandeira et al. (2004) show that in Spain a tax on CO2 emissions has 
resulted in environmental improvement. Ludewig et al. (2010) demonstrate that use of all 
motor fuels in Germany was decreasing in the period from 1995 to 2006 by an average 
rate of 0.3 percent per year. At the same time, use of public transport was rising. Based on 
an analysis of 139 simulation models, Bosquet (2000) found that a considerable drop in 
carbon dioxide emissions is among the expected effects of a green tax reform in the short 
to medium run. 
The second (economic) dividend depends mainly on the structure of the economy (e.g. 
labour market, pre-existing tax structure), time lag and explicit model assumptions. Since 
the present tax system creates significant disincentives to work and hire, virtually any 
environmental policy can compound these existing distortions (Carraro et al., 1996; Mor-
genstern, 1995; Tuladhar, Wilcoxen, 1999; Schöb, 2003). Ludewig et al. (2010) find that 
250,000 new jobs were created in Germany in this way. Experience from Denmark (Hans-
en, Holger, 2000) and Spain (Manresa, Ferran 2005) is similar. However, many authors 
argue that the “double dividend” theory oversimplifies a number of points and that certain 
conditions have to be fulfilled for a double dividend. 
4 Alternative recycling method are: (1) improvements in the energy efficiency of the building stock, (2) grants 
for improving energy efficiency in buildings, (3) recycling into local environmental projects to foster communi-
ty acceptance of ETR, (4) recycling to public transport, (5) subsidising renewable energy and combined heat and 
power production, (6) subsidising ‘cleaner’ technology in industry, (7) subsidising R&D (Clinch et al., 2006). 

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 250
Firstly, ETR is expected to improve the quality of the environment and to reduce the dis-
tortions of existing taxes. This view has been questioned in several papers (Goulder, 1995; 
Benoit, 2000; De Mooij, 1999; Li, Ren, 2012). The basic point is that the double divi-
dend hypothesis ignores the interaction between environmental taxes and pre-existing tax 
structure. If the initial tax system is suboptimal then ETR can generate a significant double 
dividend. Similarly Fraser and Waschik (2013) using a CGE model to empirically exam-
ine the double dividend hypothesis provide support for the existence of a strong double 
dividend when revenue is recycled through reductions especially in consumption taxes. 
Secondly, the outcome depends very much on labour market conditions in the country 
(Clinch et al., 2006; Carraro et al., 1996; Schöb, 2003; Koskela and Schob, 1999; Holmlund 
and Kolm, 2000; Albrecht, 2006; Ciaschini et al., 2012). If there are labour rigidities (as 
in some countries of Europe), then there will be an employment dividend resulting from 
the recycled carbon tax revenue. But in the long run, such rigidities become less relevant. 
Thirdly, green taxes represent, as a rule, a relatively small share of overall tax revenue of 
any given country
5
. Hence, a dramatic increase would be required to offset the lower per-
sonal income tax revenue. Thus, if green taxes are set high enough to achieve meaningful 
reductions in emissions, they may cause significant distortions in the tax system. Policy 
makers will then be forced to trade off cleaner environment against other policy targets 
(Coxhead, 2000).
Fourthly, Carraro et al (1996) find that the unions’ negotiating strength affects the possi-
bility of gains in employment. In the short run the employment may increase due to lower 
taxes; however, in the long run, net wages completely absorb the tax change, thus bringing 
employment back to its baseline value. Many authors argue that the effects of a green tax 
reform are doubtful in the long run.
Nevertheless, while the second dividend may be in doubt, the first dividend remains a 
powerful argument for the introduction of ETR. Obviously, a strong double dividend oc-
curs under rather “constrained” circumstances. We do not go more into the details since 
the rise and fall of the double dividend hypothesis and conditions for it has been discussed 
at length elsewhere (Bovenberg and Goulder 1997; Parry and Oates, 1998; Goulder, 1995; 
Bosquet, 2000; Fraser and Waschik, 2013). All authors agree that validity of the double-
dividend hypothesis cannot be settled as a general matter. In other words, each reform 
must be evaluated on its own merits by keeping in mind the characteristics of respective 
countries and the explicit model assumptions.
5 In most EU countries, revenue from green taxes is between 2% and 3% of GDP. There are only four EU coun-
tries where such share in lower than 2% (1.9% in Slovakia, 1.9% in Lithuania, 1.6% in Spain, 1.8% in France), 
and only three countries where this share exceeds 3.5% of GDP (4% in Denmark, 4% in the Netherlands, 3.6% 
in Slovenia). Green taxes represent the largest share of total tax revenue in Bulgaria (10.7%), the Netherlands 
(10.3%), and Slovenia (9.6%). The lowest contribution of green taxes to overall tax revenue was observed in 
France (4.2%), Belgium (4.7%), and Spain (5.2%).  Slovenia is considerably above the EU27 average (6.2%) with 
its 9.6-percent share of green tax revenue in overall tax revenue (European Commission, 2012). 

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 251
3. THE MODEL 
There are two different methodological approaches to modelling the relation between the 
environment and the rest of the economy. The first approach is based on highly precise 
modelling of a certain sector; as a rule, however, such models do not yield the best expla-
nations as to the interaction between the sector at hand and the economy as a whole. The 
other approach is based on structural macroeconomic models. A key advantage of these 
models, each of them is based on certain underlying assumptions, is that they allow a 
fairly accurate prediction of macroeconomic results in case of different scenarios. These 
models provide a better understanding of the economic consequences of environmental 
measures as they allow studying the economic processes that lead to final results. The 
downside of these models is that each sector is modelled at the aggregated level
6
. 
Our analysis is based on the latter approach. We employed the E3ME
7
 model, widely used 
among European researchers in recent years. This is a dynamic simulation econometric 
model intended for analysis of the effects of E3 policies (economy, energy, environment), es-
pecially those pertaining to environmental taxes and regulation. The model allows examin-
ing the short-term (annual) and medium-term economic effects, as well as long-term effects 
of E3 policies for a period of 20 years. Hence, E3ME combines the features of short-term 
and medium-term sector models estimated using econometric methods with the features of 
computational general equilibrium models. The E3ME model includes 42 product/industry 
sectors (OECD classification), with energy sector further disaggregated to include energy-
environment interaction and 16 service sectors. It is intended for analysis of macroeconomic 
effects (with emphasis on environmental components) of environmental economic policies, 
especially from the aspect of environmental taxation and regulation, for 33 European coun-
tries (EU27, Norway, Switzerland, Iceland, Croatia, Turkey, and Macedonia) as a whole. It 
also allows analysis of environmental effects in each country
8
.
The structure of E3ME is based on the System of National Accounts (ESA 95), with ad-
ditional links to demand for energy and environmental emissions. The model includes a 
total of 33 sets of econometrically estimated equations which also include components 
of the GDP (consumption, investment, international trade), prices, demand for energy, 
and demand for raw materials. Each set of equations is broken down by countries and 
by sectors. E3ME also allows analyzing the effects of particular scenarios as measured by 
numerous economic, energy, and environmental indicators. The model is based on the 
data for the period from 1970 to 2010 and annual projections until the year 2050. The 
main sources of data include Eurostat, AMECO DC ECFIN database, and IEA; this data 
set is further complemented by OECD STAN and other databases. Any gaps in the data 
are estimated using adjusted software algorithms. For a detailed description of the E3ME 
model, see E3ME Manual (2012). 
6 For a detailed description of methodological approaches in modelling the relations between the environ-
ment and the economy, see Ščasný et al. (2009).
7 The model was developed and is maintained by the company Cambridge Econometrics.
8 See E3ME Manual (2012) for more detailed description. 

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 252
3.1.  EFFECTS OF ECOLOGICAL TAXATION (GREEN TAX) IN THE 
  E3ME MODEL 
One of the purposes of the E3ME model is to provide consistent and coherent analysis of 
fiscal policy and its relation to greenhouse gas emissions. The E3ME model allows exam-
ining how carbon and energy taxes affect the reduction of environmental emissions, as 
well as how other taxation and economic policies affect reduction of emissions. 
The effect of a taxing carbon dioxide emissions (and energy consumption) in the E3ME 
model on prices and wages is based on two key assumptions. The first assumption is that the 
effect of tax is transmitted through the price of fuel and any use of subsequent tax revenue to 
reduce other taxes. Other effects are not modelled. The second assumption is that import of 
fuels and domestic production are taxed in proportion to the CO2 emission rate and energy 
value of the fuel, while fuel exports are not taxed. It is assumed that this tax is paid by the 
fuel producers and importers. This tax is then levied on the final users through higher fuel 
prices. Another assumption is that the industry will transmit these additional fuel costs on 
its buyers in the form of higher prices of commodities (goods and services). An increase in 
the final price is therefore a result of direct and indirect effect of tax on a particular good or 
service.  If tax revenue is used to reduce the rates of taxes levied on the employers, this will 
result in a decrease of labour costs and, in turn, a drop in production costs.  These changes, 
too, will then be transmitted forward within the E3ME model (E3ME Manual, 2012).  
Net effect of tax on prices of products and imports will be transmitted to consumer prices, 
resulting in a change in the consumption of goods and services. Such change will depend 
on individual ecotax and the price elasticity of the affected commodities. Higher prices 
of goods and services will lead to demands for higher wages. Econometric studies have 
confirmed that in the long run, entire tax is levied on the consumers. This fact is integrated 
into the E3ME model as a part of its long-term solution. 
In the E3ME model, ecotaxes indirectly influence (through direct effect on prices and 
wages) the macroeconomic parameters such as fuel consumption, production, employ-
ment in particular sectors etc.). Namely, a change in the price of fuels resulting from eco-
tax will, depending on the elasticity of substitution, lead to a change in fuel consumption. 
Increase of fuel prices due to higher taxes will cause changes in consumer prices, which 
will be reflected in substitution in consumer expenditure, change of export activity, and 
change in the relation between domestic production and imports. These changes will in 
turn affect, via feedback loop, the use of various types of fuel. A reduction in labour costs 
resulting from “recycling” of tax revenue will initially have a direct positive effect on em-
ployment, followed by an indirect effect through relative price competitiveness thereon 
as more commodities (goods and services) are produced in labour intensive industries.    
 
4. RESULTS OF THE MODEL 
Below we present the results of the introduction of the additional carbon tax. We firstly 
assume that all revenue generated from ecotax is allocated for reduction of the budget 

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 253
deficit or increase of the budget surplus. In subsequent analyses, ecotaxes will be recycled 
in various ways, e.g. they will be used to reduce the taxes levied on labour costs.
The analysis will be based in section 4.2. on a comparison to a base projection (baseline 
scenario), and in section 4.3. on a comparison to a budget recycling projection. Results 
will be presented in the form of a deviation from the base projection and the budget recy-
cling projection. Therefore, we continue by presenting the assumption underlying the base 
projection, and the way in which this projection was generated.
4.1.  DESCRIPTION OF THE BASE PROJECTION (BASELINE SCENARIO) AND 
UNDERLYING ASSUMPTIONS AND THE ESTIMATION METODOLOGY 
TOGETHER WITH PARAMETER RESULTS
It is important that the baseline projection (baseline scenario) in the framework of the 
E3ME model is consistent with the forecasts used in other analyses. The underlying as-
sumption of the baseline projection was that the E3ME projection was consistent with the 
slightly modified projection of the European commission (modified projection PRIMES 
BASELINE 2009). PRIMES BASELINE 2009 forecasts are also presented in Table A1 in 
the Appendix.
Following is a description of the key stages in modelling of the base projection. Inputs for 
the base projection include historical data (data on economic indicators, energy, and the 
environment, obtained from different sources (Eurostat, IEA etc.), estimates of param-
eters for endogenous variables, and fundamental assumptions. 
Historical data on economic indicators for Slovenia (employment, output, consumption, 
exports etc.) is used up to and including 2010. The indicators were calculated from the 
data published by Eurostat in February 2012.  Historical data on energy components (en-
ergy consumption by types of fuel etc.) and environmental components is derived from 
the World Energy Outlook for the period up to 2009.
Endogenous variables are determined using the functions estimated based on historical 
data. There are around 33 variables for which stochastic functions are estimated. However 
these variables may well be disaggregated in two dimensions (e.g. there are 19 fuel users 
and 33 countries) so we will not provide the specification of each variable. Below we first 
describe the general procedure how these stochastic functions are estimated and then 
show one example of such function and its parameters for Slovenia.
The functional form of the equations and the parameters are based on the  cointegration 
and error-correction methodology (Engle and Granger, 1987,  and Hendry et al., 1984). 
The process involves two stages. The first-stage is a levels relationship, where an attempt 
is made to identify the existence of a cointegrating relationship between the chosen vari-
ables, selected on the basis of economic theory and a priori reasoning. For example the 
aggregate energy demand (FRO) is specified as follows:

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 254
FR0
i,j,t 
=
 
a
i,j,0 
+
 
a
i,j,1
FRY
i,j,t 
+
 
a
i,j,2
PREN
i,j,t 
+
 
a
i,j,3
FRTD
i,j,t 
+ a
i,j,4
ZRDM
t 
+ a
i,j,5ZRDTt +   
                       
+ a
i,j,6
FRK
i,j,t + 
u
i,j,t
where FRY is economic output of energy users i in region j, PREN is average fuel price 
(across all fuels) deflated by unit cost in region j, FRTD is R&D expenditure by energy user 
i in region j, ZRDM is EU investment of R&D in machinery, ZRDT is EU investment of 
R&D in transport, and FRK is  investment by energy user i in region j
If a cointegrating relationship exists, then the second stage regression, known as the error-
correction representation, is implemented. It involves a dynamic, first-difference, regres-
sion of all the variables from the first stage, along with lags of the dependent variable, 
lagged differences of the exogenous variables, and the error-correction term (the lagged 
residual from the first stage regression). Due to limitations of data size, however, only 
one lag of each variable is included in the second-stage. For example in case of aggregate 
energy demand the error correction equation is specified as:
∆FR0
i,j,t 
= b
i,j,0 
+ b
i,j,1∆
FRY
i,j,t 
+ b
i,j,2∆
PREN
j,t 
+ b
i,j,3
DFRTD
i,j,t 
+ b
i,j,4
∆ZRDM
t 
+ b
i,j,5
∆ZRDT
t  
 
+ b
i,j,6
∆FRK
i,j,t 
+ b
i,j,7
∆FR0
i,j,t-1 
+ g
i,j,
ECM
i,j,t-1 ,
where ∆ is difference and ECM is error correction. 
Stationarity tests on the residual from the levels equation are performed to check whether 
a cointegrating set is obtained. Due to the size of the model, the equations are estimated 
individually rather than through a cointegrating V AR. For both regressions, the estimation 
technique used is instrumental variables, principally because of the simultaneous nature 
of many of the relationships (for example wage, employment and price determination). 
E3ME’s parameter estimate is carried out using a customised set of software routines 
based in the Ox programming language (Doornik, 2007). The main advantage of using 
this approach is that parameters for all sectors and countries may be estimated using an 
automated approach. 
The estimation produces a full set of standard econometric diagnostics, including stand-
ard errors and tests for endogeneity. However all the estimation procedures and test are 
carried out by Cambridge Econometrics, the developer of the software
9
. 
In Table A2 in appendix we provide a summary of the model equations, giving an over-
view of which variables are used, units of measurement and functional form. A full list of 
the variables included in E3ME model is available on request. In Appendix 1 we also  pre-
sent in more detail the agregate demand for energy function and the estimated parameters 
for Slovenia. The other functions and parameters for Slovenia are available upon request. 
9 A list of equation results can be made available on request. For each equation, the following information 
will be given: summary of results, full list of parameter results, full list of standard deviations.

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 255
The gaps in any of the E3ME time series was filled by software that was developed by the 
Cambridge Econometrics. This software uses growth rates and shares between sectors and 
variables to estimate missing data points, both in cases of interpolation and extrapolation. 
More precisely, “ The most straightforward case is when the growth rates of a variable are 
known and so the level can be estimated from these growth rates, as long as the initial level is 
known. Sharing is used when the time-series data of an aggregation of sectors are available 
but the individual time series is not. In this case, the sectoral time series can be calculated 
by sharing the total, using either actual or estimated shares. In the case of extrapolation, it is 
often the case that aggregate data for a number of sectors are available, although the sectoral 
disaggregation at the E3ME level is not; for example, government expenditure is a good 
proxy for the total growth in education, health and defence. A special procedure has been 
put in place to estimate the growth in more disaggregated sectors so that the sum of these 
matches the known total, while the individual sectoral growth follows the characteristics of 
each sector. Interpolation is used when no external source is available, to estimate the path 
interval, at the beginning and end of which data are available” . (E3ME, 2014, page 34) 
Basic assumptions are derived from various sources. The sources are presented in Table 
A3 in the Appendix. For Slovenia, the values of these assumptions for the period 2010–
2013 are presented in Table A4 in the Appendix. In the same table values of assumptions 
for particular commodities (e.g. energy prices, fuel prices etc.) are also presented. The 
baseline scenario is therefore based on all government measures implemented until mid 
2010. For example, the CO2 price is determined on the measures introduced by the Slove-
nian government by mid 2010.
The process of ensuring compliance of the base projection in the E3ME model involves 
three stages. This is in fact a calibration process. The first stage in reconciling the E3ME 
projections with the published and slightly modified forecast PRIMES BASELINE 2009 
(EU Energy trends to 2030, Baseline scenario 2009, European Commission, 2010). It 
includes ensuring consistency and transformation of the data into a suitable form. This 
means that different model dimensions have to be brought into line (geographic coverage, 
temporal aspect, sector coverage etc.). Transformed data are then saved in a separate file. 
In the next stage, the model is resolved in such way that model results match the slightly 
modified PRIMES BASELINE 2009 forecasts saved in a separate file. This is the calibrated 
forecasting process. In this forecast, the model solves its equations and compares the dif-
ferences in results with the data saved in the database. Model results are substituted with 
values from the forecast database. Differences between results and forecasts are saved in 
a separate database called the “residual” database.  In the last stage, the model is solved 
again using the “residual” database as well. This is the so-called endogenous baseline pro-
jection.  According the theory, the final result should be the same as in the case of calibrat-
ed forecast. In practice, the match is not 100-percent (see, E3ME manual, pages 40–41). 
In the E3ME model framework, the calibration process with modified PRIMES BASELINE 
2009 forecasts is carried out based on the trends (growth rates) rather than based on levels. 
This is because historical data in the E3ME model are newer that the data from the modified 
PRIMES BASELINE 2009. Calibrations for PRIMES BASELINE 2009 forecasts are made for 

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 256
the key economic variables and demand for energy (variables FRO, FRO1, FRO2 ... FRO12) 
and data on emissions (variables GHG, FCO2 etc.).   However, since PRIMES BASELINE 
2009 forecasts are based on the year 2010 and they do not include the most recent changes 
in the economic environment (the economic crisis), short-term calibration for macroeco-
nomic variables is conducted based on AMECO short-term forecasts. Therefore, the base-
line scenario is made based on the modified PRIMES BASELINE 2009 forecasts.
The key advantage of the endogenous baseline projection is that it allows us to analyse 
different scenarios in order to find out how the results change relative to the baseline 
scenario. There are two baseline endogenous projections: SI endogenous baseline projec-
tion and EU endogenous baseline projection.  For the SI endogenous baseline projection, 
calibration is only carried out for Slovenia while other European regions are treated as 
exogenous. This projection is used in analysis of scenarios that only affect Slovenia (e.g. a 
change in domestic tax rate). EU endogenous baseline projection involves simultaneously 
solving the E3ME model for the entire Europe. This projection is used for scenarios that 
will affect the entire Europe (e.g. a change in oil prices). If this solution is used, results for 
Slovenia will also include secondary effects from other European regions, brought about 
through international trade.
Since the introduction of the additional carbon tax in Slovenia is only affecting the Slo-
venian economy, SI endogenous projection will be used. The remaining part of Europe is 
treated as exogenous
10
. 
It is important to stress, that all scenarios that will be presented
11
 are based on (1) his-
torical data up to and including the year 2009 (energy and environmental components) 
or the year 2010 (economic components); (2) on government measures implemented by 
mid 2010; (3) and on long-term and short-term trends energy and environmental com-
ponents, that are based on the European Commission projections from 2009 (PRIMES 
BASELINE 2009). Long-term trends for macroeconomic components are also based on 
European Commission projections from 2009 (PRIMES BASELINE 2009) while short-
term macroeconomic components are based on the AMECO projections. This means that 
the effects of the economic crisis are only partially included and, as a result, the below 
results should be used with caution. 
4.2. ANALYSIS OF INTRODUCTION OF AN ADDITIONAL CARBON TAX ON 
THE SLOVENIAN ECONOMY
It is assumed within the E3ME model that payment of carbon tax (tax on carbon dioxide) 
is levied on the users of fuels based on their emissions; however, only sectors outside ETS 
are taxed in order to avoid double taxation. The cost, or burden, of the tax is then shifted 
to the consumers through higher fuel prices. 
10 We have also introduced the additional carbon tax in Slovenia by using EU endogenous baseline projec-
tion. The results were very similar. 
11 Values of particular variables for all scenarios to be used herein are presented in Table A5 in the appendix. 

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 257
In consequence, this means that we can expect the prices to rise while demand for fuel 
drops. It is assumed that higher prices will lead to a drop in real income. We can expect 
household consumption expenditure to decrease, which will in turn decrease demand 
and cause a drop in gross domestic product. As we assumed this change would not affect 
the European economy, we expect this will result in a drop of export competitiveness of 
the Slovenian economy due to higher prices, which will lead to a further decrease in GDP .
According to economic theory, the amount of carbon tax should be equal to the social 
cost incurred as a result of carbon pollution. Y ohe et al. (2007) reviewed the estimates and 
found that costs estimates are highly unpredictable as they range from USD 1 per ton of 
carbon (tC) up to USD 1,500 per ton of carbon (tC). Average estimate of social cost of 
pollution with carbon dioxide for 2005 was USD 43/tC, with a standard deviation of USD 
83/tC.  The authors found that these costs rise at a rate of 2 to 4 percent per year. Assum-
ing 4-percent annual growth since 2005, carbon pollution cost in 2012 would amount to 
an average of USD 55/tC or EUR 42/tC (i.e. EUR 11.5/tCO2. We set the amount of extra 
carbon tax to EUR 15/tCO2 (i.e. EUR 55/tC)
12
. 
In the article we compare two scenarios: baseline scenario in which no extra carbon tax 
is introduced and the projection of an introduction of an additional annual carbon tax 
in the amount of EUR 15 per ton of CO2 (EUR 15 per ton of carbon) for sectors beyond 
ETS, where all ecotax is recycled into the government budget. Comparison between the 
two projections is made for some key economic (household consumption expenditure, 
exports, gross domestic product, total manufacturing output, employment), energy (aver-
age fuel prices, demand for energy), and environmental variables (greenhouse emissions) 
which are presented in detail below.
Average fuel prices including tax (PJRT
13
) change the most in the first year following the in-
troduction of the carbon tax in the amount of EUR 15/tCO2 (EUR 55/tC) (2012) when they 
rise by 3.67% relative to the baseline scenario in which no extra carbon tax is introduced. 
After the initial price hike, the price reaches a steady state at a higher figure which is main-
tained throughout the examined period. The difference in the average fuel price between the 
baseline scenario and projection that assumes an additional carbon tax of EUR 15/tCO2 (or 
EUR 55/tC) is approximately 3.5% throughout the period at hand (until 2030). 
As expected, the introduction of an extra carbon tax of EUR 15/tCO2 (EUR 55/tC) drives 
up the average prices of fuel, which in turn causes a decrease in demand for fuels for en-
ergy production (FRO
14
). This drop relative to the baseline scenario is relatively the largest 
in the initial period, after which the decrease in demand for energy is steadied or slowed 
down. In 2013, for example, demand for energy resulting from the introduction of the car-
bon tax was projected to be lower by 0.83% compared to the baseline scenario; in 2020 by 
12 Determination of the size of the ecotax has been aligned with the Institute of Macroeconomic Analysis and 
Development (UMAR). We have also used other numbers for ecotax, but we do not report them in the article.
13 PJRT = Average fuel price including tax (in EUR/toe). The model assumes 12 different fuel consumers. 
14 FRO = Total demand for energy is in E3ME model measured in thousand tons toe. Model assumes 12 dif-
ferent fuel consumers.

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 258
1.64%; and in 2025 by 1.9%. Initial increase in prices and a considerable drop in demand 
relative to the baseline scenario are followed by a higher and steady level of fuel prices and 
accordingly lower demand for energy throughout the period of examination.
Household consumption expenditure (RSC
15
) is one of the most important macroeco-
nomic aggregates, since it takes the largest share of GDP structure. Introduction of extra 
annual carbon tax of EUR 15/tCO2 (EUR 55/tC) would lead to the highest relative drop 
of household consumption expenditure in 2013 when the decrease amounts to 0.45% rela-
tive to the baseline scenario with no introduction of carbon tax. In principle, higher av-
erage prices of fuel lead to a decrease in real income which in turn decreases household 
consumption expenditure. This would result in a drop in aggregate demand and cause 
a decrease in gross domestic product. After 2013, the difference relative to the baseline 
scenario gradually decreases and by 2020, for example, consumption is only 0.27% lower 
compared to the baseline scenario. As expected, the difference between the two scenarios 
is the largest at the beginning of the period; after 2013, it is gradually decreasing. Moreo-
ver, the data shows a relatively low effect of the introduction of the carbon tax on the 
change in consumption. The reasons can be found in the time lag as the consumers require 
some time to adjust their behaviour and consumption pattern.
If the extra annual carbon tax in the amount of EUR 15/tCO2 (EUR 55/tC) is introduced, 
exports (RSX
16
) will decrease relative to the baseline scenario in which no carbon tax is 
introduced in the short run (until 2017), and increase after 2018. Such development is ex-
pected as we assumed the change would not affect the European economy. Higher prices 
expectedly hinder the export competitiveness of the Slovenian economy; however, the ex-
port sector’s agility and dynamic character in terms of development of new technological 
solutions and updates will allow it to neutralize relatively quickly such loss of competitive-
ness. It should also be noted that changes in exports relative to the baseline scenario are 
very small (up to a maximum of 0.009%), which points to a relatively low impact of the 
carbon tax on Slovenian exports.
Introduction of extra annual carbon tax in the amount of EUR 15/tCO2 (EUR 55/tC) 
would lead to the highest drop of Slovenia’s GDP (RGDP
17
) in 2013 when the decrease 
would amount to 0.3% relative to the baseline scenario with no introduction of carbon tax. 
This is consistent with our expectations. It has been shown in our previous analysis that 
higher fuel prices lead to a decrease of real income. As a result, household consumption 
expenditure will decrease, which will in turn decrease demand and cause a drop in gross 
domestic product. As we assumed this change would not affect the European economy, 
higher prices would also result in a drop of export competitiveness of the Slovenian econ-
omy, which would lead to a further decrease in GDP . Moreover, the data shows a relatively 
low effect of the introduction of the said tax on the change in GDP . After 2013, the differ-
ence between the two scenarios gradually decreases and by 2020, for example, GDP is only 
15 RSC = Household consumption expenditure is in E3ME model measured in EUR million. The model as-
sumes 43 different types of expenditure.
16 RSX = Exports are measured in E3ME model in million euro.
17 RGDP = Gross domestic product is in E3ME model measured by the expenditure method in current mar-
ket prices in millions of euro.  

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 259
0.12% lower in case of introduction of the carbon tax compared to the baseline scenario. 
This conforms to our expectations and the theoretical findings as economic agents require 
some time to adjust to the new circumstances. Businesses need time to implement techno-
logical improvements and updates, and consumers need time to adjust their consumption 
behaviour and patterns. 
We are also interested in the effect of an extra yearly carbon tax of EUR 15/tCO2 (EUR 
55/tC) on manufacturing output (QR
18
). The highest drop relative to the baseline sce-
nario would be in 2015. In that year, the difference would amount to 0.32%. Here too, it 
is evident that introduction of carbon tax in the amount of EUR 15/tCO2 (or EUR 55/
tC) has a relatively small effect on production. The difference between the two scenarios 
is, expectedly, the highest at the start of the period. After 2013, this difference is gradu-
ally decreasing so that the deviation from the baseline scenario in 2015 is no more than 
0.01%. Technological and organizational updates allowed the enterprises to adapt to the 
new conditions after a certain period of time. According to the projection, the latter effect 
prevails in the long run, after 2027.
Employment (YRE
19
) shows a similar dynamics as manufacturing output. Employment is 
gradually decreasing relative to the baseline scenario. The highest drop in comparison to 
the baseline scenario can be seen in 2016 when it amounts to 0.36%. There are hardly any 
differences between the two scenarios at the end of the period. The effect of an additional 
carbon tax of EUR 15/tCO2 (or EUR 55/tC) on employment appears to be relatively low, 
similarly to the effect on GDP and manufacturing output.
As expected, the introduction of an extra carbon tax of EUR 15/tCO2 (EUR 55/tC) grad-
ually decreases greenhouse gas emissions (RGHG
20
) in CO2 equivalents. This includes 
emissions of CO2, CH4, N2O, HFCs, PFCs and SF6. For example, the highest drop in 
emissions relative to the baseline scenario is seen in 2012 (by 0.6%) and 2013 (by an extra 
0.5%) to –1.2%. The decrease in emissions in comparison to the baseline scenario is stead-
ied at approximately 2% after 2020. 
4.3.  ANALYSIS OF DIFFERENT FORMS OF REVENUE RECYCLING IN CASE 
OF EXTRA CARBON TAX IN THE SLOVENIAN ECONOMY
Introduction of an extra annual carbon tax of EUR 15/tCO2 (EUR 55/tC) on an annual 
basis for the period 2012–2030 would result in additional annual tax revenue ranging 
from a minimum amount of EUR 144.6 million in year 2012 to a maximum amount of 
EUR 160.1 million  in year 2020. The additional tax revenue can be allocated to the econ-
omy through different revenue recycling options. We compare the following five revenue 
recycling options (in each option we have introduced a yearly carbon tax of EUR 15/tCO2 
(EUR 55/tC), while other assumptions remain the same as in the baseline scenario):
18 QR = total manufacturing output (EUR million). The model is based on an analysis of 42 different sectors.
19 YRE = Employment (thousands). The model is based on an analysis of 42 different industries. 
20 RGHG = Greenhouse gas emissions (in CO2 equivalent thousands of tons)

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 260
a) The first scenario analyses the effects of introduction of the extra carbon tax and revenue 
recycling through a decrease in the budget deficit and tax revenue.
b) In the second scenario, we study the effects of revenue recycling through a decrease in 
social security contributions for the workers/employees, equivalent to the amount of 
green tax revenue (fiscal neutrality). Although the yearly decrease of workers’ social con-
tributions varies by year, depending on the green tax collected, the average decrease in 
the period 2012-2030 was 0.6 percentage points i.e. the worker social contributions were 
on average equal to 18.0% in the observed period (2012-2030).
c) In the third scenario we analyse the effects of revenue recycling through a corresponding 
decrease in social security contributions payable by the employers subject to the princi-
ple of fiscal neutrality. Although the yearly decrease of employers’ social contributions 
varies by year, depending on the green tax collected, the average decrease in the period 
2012-2030 was 0.6 percentage points i.e. the employers’ social contributions were on 
average equal to 13.0% in the observed period.
d) In the fourth scenario we allocate the green tax revenue for covering the budget deficit in 
the period from 2012 to 2016, and for a decrease in workers’ social security contributions 
in 2017 and thereafter. Assuming fiscal neutrality, green tax revenue were first allocated 
to the budget (period 2012-2016) and for the period 2017-2030 we decreased the work-
ers’ social security contributions on average to 18.1%. 
e) In the fifth scenario, revenue is recycled through a decrease in budget deficit in the first 
five years (2012–2016); then, social security contributions payable by the employers are 
decreased by the relevant amount. Applying the principle of fiscal neutrality, the latter 
were decreased on average to 13.1% (0.5 percentage points) in the period 2017–2030.
 
A comparison between different types of recycling will be made especially for some 
key economic variables (household consumption expenditure, gross domestic product, 
manufacturing output, employment). Analysis of revenue recycling will be based on a 
comparison of the second, third, fourth, and fifth scenario, respectively, to the first one. 
We wish to determine the existence of the double dividend based on a decrease of some 
social security contributions, improvement in cost competitiveness and the resulting rise 
in GDP and employment.
Effect on household consumption expenditure 
Figure 1 presents the effect on household consumption expenditure (RSC) in case of dif-
ferent options of recycling of the revenue generated by the extra yearly carbon tax in the 
amount of EUR 15/tCO2. In our analysis, four scenarios (2nd, 3rd, 4th, and 5th scenario) 
are compared to the projection in which all carbon tax revenue is allocated exclusively 
for covering the budget deficit (first scenario). Figure 1 shows that the positive effect on 
household consumption expenditure in all four scenarios is stronger than in case of the 
projection in which all generated tax revenue is allocated exclusively for covering the 
budget deficit (first scenario). This is expected as additional relief through lower social 
contributions may increase the general population’s purchasing power as net wages rise. 
Furthermore, it can be observed that revenue recycling through workers’ social contribu-
tions has a higher effect on household consumption expenditure than recycling through 

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 261
social security contributions payable by the employers in the entire period at hand (both rel-
ative to the first scenario). The difference in household consumption expenditure between 
the two revenue recycling options is decreasing through the years.  The reasons can be found 
in the fact that a decrease in employers’ social security contributions would translate to a 
lower extent into an increase in net wages and the resulting increase in consumption than it 
would be the case if social security contributions were decreased for the workers.
The result is similar in the case where we allocate the green tax revenue for covering the 
budget deficit in the period from 2012 to 2016, and for a decrease in workers’ social se-
curity contributions in 2017 and thereafter. In this case, too, decrease of social security 
contributions for the workers has a stronger positive effect on household consumption ex-
penditure than a decrease of social security contributions for the employers (both in com-
parison to the first scenario). Similar as before, the differences between the two scenarios 
through the years are gradually decreasing. Figure 1 also shows that the best scenarios 
from the aspect of revenue recycling are the ones that decrease social security contribu-
tions for the workers (scenarios 2 and 4). These two scenarios are only different in the first 
five years; after that, their results tend to match. Similar match can be seen between the 
two scenarios in which the employer’s social security contributions are reduced. It should 
also be noted that the differences between all scenarios referred to are relatively small.
Figure 1: Comparison between different forms of carbon tax revenue recycling from the 
aspect of effect on household consumption expenditure, RSC.
Source: E3ME program and own calculations.
Effect on gross domestic product
Figure 2 shows the effect of introduction of a yearly carbon tax in the amount of EUR 15/
tCO2 on GDP (RGDP) in different cases of tax revenue recycling. In our analysis, four 
scenarios (2nd, 3rd, 4th, and 5th scenario) are compared to the first scenario in which 
1 
 
 
 
 
 
 
  
 
0
0,1
0,2
0,3
0,4
0,5
0,6
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
 % change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first 5
years, followed by a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first 5
years, followed by a decrease of
employers' social security
contributions
0
0,05
0,1
0,15
0,2
0,25
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
% change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of workers' social
security contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of employers'
social security
contributions

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 262
all carbon tax revenue is allocated exclusively for covering the budget deficit. It is evi-
dent from Figure 2 that the positive effect on GDP in all four scenarios is stronger than 
in case of the projection in which all generated tax revenue is allocated exclusively for 
covering the budget deficit. This matches our expectations as additional relief of labour 
costs through a decrease in social security contributions payable by the employers or the 
workers translates into an increase in household purchasing power and in turn an increase 
in GDP . The positive effect is stronger in case of revenue recycling through a decrease in 
worker’s social security contributions in the entire period at hand (both relative to the first 
scenario). The difference between the two revenue recycling options is decreasing through 
the examined period. The reasons for this can be found in higher household consumption 
expenditure (see previous section) which is the largest component of GDP . 
The result is similar in the case where green tax revenue is allocated for covering the 
budget deficit in the period from 2012 to 2016, and for a decrease in social security contri-
butions in 2017 and beyond. Decrease of social security contributions for the workers has 
a stronger positive effect on household consumption expenditure than a decrease of social 
security contributions for the employers (both in comparison to the first scenario). In this 
case, too, the differences between the two scenarios are gradually decreasing through the 
years. Figure 2 also shows that the best scenarios from the aspect of revenue recycling are 
the ones that decrease social security contributions for the workers (scenarios 2 and 4). 
These two scenarios are only different in the first five years; after that, their results tend 
to match. Similar match can be seen between the two scenarios in which the employer’s 
social security contributions are reduced. It should again be noted that the differences be-
tween all scenarios in terms of discrepancy relative to the first scenario are relatively small.
Figure 2: Comparison between different forms of carbon tax revenue recycling from the 
aspect of effect on gross domestic product, RGDP .
Source: E3ME program and own calculations.
1 
 
 
 
 
 
 
  
 
0
0,1
0,2
0,3
0,4
0,5
0,6
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
 % change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first 5
years, followed by a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first 5
years, followed by a decrease of
employers' social security
contributions
0
0,05
0,1
0,15
0,2
0,25
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
% change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of workers' social
security contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of employers'
social security
contributions

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 263
Effect on total manufacturing output 
Following is a presentation of the effect of carbon tax introduction on manufacturing output 
(QR) in case of different forms of recycling. Figure 3 compares four scenarios to the projection 
in which all carbon tax revenue, is allocated exclusively for covering the budget deficit (first 
scenario). It is evident from Figure 3 that the positive effect on manufacturing output in all 
four scenarios (2nd, 3rd, 4th, and 5th) is stronger than in case of the projection in which all 
generated tax revenue is allocated exclusively for covering the budget deficit. Higher cost relief 
through a decrease in social security contributions of the employer or the worker and the re-
sulting improvement in cost efficiency appears to motivate total manufacturing output as well. 
Recycling through a reduction in social security contributions of the workers has a more posi-
tive effect on production than recycling through decrease in social security contributions for 
the employers in the period 2012–2030 (both relative to the first scenario). The result is similar 
in the case where we allocate the green tax revenue for covering the budget deficit in the pe-
riod from 2012 to 2016, and for a decrease in social security contributions in 2017 and there-
after. In both cases, decrease of social security contributions for the workers has a stronger 
positive effect on manufacturing output than a decrease in the employer’s social security con-
tributions. Again, the differences between all scenarios in terms of discrepancy relative to the 
first scenario are relatively small.
Figure 3: Comparison between different forms of carbon tax revenue recycling from the 
aspect of total manufacturing output, QR.
Source: E3ME program and own calculations.
2 
 
 
 
 
 
 
 
 
0
0,05
0,1
0,15
0,2
0,25
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
% change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of workers' social
security contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of employers' social
security contributions
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
% change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first
5 years, followed by a decrease
of workers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first
5 years, followed by a decrease
of employers' social security
contributions

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 264
Effect on employment
Following is a presentation of the effect of carbon tax introduction on employment (YRE) 
in case of different forms of recycling. Four scenarios are compared to the projection in 
which all carbon tax revenue, is allocated exclusively for covering the budget deficit (first 
scenario). Figure 4 shows that the positive effect on employment in all four scenarios 
(2nd, 3rd, 4th, and 5th) is stronger than in case of the projection in which all generated tax 
revenue is allocated exclusively for covering the budget deficit. Higher cost relief through 
a decrease in social security contributions (of the employer or the worker) evidently has 
a positive effect on employment, which is also consistent with the previous two figures. 
Revenue recycling through a decrease of the employer’s social security contributions has 
a stronger effect on employment than revenue recycling through worker’s social security 
contributions, but only in the short run until the year 2014. In the long run, the opposite 
is true; after 2015, the difference between the second and the third scenario is constant. 
If carbon tax revenue is allocated for covering the budget deficit in the period 2012–2016 
and for a decrease in social security contributions in 2017 and beyond, the conclusion 
is similar. In this case, too, revenue recycling has a stronger effect in the short run (until 
2018) if the employer’s social security contributions are decreased. Differences between all 
analyzed scenarios are relatively small in terms of discrepancy relative to the first scenario.
Figure 4: Comparison between different forms of recycling in case of carbon tax introduc-
tion from the aspect of employment, YRE
Source: E3ME program and own calculations.
2 
 
 
 
 
 
 
 
 
0
0,05
0,1
0,15
0,2
0,25
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
% change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of workers' social
security contributions
% change between revenue
recycling into budget and
recycling into budget in the
first 5 years, followed by a
decrease of employers' social
security contributions
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
20 11 20 13 20 15 20 17 20 19 20 21 20 23 20 25 20 27 20 29
% change 
year 
% change between revenue
recycling into budget and
recycling into a decrease of
workers' social security
contributions
% change between revenue
recycling into budget and
recycling into a decrease of
employers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first
5 years, followed by a decrease
of workers' social security
contributions
% change between revenue
recycling into budget and
recycling into budget in the first
5 years, followed by a decrease
of employers' social security
contributions

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 265
5. CONCLUSION
The main goal of the environmental tax reform is economic and environmental improve-
ment. Environmental dividend involves reduction in emissions, while economic dividend 
has to do with improved cost competitiveness, higher growth, and higher employment. 
Our primary goal was to determine the effect of an extra carbon tax (EUR 15 per ton of 
CO2 i.e. EUR 55 per ton of carbon) in the period 2012–2030 on Slovenian economy, in 
order to determine whether a carbon tax would indeed yield a double dividend.
In the first section, we analysed the effects of the introduction of a yearly carbon tax (EUR 
15 per ton of CO2) relative to the baseline projection (in which no tax is introduced) in the 
period 2012–2030, using the E3ME model. Our analysis has shown that average prices of 
fuels will increase which will reduce demand for fuels. Higher prices will also lead to lower 
household consumption expenditure, which would decrease aggregate demand and result 
in a drop of GDP. GDP would be additionally decreased in the short run by lower export 
competitiveness of the Slovenian economy, resulting from higher prices, as we assumed that 
the change in prices would not affect the European economy. In the medium and long run, 
the effect of carbon tax on the change in GDP , relative to the baseline scenario (i.e. no car-
bon tax), is always lower. This conforms to our expectations and the theoretical findings as 
economic agents require some time to adjust to the new circumstances. The E3ME model 
has shown that Slovenian export sector would look to introduce new technological solutions 
and updates, thereby neutralizing relatively quickly the negative effects of the introduction 
of the carbon tax on the competitiveness of the Slovenian economy. Similar dynamics and 
oscillation as in GDP can be observed in manufacturing output and employment. Green-
house emissions, too, are reduced in the model, at approximately the same rate. 
Economic policy developers in Slovenia, as in many other European countries with imple-
mented environmental tax reform, should be aware that introduction of a carbon tax in 
Slovenia would have more negative effects in the short run than in the medium and long 
run. It is therefore of key importance for the success of the green tax reform to introduce 
the extra carbon tax gradually, transparently, and predictably. This would allow enough 
time for economic agents to adapt, and for economic policy developers to evaluate the 
first effects of the green tax reform and to make any adjustments if discrepancies from the 
planned goals are identified in the course of the reform. This would also prevent recurring 
discussions as to the urgency of increase of some tax rates and political pressure to de-
crease such rates as a result of higher prices of oil and petrochemicals in the global market. 
In the second section, we used the E3ME model to analyze the effects of different forms of 
tax revenue recycling, either through a decrease in the budget deficit or through a decrease 
of social security contributions payable by either the employers or the workers, in case of 
a yearly carbon tax in the amount of EUR 15 per ton of CO2 in the period 2012–2030. 
Our analysis has shown that recycling through lowering the social security contributions for 
workers (2nd and 4th scenario) and employers (3rd and 5th scenario) have a stronger posi-
tive effect on household consumption expenditure than the scenario in which all revenue is 
allocated exclusively for covering the budget deficit (first scenario). Differences between the 

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 266
recycling scenarios are relatively small. Additional relief through a decrease in social security 
contributions in case of an extra carbon tax would increase the purchasing power of the gen-
eral population (household consumption expenditure), which would in turn increase the 
GDP . Higher cost relief through a decrease in social security contributions also has a positive 
effect on total manufacturing output and employment. We have also shown that recycling 
through a decrease in social security contributions of workers has a stronger positive eco-
nomic effect than recycling through a decrease in employers’ social security contributions 
in the entire period at hand. The result is similar in the case where we allocate the green tax 
revenue for covering the budget deficit in the period from 2012 to 2016, and for a decrease 
in workers’ or employers’ social security contributions in 2017 and thereafter. 
Policy implications for the Slovenian government are twofold. Firstly, scenarios in which all 
revenue is allocated exclusively for lowering the social security contributions for workers/em-
ployers have a stronger positive economic effect than the scenario in which all revenue is al-
located exclusively for covering the budget deficit. Secondly, the optimal fiscal instrument for 
improving the environmental (first dividend) and economic welfare (second dividend) seems 
to be recycling through a decrease in social security contributions of workers. The reasons can 
be found in the fact that a decrease in employers’ social security contributions would translate 
to a lower extent into an increase in net wages and the resulting increase in consumption than 
it would be the case if social security contributions were decreased for the workers. 
However, an environmental tax reform cannot be successful if the political reality in Slo-
venia is disregarded. As a rule, economists design optimum policy mixes for the attain-
ment of certain goals; however, politics often requires compromises. Experience from 
other countries has shown that the key to their success was the high rate of consent of all 
political parties and civil society regarding the urgency of an environmental tax reform. 
Therefore, the Slovenian government should inform the public about the negative effects 
of an extra carbon tax. Public support will be higher, if an effective system of measures is 
put into place to neutralize the harmful effects of the additional carbon tax. 
REFERENCES
Albrecht, J. (2006). The use of consumption taxes to re-launch green tax reforms. International Review of Law 
and Economics, 26(1), 88–103.
Barker, T.S. & Ekins, P . & Johnstone, N. (1995). Global Warming and Energy Demand. London: Routledge.
Baron, R. (1997). Economic/Fiscal Instruments: Competitiveness Issues Related to Carbon/Energy Taxation Policies 
and Measures for Common Action. Working Paper No. 14. Paris: OECD.
Benoit, B. (2000). Environmental tax reform: does it work? A survey of the empirical evidence. Ecological Eco-
nomics, 34(1), 19–32.
Berkhout, G., Ferrer-i-Carbonell A. & Muskens, A.C. (2001). Het effect van de REB op huishoudelijk energie-
gevruik. EEN econometrische analyse. Amsterdam.
Bentzen, J. & Engsted, T. (1993). Short- and long-run elasticities in energy demand: a cointegration approach. 
Energy Economics, 15(1) 9-16.
Bovenberg, L. & Goulder, L.H. (1997). Cost of environmentally motivated taxes in the presence of other taxes: 
general equilibrium analyses. National Tax Journal, 50(1), 59–87.

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 267
Camecon (2012). Teaching course. 15th-16th March, 2012.
Carraro, C., Galeotti, M. & Gallo, M. (1996). Environmental taxation and unemployment: Some evidence on the 
‘double dividend hypothesis’ in Europe. Journal of Public Economics, 62(1/2), 141-181.
Ciaschini, M., Pretalori, R., Severini, F., Socci, C. (2012): Regional Double Dividend from Environmental Tax 
Reform: An Application from the Italian Economy. Research in Economics, 66(3): 273-283.
Clinch, P., Dunne, L. & Dresner, S. (2006). Environmental and wider implications of political impediments to 
environmental tax reform. Energy Policy, 34(8), 960–970.
Coxhead, I. (2000). Tax Reform and the Environment in Developing Economies: Is a Double Dividend Possible? 
Wisconsin-Madison Agricultural and Applied Economics Staff Paper No. 431. Madison: University of Wiscon-
sin-Madison. 
Doornik, J. (2007). Ox: An Object-Oriented Matrix Language. London: Timberlake Consultants Press.
De Mooij, R.A. (1999). The double dividend of an environmental tax reform. In Bergh, V. (Ed.), Handbook of 
Environmental and Resource Economics (pp. 600-613). London: Edward Elgar. 
E3ME Manual (2012).
http://www.camecon.com/Libraries/Downloadable_Files/E3ME_Manual.sflb.ashx (Accessed on 20.12.2012)
E3ME (2014). Technical Manual, Version 6.0. Cambridge: Cambridge Econometrics.
Ekins, P. (2009). Resource Productivity, Environmental Tax Reform and Sustainable Growth in Europe. Berlin: 
Anglo-German Foundation.
Engle, R.F. & C.W .J. Granger, C.W .J. (1987). Cointegration and error correction: representation, estimation and 
testing. Econometrica, 55, 251-76.
European Commission (2010). EU energy trends to 2030 – update 2009. Brussels: European Commission.
European Commission (2012). Taxation trend in the EU. Brussels: European Commission.
Franzén, M. & Sterner, T (1995). Long-run Demand Elasticities for Gasoline. In  Barker, T. & Johnstone, N. & 
Ekins, P . (eds.), Global Warming and Energy Elasticities. London: Routledge.
Fraser, I. & W aschik, R. (2013): The Double Dividend Hypothesis in CGE Model: Specific factors and the Carbon 
Case. Energy Economics, 39, 283-295.
Gimenez, E. & Rodriguez, M. (2010). Reevaluating the first and the second dividends of environmental tax 
reforms. Energy Policy, 38(11), 6654–6661.
Glomm, G., Kawaguchi, D. & Sepulveda, F. (2008). Green taxes and double dividends in a dynamic economy. 
Journal of policy modeling, 30(1), 19-32.
Goulder, L.H. (1995). Environmental taxation and the ‘‘double dividend’’: a reader’s guide. International Tax and 
Public Finance, 2(2), 155–182.
Hansen, M. & Holger, H. (2000). Green tax reform in Denmark. In Schlegelmilch, Kai (Ed.), Green Budget reform 
in Europe: Countries at the forefront (pp. 139-145). Berlin: Springer.
Hendry, D. F. & Pagan, A. & Sargan, J.D. (1984). Dynamic specification.  In Griliches, Z. & Intriligator, M.D. 
(eds), Handbook of Econometrics, Vol II (pp. 1023-1100).  Amsterdam: Elsevier.
Hoerner, B. (2001). Environmental tax reform: The European Experience. Washingtom DC: Center for Sustain-
able Economy. 
Holmlund, B. & Kolm, A.S. (2000). Environmental tax reform in a small open economy with structural unem-
ployment. International Tax and Public Finance, 7(3), 315–333.
Hunt, L. & Manning, N. (1989). Energy price- and income-elasticities of demand: some estimates for the UK 
using the cointegration procedure. Scottish Journal of Political Economy, 36(2), 183-193.
Johansson, O. & Schipper, L. (1997). Measuring the long-run fuel demand of cars. Journal of Transport Econom-
ics and Policy, 31(3), 277-292.
Johansson, B. (2000). The carbon tax in Sweden. In: OECD, Innovation and the Environment (pp. 85-94). Paris: 
OECD.

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 268
Koskela, E. & Schob, R. (1999). Alleviating unemployment: the case for green tax reform. European Economic 
Review, 43(9), 1723–1746.
Kosonen, K. & Nicodème, G. (2009). The role of fiscal instruments in environmental policy. Working paper No. 
2719, Munchen: CESifo.
Labandeira, X. & Labeaga, J. (1999). Combining inpt-ouput analysis and microsimulations to acces the effects of 
carbon taxation on spanish households. Fiscal studies, 20(3), 305-320.  
Labandeira, X., Labeaga, J.M. & Rodríguez, M. (2004). Green tax reforms in Spain. European Environment, 14(5), 
290–299.
Hua Li, Longyang Ren (2012). China’s Provincial Tax Structure Optimization: Double Dividend of Energy and 
Efficiency. Finance & Trade Economics, 10(3).
Ludewig, D., Meyer, B. & Schlegelmilch, K. (2010). Greening the Budget: Pricing Carbon and Cutting Energy Sub-
sidies to reduce the financial deficit in Germany. Washington DC: Heinrich Boll Stiftung.
Manresa, A. & Sancho, F. (2005). Implementing a double dividend: recycling ecotaxes towards lower labour 
taxes. Energy Policy, 33(12), 1577–1585
Morgenstern, R. (1995). Environmental Taxes: Dead or Alive? Discussion Paper 95-03. Washington, DC: Re-
sources for the Future.
OECD (2007). Employment Outlook. Paris: OECD.
Parry, I. & Oates, W . (1998). Policy Analysis in a Second Best World. Discussion paper No. 98-48. Washington, 
DC: Resources for the Future.
Patuelli, R., Nijkamp, P . & Pels, E. (2005). Environmental tax reform and the double dividend: A meta-analytical 
performance assessment. Ecological Economics, 55(4), 564-583.
Pigou, A.C. (1920). The Economics of Welfare. London: Weidenfeld and Nicolson.
Roed, L. (2006). Distributional effects of environmental taxes on transportation – evidence from Engel curves in 
the USA. Journal of Consumer Policy, 29(39), 301-318.
Sandmo, A. (1975). Optimal taxation in the presence of externalities. Swedish Journal of Economics, 77(1), 86–98.
Schöb, R. (2003). The Double Dividend Hypothesis of Environmental Taxes: A Survey. Working paper No. 946. 
Munchen: CESifo. 
Speck, S., Jilkova, J. (2009). Design of environmental tax reforms in Europe. Carbon energy taxation: lessons from 
Europe. Oxford: Oxford University Press.
Ščasný, M., Píša, V ., Pollitt, H. & Chewpreecha, U. (2009). Analyzing Macroeconomic Effects of Environmental 
Taxation in the Czech Republic with the Econometric E3ME Model. Journal of Economics and Finance, 59(5), 
460-491.
Tiezzi, S. (2001). The welfare effects of carbon taxation in Italian households. Working paper No. 337. Siena: Di-
partimento di Economica Politica, Universita degli Studi di Siena.
Tuladhar, S.D. & Wilcoxen, P .J. (1999). An Econometric Look at the Double Dividend Hypothesis. National Tax 
Association Proceedings, 57-62.
Turner, K., Pearce, D. & Bateman, I. (1994). Environmental economics: An elementary introduction. Prentice Hall: 
Pearson Education.
West, S. & Williams, R. (2004). Estimates from a Consumer demands system: Implication for the incidence of 
environmental taxes. Journal of Environmental Economics and Management, 47(3), 535-558. 
Yohe, G.W ., Lasco, R.D., Ahmad, Q.K., Arnell, N.W ., Cohen, S.J., Hope, C., Janetos, A.C. & Perez, R.T. (2007). 
Perspectives on climate change and sustainability. In Parry, M.L., Canziani, O.F., Palutikof, J.P ., van der Linden, 
P .J. & Hanson, C.E. (Eds.), Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Work-
ing Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 811-841). 
Cambridge: Cambridge University Press.

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 269
APPENDIX
Table A1: PRIMES (Baseline 2009) for Slovenia. 
3 
 
 
 
Table A1: PRIMES (Baseline 2009) for Slovenia. 
 
  

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 270
Source: EU energy trends to 2030 – update 2009 (2010), pp. 114-115.
 
4 
 
 
 
 
 
 
 
 
 

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 271
Table A2: Equation summary.
Source: E3ME Manual (2012).
Table A3: Baseline assumptions, complete with sources.
  DATA SOURCES
World assumptions
1. Commodity prices
     - food CE own assumptions
     - beverages CE own assumptions
     - agricultural raw materials CE own assumptions
     - metals CE own assumptions
     - energy IEA, PRIMES
     - oil IEA, PRIMES
     -  global inflation CE own assumptions
Region specific assumptions
1. Exchange rates DG ECFIN  AMECO database over  
historical, fixed afterwards
5 
 
 
 
Table A2: Equation summary. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 272
     - euro exchange rates (WREX)
     - purchasing power standard (WRPX)
2. Interest rates DG ECFIN  AMECO database over  
historical, fixed afterwards
    - short-term rate (WRSR) 
    - long-term rate  (WRLR)
3. Macro variables Not use for E3ME regions (endogenous) 
forecasts calibrated to PRIMES 2009 
projection
Historical data stored in databank from 
Eurostat 
Other Regions (CE own assumptions + 
results from E3MG modelling)
    - GDP (WGDP)
    - GDP deflator (WHUC)
4. Government consumption (WRSG, 
GW01, GW02,GW03)
Eurostat, Cambridge Econometrics
   - defence     - fixed after last year of historical data
   - education      - fixed after last year of historical data
   - health     - fixed after last year of historical data
5. Fiscal policy DG ECFIN AMECO database, DG TAX 
AND CUSTOMS “T axes in Europe” database 
over historical period,  fixed afterwards
- taxes on goods and services (WITR)
- standard rate on V AT  (WSVT)
- taxes on income and capital gains 
(WDTR)
- taxes on international trade (WTTR) 
- subsidies and other transfers to 
households (WBNR)
- social security taxes paid by employees 
(WSSR)
- social security taxes paid by employers 
(WERS)
6.  Population (WRPO, P AR1…. P AR6) Eurostat population projections
   - total population
   - male/female split
   -  children/working-age/ old-age 
pensioner split 

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 273
7. Labor force (LRP1, LRP2) Not use for E3ME regions (endogenous)
Historical data stored in databank from 
Eurostat LFS
   - male/female participation rates
Source: E3ME program.
Table A4: Baseline assumptions for Slovenia and the world in the E3ME model. 
Source: E3ME program. 
7 
 
 
 
Table A4: Baseline assumptions for Slovenia and the world in the E3ME model.  
 
SLOVENIA
 Code Description unit 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
WREX01            Ex hange rate local currency per euro 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
WRPX01            Ex hange rate: PPP (not used) local currency per euro 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386 1.386
WRSR01            Interest rate: short run (not used) percent 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046 0.046
WRLR01            Interest rate: long run percent 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038 0.038
WGDP01_MA06       G DP (not used for E3ME regions) year on year growth 1.379 1.937 2.451 3.248 3.248 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541 3.541
WHUC01_MA12       inflation (not used for E3ME regions) annual rate 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
WRSG01_MA02       G overnm ent spending year on year growth 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029 1.029
GW 01_DEFENCE      G overnm ent spending: Defence share of total governm ent spending 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06
GW02_EDUCATION    G overnm ent spending: Education share of total governm ent spending 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245 0.245
GW 03_HEA LTH       G overnm ent spending: Health share of total governm ent spending 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297 0.297
WITR01_TAX_G&S    Tax : Indirect 1+share of household spending 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184 1.184
WSVT01_TAX_VAT    T a x: VA T Rate 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
WDTR01_TAX_INC    Tax : Direct Rate (wages) 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174 0.174
 WTTR01_TAX_TRADE  Tax: Import tariffs (not used) Rate 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002 1.002
WBNR01_SUBS&TRANS Benefit Payment share of wage 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366 0.366
WSSR01_SS_TOTAL   Soc. sec em ployees' contibution rate 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186 0.186
WERS01_SS_ERS     Soc. sec em ployers' contibution rate 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136 0.136
WRPO_POP_TOTAL    Population year on year growth 0.276 0.237 0.209 0.198 0.168 0.14 0.116 0.087 0.056 0.03 -0.008 -0.035 -0.07 -0.104 -0.138 -0.161 -0.185 -0.208 -0.225 -0.245 -0.264
PAR1_M_CHILD      Population: m ale 0-15 share of total population 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.069 0.068 0.067 0.067 0.066 0.065 0.064 0.063
PAR2_F_CHILD      Population: fem ale 0-15 share of total population 0.066 0.066 0.067 0.067 0.067 0.068 0.068 0.068 0.068 0.069 0.069 0.068 0.068 0.067 0.067 0.066 0.065 0.064 0.063 0.063 0.062
PAR3_M_WORK_AG E   Population: m ale 16-64 share of total population 0.357 0.357 0.356 0.354 0.352 0.35 0.347 0.345 0.342 0.339 0.336 0.333 0.331 0.329 0.327 0.325 0.324 0.322 0.32 0.319 0.317
PAR4_F_WORK_AG E   Population: fem ale 16-64 share of total population 0.339 0.339 0.338 0.336 0.334 0.332 0.329 0.327 0.324 0.321 0.319 0.317 0.314 0.313 0.311 0.31 0.308 0.307 0.305 0.304 0.303
PAR5_M_OLD        Population: m ale 65+ share of total population 0.066 0.066 0.067 0.069 0.071 0.074 0.076 0.079 0.082 0.085 0.088 0.091 0.094 0.097 0.099 0.102 0.104 0.107 0.109 0.112 0.114
PAR6_F_OLD        Population: fem ale 65+ share of total population 0.102 0.102 0.102 0.104 0.105 0.107 0.109 0.111 0.113 0.116 0.118 0.12 0.123 0.125 0.127 0.13 0.132 0.134 0.137 0.139 0.141
LRP1_M_PARTN_RATE Participation rate: male (not used) percent of m ale working population 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761 0.761
LRP2_F_PARTN_RATE Participation rate: female (not used) percent of fem ale working poulation 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675 0.675
W ORLD
 Code Description unit 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
PFMG(01)          Com m odity Price: Food year on year growth 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8
PFMG(02)          Commodity Price: Beverages year on year growth 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
PFMG(03)          Commodity Price: Agriculture Raw Material year on year growth 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4 2.4
PFMG(04)          Commodity Price: Metals & Minerals year on year growth 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2
PFMG(05)          Com m odity Price: Energy year on year growth 6.031 6.929 6.929 6.929 6.929 6.929 5.861 5.861 5.861 5.861 5.861 4.566 4.566 4.566 4.566 4.566 4.566 4.566 4.566 4.02 4.02
PFMG(06)          Com m odity Price: Brent oil year on year growth 20.586 2.27 2.27 2.27 2.27 2.27 6.367 6.367 6.367 6.367 6.367 5.126 5.126 5.126 5.126 5.126 5.126 5.126 5.126 4.814 4.814
PFMG(07)          Agregate G lobal Inflation year on year growth 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
 
 
 
  

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 274
Table A5: Values of economic, environmental, and energy variables in different scenarios 
for the period 2011-2030. 
Source: E3ME program and own calculations. 
Appendix 1: Aggregate demand for energy and its parameters for Slovenia.
In Table A6 we show the specification of aggregated demand for energy that is used in the 
E3ME model. The equation is based on the work of Barker, Ekins and Johnston (1995), 
Hunt and Manning (1989) and  Bentzen and Engsted (1993). 
»The aggregate energy equation considers the total fuel used (summation of 12 fuel types) 
in thousand tonnes of oil equivalent (th.toe) by 19 fuel users. The demand for energy by a 
fuel user is dependent on the ‚activity‘ for the fuel user. This is chosen as gross economic 
8
SCENARIO RECYCLING / YEAR 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Green revenues from a cabon tax                  
baseline scenario / 00000000000000000000
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) all 0 144.639 147.335 150.279 153.514 154.514 155.882 157.703 159.536 160.94 159.514 158.126 157.011 156.118 155.148 153.213 151.235 149.65 148.493 147.099
baseline scenario / 28749.77 29436.21 30415.7 31443.8 32502.98 33411.88 34355.28 35349.36 36385.24 37449.73 38176.75 38926.43 39697.25 40507.05 41345.79 41865.08 42408.16 42959.87 43502.27 44051.21
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 28749.77 29415.28 30322.49 31386.87 32439.38 33356.36 34305.07 35300.83 36336.85 37402.58 38133.77 38889.22 39663.53 40473.73 41309.04 41825.57 42374.68 42926.33 43462.15 44009.93
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 28749.77 29478.9 30380.23 31450.78 32501.63 33419.58 34367.1 35350.45 36384.39 37445.14 38171.98 38916.1 39691.16 40504.11 41344.97 41858.34 42407.76 42959.18 43488.57 44029.78
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 28749.77 29446.68 30353.49 31427.92 32481 33394.67 34337.66 35326.3 36362.72 37426.74 38155.76 38906.42 39681.3 40491.9 41329.51 41846.16 42394.95 42944.81 43476.59 44021.18
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 28749.77 29415.28 30322.49 31386.87 32439.38 33356.36 34360.18 35339.48 36378.32 37451.36 38185.97 38932.13 39706.17 40515.79 41349.45 41852.42 42399.23 42950.67 43487.95 44039.86
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 28749.77 29415.28 30322.49 31386.87 32439.38 33356.36 34328.9 35318.64 36363.46 37433.86 38163.37 38911.42 39684.97 40495.51 41331.45 41843.31 42391.35 42941.27 43476.87 44026.62
baseline scenario / 16705.19 16945.31 17470.62 18004.52 18550.77 19074.5 19612.02 20159.48 20729.08 21316.29 21789.22 22266.06 22770.05 23289.98 23818.74 24183.24 24543.27 24956.55 25382.57 25810
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 16705.19 16891.03 17392.22 17931.52 18477.28 19004.83 19545.53 20096.52 20667.77 21256.98 21732.45 22210.29 22713.19 23230.92 23758.01 24122.45 24495.48 24913.65 25332.61 25754.82
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 16705.19 16975.57 17477.51 18022.34 18559.2 19085.13 19623.32 20160.12 20731.11 21320.4 21797.08 22264.55 22769.02 23288.56 23818.11 24170.78 24539.25 24958.42 25376.35 25796.2
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 16705.19 16944.68 17447.24 17989.73 18527.51 19052.41 19590.35 20132.39 20703.96 21293.73 21770.3 22242.09 22746.6 23265.72 23794.24 24151.53 24521.36 24939.85 25358.64 25779.81
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 16705.19 16891.03 17392.22 17931.52 18477.28 19004.83 19623.81 20164.48 20737.44 21328.23 21803.74 22267.61 22768.59 23287 23817.9 24169.99 24541.67 24960.25 25379.2 25801.98
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 16705.19 16891.03 17392.22 17931.52 18477.28 19004.83 19595.12 20140.13 20712.34 21300.78 21774.67 22242.68 22744.26 23262.72 23792.73 24149.85 24522.38 24941.19 25360.65 25783.29
baseline scenario / 16705.19 16945.31 17470.62 18004.52 18550.77 19074.5 19612.02 20159.48 20729.08 21316.29 21789.22 22266.06 22770.05 23289.98 23818.74 24183.24 24543.27 24956.55 25382.57 25810
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 16705.19 16891.03 17392.22 17931.52 18477.28 19004.83 19545.53 20096.52 20667.77 21256.98 21732.45 22210.29 22713.19 23230.92 23758.01 24122.45 24495.48 24913.65 25332.61 25754.82
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 16705.19 16975.57 17477.51 18022.34 18559.2 19085.13 19623.32 20160.12 20731.11 21320.4 21797.08 22264.55 22769.02 23288.56 23818.11 24170.78 24539.25 24958.42 25376.35 25796.2
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 16705.19 16944.68 17447.24 17989.73 18527.51 19052.41 19590.35 20132.39 20703.96 21293.73 21770.3 22242.09 22746.6 23265.72 23794.24 24151.53 24521.36 24939.85 25358.64 25779.81
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 16705.19 16891.03 17392.22 17931.52 18477.28 19004.83 19623.81 20164.48 20737.44 21328.23 21803.74 22267.61 22768.59 23287 23817.9 24169.99 24541.67 24960.25 25379.2 25801.98
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 16705.19 16891.03 17392.22 17931.52 18477.28 19004.83 19595.12 20140.13 20712.34 21300.78 21774.67 22242.68 22744.26 23262.72 23792.73 24149.85 24522.38 24941.19 25360.65 25783.29
baseline scenario / 21531.41 22199.45 22996.59 23848.7 24729.97 25510.09 26305 27079.24 27880.9 28705.55 29188.84 29579.96 30044.5 30518 31034.2 31365.94 31634.17 31984.08 32300.69 32646.13
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 21531.41 22214.99 22980.56 23826.07 24692.9 25477.39 26276.24 27057.75 27861.56 28683.53 29166.47 29562.1 30035.12 30511.67 31026.06 31348.3 31652.37 32012.02 32319.98 32665.44
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 21531.41 22229.12 22990.89 23850.03 24719.05 25502.39 26295.74 27072.12 27872.67 28688.84 29167.51 29558.76 30035.01 30513.88 31034.17 31358.98 31651.61 32016 32318.01 32659.15
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 21531.41 22211.22 22976.19 23839.81 24714.27 25495.9 26284.69 27062.03 27864.91 28685.59 29167.42 29564.31 30038.1 30513.06 31029.63 31359.26 31654.73 32013.31 32315.45 32658.35
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 21531.41 22214.99 22980.56 23826.07 24692.9 25477.39 26288.89 27055.07 27870.52 28696.19 29179.96 29571.86 30047.22 30522.01 31034.96 31350.33 31652.39 32008.75 32315.32 32662.03
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 21531.41 22214.99 22980.56 23826.07 24692.9 25477.39 26266.32 27043.07 27865.38 28694.51 29175 29566.49 30037.94 30513.28 31029.36 31355.16 31657.81 32011.4 32315.07 32660.95
baseline scenario / 936.692 950.426 955.094 955.764 958.411 952.612 946.927 938.577 932.254 925.688 920.588 913.883 909.383 905.041 901.945 894.97 887.232 880.346 873.976 869.069
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 936.692 950.497 953.889 953.042 955.141 949.1 943.516 935.464 929.486 923.195 918.308 911.838 907.599 903.398 900.288 893.212 886.057 879.566 873.35 868.53
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 936.692 952.573 957.562 958.191 960.549 954.568 948.615 939.884 933.233 926.34 921.176 914.372 910.114 905.951 903.019 895.697 888.096 881.357 874.887 869.867
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 936.692 953.847 957.965 957.848 959.617 953.441 947.391 938.413 931.928 925.27 920.258 913.345 909.153 905.006 902.028 894.562 887.215 880.625 874.298 869.396
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 936.692 950.497 953.889 953.042 955.141 949.1 945.369 938.196 933.23 927.152 922.413 915.526 910.96 906.343 902.984 895.451 888.059 881.422 875.185 870.382
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 936.692 950.497 953.889 953.042 955.141 949.1 946.645 938.51 932.958 926.643 921.709 914.505 909.912 905.366 902.092 894.496 887.281 880.734 874.513 869.717
baseline scenario / 5588.252 5737.804 5874.773 5998.528 6122.417 6186.53 6258.1 6330.867 6398.466 6463.33 6332.618 6210.716 6091.337 5971.548 5854.552 5768.035 5689.808 5612.959 5535.926 5457.153
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 5588.252 5699.225 5801.967 5910.516 6027.753 6082.999 6146.05 6218.732 6292.517 6359.086 6221.624 6089.353 5966.265 5851.007 5738.6 5651.438 5565.019 5486.641 5416.851 5343.681
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 5588.252 5699.981 5805.95 5914.157 6031.337 6086.616 6150.177 6222.895 6296.066 6362.511 6224.386 6091.057 5966.674 5851.668 5740.395 5654.128 5568.177 5487.795 5416.669 5343.986
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 5588.252 5703.375 5806.024 5908.469 6024.208 6085.302 6154.044 6224.437 6292.099 6356.679 6222.169 6092.201 5968.397 5851.377 5738.25 5651.396 5567.058 5489.025 5417.842 5343.073
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 5588.252 5699.225 5801.967 5910.516 6027.753 6082.999 6146.645 6223.044 6296.215 6362.05 6224.478 6092.933 5969.082 5853.194 5740.442 5653.166 5566.013 5487.111 5417.339 5344.526
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 5588.252 5699.225 5801.967 5910.516 6027.753 6082.999 6150.509 6222.179 6289.215 6355.302 6224.07 6095.759 5969.932 5849.961 5736.272 5651.136 5567.327 5489.403 5417.561 5342.396
baseline scenario / 6528.858 6735.286 6929.519 7113.159 7305.409 7421.845 7549.201 7677.035 7798.161 7916.722 7704.82 7521.981 7359.104 7208.643 7072.313 7047.299 7031.582 7019.494 7007.009 6991.417
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 6528.858 6679.237 6836.475 7007.4 7193.944 7299.16 7414.321 7541.112 7670.133 7791.199 7571.272 7375.161 7207.594 7063.331 6933.324 6906.806 6879.445 6864.735 6861.451 6852.784
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 6528.858 6680.491 6841.556 7011.197 7197.421 7303.375 7420.242 7547.516 7675.336 7795.493 7574.27 7376.883 7208.198 7064.885 6936.641 6911.035 6884.056 6866.505 6861.389 6853.399
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 6528.858 6685.019 6841.814 7004.185 7188.617 7301.797 7425.046 7549.017 7669.717 7787.832 7571.796 7378.862 7210.618 7064.275 6933.412 6907.077 6882.391 6868.113 6862.943 6852.137
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 6528.858 6679.237 6836.475 7007.4 7193.944 7299.16 7415.302 7546.529 7674.064 7794.156 7574.749 7380.27 7211.954 7066.593 6935.68 6908.626 6880.418 6865.489 6862.725 6854.685
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 6528.858 6679.237 6836.475 7007.4 7193.944 7299.16 7420.551 7545.693 7665.209 7785.534 7574.371 7383.883 7212.714 7061.986 6930.139 6906.192 6882.481 6868.658 6862.821 6851.513
baseline scenario / 3144.918 3140.383 3144.66 3156.688 3174.105 3182.884 3195.34 3212.91 3235.262 3260.783 3303.043 3347.844 3396.56 3449.189 3506.829 3552.643 3598.905 3642.847 3684.911 3730.16
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget 3144.918 3255.683 3262.142 3270.724 3287.534 3298.071 3312.47 3330.179 3351.101 3375.623 3418.701 3464.836 3514.412 3566.701 3623.994 3670.967 3719.383 3764.525 3807.132 3853.669
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) workers' social security contributions 3144.918 3255.151 3261.36 3271.114 3288.359 3298.943 3312.897 3330.404 3351.084 3375.115 3417.795 3463.749 3513.401 3565.553 3622.774 3669.708 3718.03 3763.33 3805.992 3852.584
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) employers' social security contibutions 3144.918 3254.813 3261.43 3271.553 3288.754 3298.492 3311.932 3329.914 3351.437 3375.757 3418.011 3463.634 3513.341 3565.882 3623.334 3670.273 3718.387 3763.46 3806.219 3853.038
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, workers' social security contributions 3144.918 3255.683 3262.142 3270.724 3287.534 3298.071 3312.327 3329.656 3351.871 3376.649 3419.513 3465.23 3514.767 3566.705 3623.515 3670.092 3718.494 3763.498 3806.108 3852.698
carbon tax of EUR 55 per ton of carbon (EUR 15 per ton of CO2) budget, employers' social security contibutions 3144.918 3255.683 3262.142 3270.724 3287.534 3298.071 3311.844 3329.706 3352.208 3376.789 3418.878 3464.365 3514.366 3567.033 3624.048 3670.382 3718.387 3763.411 3806.358 3853.222
Total demand for energy in thousand toe (FRO) 
Average fuel (energy) price including taxes in EURO/toe (PJRT)
Gross domestic product in million of euro (RGDP)
Household consumption expenidtores in million of euro (RSC)
Export in million of euro (RSX)
Total manufacturing output in million EUR (QR)
Emplyment in thousands (YRE)
Greenhouse gas emissions in CO2 equivalent thousands of tons of carbon

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 275
output for most sectors, but household fuel demand is a function of total consumers‘ 
expenditure. A restriction is imposed such that as activity increases then demand for 
energy use will not decline (all other factors being equal). 
The average price ratio captures the effect of prices relative to the fuel used, and is de-
flated by unit costs. The equations have been tested so that relative price increases cause 
demand to fall but relative price decreases have no effect. Such asymmetrical price ef-
fects in aggregate energy demand equations have been the subject of other research 
(Gately, 1993; Walker and Wirl, 1993; Grubb, 1995). The idea is that because energy is 
used via capital stock with a long lifetime, and since technical change is progressive and 
is not generally reversed, when energy prices rise and energy savings are introduced, 
then when energy prices fall again, these savings are not reversed i.e. energy demand 
responds to rises in real prices, but not falls. The effect changes the properties of the 
model in a non-linear fashion: if in the base run real energy prices fall over the projec-
tion period, then increases in energy taxes will have no effect until they start to increase 
real prices (one year to the next, not compared to the base).
The long-run price elasticity for road fuel is imposed at -0.7 for all regions, also Slovenia, 
following the research on long-run demand (Franzen and Sterner, 1995) and (Johansson 
and Schipper, 1997). 
The measures of research and development expenditure and investment capture the ef-
fect of new ways of decreasing energy demand (energy saving technical progress) and 
the elimination of inefficient technologies, such as energy saving techniques replacing 
the old inefficient use of energy. Research and development expenditure in industries 
16-18 (machinery) and 19 (motor vehicles) for the EU as a whole take into account 
spillover effects from international companies.« (E3ME Manual, 2012, page 49-50).
Tabel A6: Specification of agregate demand for energy.
Co-integrating dynamic equation: 
DLN(FR0(.))  [total fuel used by fuel users] 
= BFR0(,.1) [constant]
+  BFR0(.,2) * DLN(FRY(.))  [activity measure] 
+  BFR0(.,3) * DLN(PREN(.))  [average price ratio] 
+  BFR0(.,4) * DLN(FRTD(.))  [R&D by fuel user] 
+  BFR0(.,5) * DLN(ZRDM)  [EU R&D in machinery] 
+  BFR0(.,6) * DLN(ZRDT)  [EU R&D in transport] 
+  BFR0(.,7) * DLN(FRK(.))  [investment by fuel user] 
+  BFR0(.,8) * DRDEU  [German unification] 
+  BFR0(.,9) * D09R  [2009 recession dummy] 
+  BFR0(.,10) * DLN(FR0(-1))  [lagged changes in fuel use] 
 

ECONOMIC AND BUSINESS REVIEW  |  VOL. 16  |  No.  3  |  2015 276
Co-integrating long-term equation: 
DLN(FR0(.))  [total fuel used by fuel users] 
= BFR0(.,11) * ECM(-1)  [lagged error correction] 
+  BFR0(,.12)  [constant]
+  BFR0(,.13) * LN(FRY(.))  [activity measure] 
+  BFR0(.,14) * LN(PREN(.))  [average price ratio] 
+  BFR0(.,15) * LN(FRTD(.))  [R&D by fuel user] 
+  BFR0(.,16) * LN(ZRDM)  [EU R&D in machinery] 
+  BFR0(.,17) * LN(ZRDT)  [EU R&D in transport] 
+  BFR0(.,18) * LN(FRK(.))  [investment by fuel user] 
+  BFR0(.,19) * RDEU  [German unification] 
+  BFR0(.,20) * D09R  [2009 recession dummy] 
+  ECM  [error] 
Identity: 
PREN           =  PFR0(.)/PRYM  [average price ratio] 
Restrictions: 
BFR0(.,3 .,4 .,5 .,6 .,7 .,14 .,15.,16 .,17 .,18) <=0  [‘right sign’] 
BFR0(.,2), BFR0(.,13) >=0  [modeling energy  
  demand/activity ratio] 
0>BRF0(.,11)>-1  [‘right sign’] 
Definitions: 
BFR0  is a matrix of parameters 
FR0  is a matrix of total fuel used by 22 fuel users for 33 regions, th toe. 
PREN  is a matrix of average price used deflated by unit cost for 33 regions, euro/toe 
FRY  is a matrix of activity for 22 fuel users and 33 regions, m euro at 2005 prices 
FRTD  is R&D in machinery by the EU, m euro at 2005 prices 
ZRDM  is R&D in transport by the EU, m euro at 2005 prices 
ZRDT  is a matrix of investment by 22 fuel users for 33 regions, m euro  
 at 2005 prices 
FRK  is a matrix of prices of value added at market prices for each region  
 (2005 = 1.0, local price) 
PRYM         is a matrix of average prices in euro/tonne of all fuels used by each fuel user 
PFR0  is a matrix of average prices in euro/tonne of all fuels used by each fuel user 
RDEU  is a dummy matrix for German unification (=0 for other countries) 
D09R  is a dummy matrix for 2009 recession (=0 until 2008, =1  
 from 2009 onward) 
(.)  indicates that a matrix is defined across sectors 
LN  indicates natural logarithm 
DLN  indicates change in natural logarithm 
ECM  [error] 
Source: E3ME Manual (2012).

A. KEŠELJEVIĆ, M. KOMAN | ANALYSIS OF THE EFFECTS OF INTRODUCTION... 277
In Table A7 we show the values of estimated paramaters of agregated demand for energy 
for Slovenia. 
Tabel A7:  Values of parameters of agregated demand for energy function for Slovenia.
 Source: E3ME model.
The price elasticities of energy demand for fuel users are for example shown in column 
3 and 14. Column 3 shows price elasticites of demand based on co-integrating dynamic 
equation, while column 14 shows long term elasticites of demand based on co-integrating 
long-term equation.. For example, 1% increase of average price ratio (variable PREN) 
causes decrease in quantity demanded for energy in road transportation for 0.7%. 
9
 
FUEL USERS 123456789 10 11 12 13 14 15 16 17 18 19 20
 1 Power own use & trans. 0.055 0 -0.328 0 -0.456 0 -0.473 0 0 -0.2 -0.95 7.964 0.247 -0.177 0 -0.088 -0.058 -0.027 0 0
 2 O.energy own use & tra -0.064 0 0 -0.031 0 -0.64 0 0 0 -0.2 -0.2 5.337 0.232 -0.331 -0.086 -0.019 -0.056 -0.044 0 0
 3 Iron & steel          0.02 0 0 0 0 -1 0 0 0 0.6 -0.95 9.106 0.117 -0.263 -0.091 -0.249 -0.037 -0.013 0 0
 4 Non-ferrous metals    0.005 0 -0.85 0 0 0 -0.169 0 0 0.095 -0.216 7.931 0.297 -0.311 -0.015 0 -0.184 -0.208 0 0
 5 Chemicals             0.008 1.2 -1.3 0 0 0 0 0 0 0.093 -0.417 7.69 0.432 -0.253 -0.135 -0.073 -0.308 -0.011 0 0
 6 Non-metallics nes     0.138 0.06 -0.273 -0.032 0 0 -0.021 0 0 0.01 -0.799 6.685 0.292 -0.279 -0.05 -0.027 -0.132 0 0 0
 7 Ore-extra.(non-energy) 0.153 0 -1.3 0 0 0 0 0 0 -0.2 -0.2 9.544 0.751 -0.331 0 -0.166 -0.653 -0.026 0 0
 8 Food, drink & tob.    -0.008 1.2 -1.3 0 0 0 0 0 0 -0.2 -0.936 4.555 0.609 -0.221 -0.003 -0.14 -0.061 -0.251 0 0
 9 Tex., cloth. & footw. -0.078 1.2 -0.504 -0.295 -1 0 -0.111 0 0 -0.2 -0.2 7.24 0.546 -0.269 -0.015 -0.049 -0.44 -0.08 0 0
10 Paper & pulp          0.049 0 -1.024 0 0 -1 -0.06 0 0 0.159 -0.2 4.684 0.635 -0.387 -0.005 -0.029 -0.106 -0.091 0 0
11 Engineering etc       0.065 0 -0.871 0 -1 0 -0.27 0 0 0.134 -0.2 6.39 0.406 -0.214 -0.005 -0.162 -0.155 -0.04 0 0
12 Other industry        0.076 0 0 0 0 0 -1.56 0 0 0.228 -0.95 12.476 0.709 -0.492 -0.02 -0.512 -0.278 -0.358 0 0
13 Rail transport        -0.042 0.844 -0.344 0 0 0 -0.024 0 0 -0.2 -0.723 5.764 0.19 -0.212 0 -0.136 -0.043 -0.016 0 0
14 Road transport        -0.107 0 -0.095 0 0 0 0 0 0 0.454 -0.574 6.184 0.602 -0.7 0 0 -0.021 -0.008 0 0
15 Air transport         0.035 0 0 0 0 0 -0.013 0 0 0.249 -0.2 5.399 0.457 -0.403 0 -0.174 0 -0.065 0 0
16 Other transp. serv.   0 0 0 0 0 0 0 0 0 0 0 0 0.146 -0.359 0 -0.08 -0.38 -0.327 0 0
17 Households            -0.004 0 0 0 0 0 0 0 0 -0.2 -0.2 3.875 0.718 -0.217 0 -0.026 -0.072 -0.258 0 0
18 Other final use       0.362 0 -0.91 -0.228 0 0 -3 0 0 0.6 -0.95 5.73 0.666 -0.248 -0.049 -0.085 -0.038 -0.361 0 0
19 Non-energy use 0.124 0 -0.681 0 -1 0 0 0 0 -0.2 -0.2 7.721 0 -0.221 0 -0.003 -0.133 0 0 0
COEFFICIENTS