N.M. Christodoulakis et al. r Energy Economics 22 2000 395]422 396
1. Introduction
The importance of energy as a key factor in explaining long-run trends in economic output growth was } to a large extent } neglected until the first oil
price shock in 1973. At that time, it became evident that the slowdown in growth and productivity observed in all industrialised countries was closely associated with
developments in the energy sector. The investigation of the macroeconomic links, which were revealed then, between energy and the rest of the economy has since
offered much scope for economists and policy makers to explain the growth process.
Lately, economic growth and energy consumption are also closely related to environmental considerations, because of concern caused by evidence of anthro-
pogenic influence on climate, resulting in the international treaties signed in Ž
. Ž
. Ž
. Toronto 1988 , Rio de Janeiro 1992 and Kyoto 1997 controlling atmospheric
gases. In particular, special attention is starting to be paid to the decrease of carbon
Ž .
dioxide CO and other ‘greenhouse’ gases emissions that primarily arise from the
2
Ž .
consumption of energy. The Council of Ministers of the European Union EU decided in October 1990 that CO emissions should be stabilised by 2000 at the
2
level of 1990.
1
Convinced of the necessity of further action on this matter, the world community adopted, in December 1997, the Kyoto Protocol to the United
Ž .
Nations Framework Convention on Climate Change UNFCCC . The Kyoto Proto- col sets out further commitments of the developed country Parties of the UNFCCC
for the period 2008]2012. The overall target agreed is a reduction of the aggregate Ž
. Ž
. anthropogenic emissions of carbon dioxide CO , methane CH , nitrous oxide
2 4
Ž .
Ž .
Ž .
N O , hydrofluorocarbons HFCs , perfluorocarbons PFCs and sulphur hexafluo-
2
Ž .
ride SF by at least 5 below the 1990 levels for the period 2008]2012.
6
The Protocol also sets out a differentiated approach for the achievement of such a reduction target among the Parties involved. Under the provisions of Article 4 of
the Protocol that enables the joint fulfilment of the commitments undertaken, the Ž
. overall target for the EU y8 will be achieved by an equitable burden sharing
between EU member states. The burden sharing should take into account differ- ences in the starting point, national circumstances and historic responsibility of
each member state in the occurrence of the ‘greenhouse’ effect. Negotiations amongst EU member states were finalised during the Environment Council in June
1998 and the burden sharing agreement specifies that Greece will have to restrict the average growth of the emissions of all six gases, for the period 2008]2012, to
1
Ž . Ž .
The decision was prompted among others, by the following factors see Surrey, 1992 : a the Ž
. consequent action of other policy measures e.g. traffic jams , which are related to the energy
Ž . Ž .
consumption and which have to in any case be enforced ‘no regrets’ policy ; b the wide differences Ž .
among interested countries, which call for common policy measures; c the necessity of switching to a more efficient use of energy sources, given the inability of substitution with other less pollutive ones;
Ž . and d the importance of developing a European know-how in this sector.
N.M. Christodoulakis et al. r Energy Economics 22 2000 395]422 397
q 25 compared to 1990 levels. This level of ambition will be achieved through a
Ž number of interventions national and EU common and co-ordinated policies and
. measures in all sectors of the economy, and particularly in the energy sector.
In this context, the purpose of this work is to provide estimates of energy consumption in Greece by sector of economic activity and by type of energy form.
Next, based on these estimates forecasts of the demand for energy will be generated for the period up to 2012, assuming a number of alternative scenarios
for the prospects of the Greek economy. The forecasts of energy consumption are then used to derive CO emissions levels, which are closely related to economic
2
growth and energy consumption. Also, attention is focused on the changes in energy consumption and CO emissions brought about by the Community Support
2
Ž .
Ž .
Framework CSF II. The CSF II also known as Delors’ II Package is an intervention plan deemed necessary in order to assist the less developed members
Ž .
of the European Union Greece, Spain, Portugal, Ireland to modernise their economies, foster growth and approach the welfare and efficiency of its most
developed members. Its second phase, CSF II, which is operational during 1994]1999, is substantially more extensive in actions and far-reaching in impact
Ž .
than the first CSF phase CSF I implemented in 1989]1993, and is expected to have a significant impact on the growth process in these countries.
One of the major CSF II interventions in Greece is the introduction of natural Ž
. gas NG to the Greek energy system. The actual cost for the NG penetration is
estimated at 2 billion ECU and is largely financed by the European Investment Ž
. Bank. According to the works programme of Public Gas Corporation PGC , total
absorption of NG will reach 3.5 billion Nm
3
per year by 2005.
2
An extended infrastructure system has been designed and is now near completion in order to
carry imported NG to the main consumption areas: first, in the electricity genera- tion sector and large industrial units and, next, in the residential and commercial
sectors } though at slower rates.
3
To enable the analysis of the above effects, it is better to integrate the demand and supply of energy into a full-developed macroeconomic model, so that
price]quantity interactions are properly taken into account. Therefore, we employ a sectoral macroeconometric model of Greece and after estimating demand
2
Approximately 85 of this total will be imported from Russia, via Bulgaria, by pipeline and the remaining 15 will be imported from Algeria in liquefied form. The construction of the main
transmission pipeline was completed in July 1995, while the construction of secondary pipelines is Ž
. progressing. The supply of natural gas to the first Public Power Corporation PPC gas-fired power
station began in 1997, while two more power stations in South Attica are scheduled to come on-line by the end of 1998. Furthermore, another natural gas fired station in northern Greece is expected to come
on-line by the end of 2001. According to the specified timetable, it is estimated that the PPC will be in a position to absorb approximately 1.1 billion Nm
3
of natural gas by the year 2000 and 1.5 billion Nm
3
by the year 2001.
3
In the industrial sector, PGC has already formulated a pricing policy and the general terms of gas supply to industrial customers, and is aggressively pursuing contracts with industrial customers. Nine
contracts have already been signed for more than 0.3 billion Nm
3
r year and a few large industrial units
are already connected to the distribution network, while negotiations with other industries are in progress. City networks are presently under construction and will be completed by the year 2000.
N.M. Christodoulakis et al. r Energy Economics 22 2000 395]422 398
functions by sector of economic activity and type of energy, the whole model is used to forecast how these variables are likely to develop in the medium-term.
In a related macroeconomic study without including energy modelling, Christo- Ž
. doulakis and Kalyvitis 1998a analysed a number of scenarios for the growth
patterns of Greece for the period up to the year 2010. The same authors Ž
. Christodoulakis and Kalyvitis, 1997 analysed the macroeconomic aspects of CSF
II effects on energy demand in Greece, but without including the introduction of NG. The present paper extends this work, first, by explicitly including the penetra-
tion of NG, in conjunction with CSF effects in the energy system, second, by employing more powerful econometric modelling techniques to the key equations
of the energy model, third, by using an updated data set covering the period 1974]1994 and extending the time horizon to 2012 and, fourth, by computing
expected CO emissions in view of the energy demand forecasts.
2
In particular, the forecasting exercise takes place under alternative assumptions about the effects that CSF II is likely to have on the Greek economy. First, a
benchmark forecast presenting economic developments in the absence of CSF interventions and NG penetration in the energy system is examined. Next, a case is
examined based on the benchmark forecast but with the assumption that NG introduction satisfies additional electricity demand. Finally, two scenarios that take
into account the impact of total CSF actions without and with NG penetration, respectively, on energy consumption and CO emissions are examined.
2
In Section 2 of the paper, the structure of energy system and the estimation Ž
. equations are discussed and presented and are given analytically in Appendix A
together with the procedure for calculating the related CO emissions. In Section
2
3, the four scenarios for the derivation of forecasts of final energy demand and CO emissions in Greece until the year 2012 are described. In Section 4 we assess
2
the quantitative effects under these scenarios on energy consumption and CO
2
emissions, and, finally, in Section 5 the main conclusions are presented.
2. Modelling energy consumption in Greece
Kategori