Internalisation of external costs in the Polish power generation sector: A partial equilibrium model

ARTICLE IN PRESS

Energy Policy 34 (2006) 3409–3422 www.elsevier.com/locate/enpol

Internalisation of external costs in the Polish power generation sector: A partial equilibrium model Mariusz Kudelko MEERI, Polish Academy of Sciences, Mineral and Energy Economy Research Institute, ul. Wybickiego 7, 30950 Krakow, Poland Available online 29 August 2005

Abstract This paper presents a methodical framework, which is the basis for the economic analysis of the mid-term planning of development of the Polish energy system. The description of the partial equilibrium model and its results are demonstrated for different scenarios applied. The model predicts the generation, investment and pricing of mid-term decisions that refer to the Polish electricity and heat markets. The current structure of the Polish energy sector is characterised by interactions between the supply and demand sides of the energy sector. The supply side regards possibilities to deliver fuels from domestic and import sources and their conversion through transformation processes. Public power plants, public CHP plants, industry CHP plants and municipal heat plants represent the main producers of energy in Poland. Demand is characterised by the major energy consumers, i.e. industry and construction, transport, agriculture, trade and services, individual consumers and export. The relationships between the domestic electricity and heat markets are modelled taking into account external costs estimates. The volume and structure of energy production, electricity and heat prices, emissions, external costs and social welfare of different scenarios are presented. Results of the model demonstrate that the internalisation of external costs through the increase in energy prices implies significant improvement in social welfare. r 2005 Elsevier Ltd. All rights reserved. Keywords: Energy system; Partial equilibrium model; External costs of energy

1. Introduction The long life of energy technologies and long-term impact of energy investment decisions on the economy, resource depletion and the environment necessitate the consideration of a long-planning horizon and interlinkages between the energy, economy and environment sectors. Literature demonstrates that welfare economics has been a fundamental framework enabling the invention and implementation of efficient tools and answers how to maintain the sustainable development of national energy systems. It has been proven that the inclusion of external costs in the decision-making process seems to be a crucial point in achieving a social optimum. Different types of energy models have been Tel.: +48 126330296; fax: +48 12323524.

E-mail address: [email protected]. 0301-4215/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.enpol.2005.01.005

rapidly developed to provide guidelines and principles for addressing various policy and planning concerns. The bottom-up modelling approach is focused mainly on micro-level technological issues and does not capture important macroeconomic inter-links within the economy. These models are mainly concentrated on the leastcost energy planning with reference to environmental constraints. They are limited for policy goals since they do not analyse the effects of price changes on other markets. Examples of this type of approach are MARKAL (Berger and Haurie, 1987), EFOM (Finon, 1974), LEAP (Raskin, 1986), and AIM (Morita et al., 1996). Conversely the top-down approach is explicitly addressed to reflect the relationship between the energy system and the rest of the economy. CGE models are particularly applied in the mid- and long-term analysis. Examples of such tools are GLOBAL 2100 (Manne, Richels, 1990), PRIMES (Rose et al., 1996), GREEN

ARTICLE IN PRESS 3410

M. Kudelko / Energy Policy 34 (2006) 3409–3422

(Burniaux et al., 1992) and Dynamic General Equilibrium Model (Jorgenson and Wilcoxen, 1990). Both types of models are commonly used in the energy policy planning but it is stressed that CGE models should be preferred for the long-term approach. Top-down models can answer energy policy issues concerning the implications for macroeconomic indicators end economy-wide emissions, due to their explicit inclusion of inter-linkages between the energy sector and the rest of the economy (Pandey, 2001). Unfortunately for developing countries their use is somewhat limited mainly due to economic instability. The lack of reliable economic data on production and price feedbacks within sectors makes it difficult to determine useful and constructive final conclusions. There are many examples of comprehensive research of the Polish energy system development. Polish energy policy goals have been studied thoroughly on several occasions. Models that have been used so far are tools created abroad and adjusted to the Polish conditions. Most of them use the least-cost method to develop mid-term electricity and heat generation strategy according to imposed environmental constrains. Typical tools that have been used by the Polish governmental agencies for energy and environmental policy planning are EFOM-ENV, ENPEP and IPM (Jankowski, 1997; ICF’s Integrated Planning Model, 1999). They have been mostly employed to examine the official government programs concerning the national energy policy, the micro-level effects of greenhouse gases reduction, costs of implementing the LCP Directive or CO2 trading scheme, etc. Attempts to link these models with the CGE tool of Polish economy have also been observed (Hille and van Leeuwen, 1993, Kiuila, 2003). The focus of this paper is to show the framework and results of the partial equilibrium model of the Polish energy sector. One of the main goals of the model concerns the verification of the mid-term Polish energy policy elaborated and approved by official government agencies. Instead of focusing on macro-economic relations the model is mainly concentrated on the electricity and heat markets. The models applied so far in Poland (for example EFOM-ENV, ENPEP and IPM) have been mostly focused on minimisations of energy supplies to the consumers by adjustment to the imposed constraints based on domestic and EU environmental legislations. They suffer from very few concerns on environmental issues, i.e. external costs of energy production. The presented research goes beyond that approach and the external costs of air emissions from energy conversion technologies are considered. That is why the verification of the Polish energy policy is here demonstrated in the spectrum of social external effects of energy production, which seems to be the first such attempt in modelling approach in Poland.

The partial equilibrium model presented here uses as an objective function the maximisation of social welfare that is defined as a sum of producers’ and consumers’ surplus minus external costs. As proven (Bigano et al., 2000; Andersson and Haden, 1997) this approach can be successfully used in assessing the optimal future structure of energy production and prices taking into account gains and costs referred to the energy production. It requires identifying and assessing microeconomic components of social benefits and costs of energy production. The model is precisely scaled and its base scenario reflects the actual performance of the Polish energy sector. Furthermore, the results of the base scenario are consistent with the official mid-term energy policy goals approved by the government (Energy Policy Assumptions, 2002). In order to examine the effects of internalisation of external costs the relevant negative impact estimates have been taken from the ExternE project (European Commission, 1998; Krewitt, 2002). The volumes of externalities were adjusted by the PPP ratio to keep the analysis more precise with respect to Polish economic conditions. The structure of the paper is as follows. In the first section the Polish energy system indicators are presented. The primary energy production structure, the main energy producers and emissions characteristics are discussed. The second part concerns the description of the model. Its structure and links within the model, main equations, constraints, variables, technological, economic and emissions data used in the model are described. The sensitivity analysis and the main scenarios of the model are presented as well. The third part of the paper focuses on the core results of the model. Since appropriate calculations have been carried out using two major criteria of effective allocation of resources, the concluding remarks are formulated according to such a scheme. The paper ends with the conclusions and recommendations regarding the optimal structure of energy production in Poland, possible electricity and heat prices, the level of external costs related to energy production in Poland and the volume of social welfare referred to each scenario.

2. The Polish energy system—the main indicators The energy sector is a key branch of the Polish economy. Its share in GDP was 7.0–5.5% in 1995–2000 (Jankowski et al., 2002). Poland’s accession to the EU creates new conditions and challenges facing the Polish energy sector. The most important issues seem to be a diversification of energy supplies (a gradual increase of renewables for electricity production), improvement of energy productivity and substantial emissions reduction. These challenges are likely to be very hard to meet due

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

to the fact that 97% of electricity in Poland is produced from fossil fuels. Domestic sources of primary energy dominate total supply. However, virtually all consumed oil and half of the natural gas are imported. Despite the sharp downward shift of hard coal production over the past 8 years—from 81 in 1992 to 60.4 Mtoe in 2000 (EIC, 2002)—it is by no means the main primary energy source in Poland. Domestic reserves that amount to 11 billion tons are likely to be sufficient for about 60 years at the current rate of exploitation. Production and consumption of lignite are rather stable (12.5–13.5 Mtoe) as a result of constant demand from lignite power stations. Non-fossil energy sources are marginal and likely to remain so. Renewable sources still occupy a small position in the Polish energy balance (4–5 Mtoe) (EIC, 2002). The Polish electricity sector consists of 15 large public power plants and 30 public CHP plants. They are mostly state owned companies. The Polish Power Grid Company monopolises high-voltage transmission services. The district heat sector is more decentralised and is characterised by companies owned generally by local authorities. The majority of industrial plants in Poland receive their heat and electricity supply from their own boilers and generators. The Polish hard coal sector is organised in four coal companies—each of them owning and managing a group of hard coal mines. It is now undergoing a restructuring process aimed at reducing costs and capacity. The transformation of the Polish economy had a positive effect on the environment and air quality. Initially a very high level of sulphur dioxide emissions (3.2 Mtons in 1990) was reduced by approximately two times by the year 2000 (1.5 Mtons) (CSO, 2001). All sources of emissions contributed to such significant reduction of SO2 emissions, but the highest reduction is attributed to the energy sector. However, the present level of sulphur emissions is still regarded as very high, mainly owing to Poland’s heavy reliance on coal combustion. The level of NOx emissions from the energy sector—0.8 Mtons in 2000—has been rather stable in recent years owing to technological constraints (CSO, 2001). The greatest progress in emissions reduction has been observed in the case of particulates. The total volume of TSP emissions dropped by over two times—from 1.95 Mtons in 1990 to 0.8 Mtons in 2000 (CSO, 2001). This sharp fall in TSP emissions is caused by the relatively low cost of abatement equipment. Nevertheless, the bulk of air pollution in Poland is still caused by the energy sector. Public power plants generated 53% of Poland’s total SO2 emissions, 28% of NOx emissions and 11% of particulates in 2000. Including industrial CHP plants appropriate values equal 71%, 38% and 61%, respectively (CSO, 2001). Poland produces a significant amount of greenhouse gases. Unlike other air pollutants, CO2 emissions

3411

stabilised in the mid-1990s at the level of 360–370 Mtons and fell to 315 Mtons by 2000. Fuels combustion remains the major source of CO2 emissions. In the year 2000 nearly 97% of CO2 emissions (302 Mtons) originated from these processes. The energy sector, which is the main consumer of solid fuels, is responsible for 56% of CO2 emissions (176 Mtons) (CSO, 2001). It is worth noting that Poland will probably meet the Kyoto emission target simply by continuing the current energy and environmental policy. Moreover, different studies (i.e. GCSS, 2000) demonstrate that by 2010 GHG emissions could be below the Kyoto limit by up to approximately 70 Mtons of CO2. Thus a certain amount of CO2 emissions may be the subject of international emission trading based on the flexible instruments of the Kyoto Protocol. The improvement of air quality is one of the priorities of the Polish environmental policy. Poland approved and brought into force new regulations on air protection in 2001, which impose a new method of defining obligations of companies emitting pollutants, including the requirement of having integrated pollution prevention and control (IPPC) permits by the largest plants. Poland also incorporated new directives in its legislation (2001/77/EC, 2001/80/EC, 2001/81/EC) connected directly and indirectly with the issues of air protection. The consistent implementation of obligations on air protection by Poland is also required in the case of two international agreements: the Convention on Longrange Transboundary Air Pollution (UNECE, 1979) and the UN Framework Convention on Climate Change (UNFCCC, 1992).

3. The model The tool presented here is the dynamic partial equilibrium model of the mid-term development of the Polish power sector. The model focuses on detailed issues related to energy production capabilities, the electricity and heat markets, without capturing other macro-economic links. It equilibrates prices and volumes of electricity and heat production taking into account external costs related to emissions generated by energy technologies. The demand for final energy is estimated on the basis of market relations, i.e. price and income elasticities. In this representation the buyer of energy consumes electricity and heat up to the point where his marginal willingness to pay equals the marginal cost of production. Hence consumers maximise their discounted surpluses and the behaviour of the producers is modelled as a profit maximising firms in the energy market. The markets will be in equilibrium if the activities of different ‘‘agents’’ are compatible. This means that the total demand for electricity and heat equals their supplies. Unlike in other approaches

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

3412

environmental constraints in the form of emissions limits are not imposed here. In fact one of the most useful features of that tool is the assignment of an optimal emissions level on the base of cost-benefit analysis. The structure of the Polish energy sector is characterised by the typical interactions between supply and demand for primary and final energy carriers (Fig. 1). The supply side considers possibilities to deliver energy carriers from domestic and import sources and their conversion through the energy processes. Public power plants, public CHP plants, industry CHP plants and municipal heat plants are the main producers of energy represented in the model. The technology variables are electricity and heat production and the level of environmental investments. It was impossible to consider each individual power generation plant separately because of computational limits. Given this the main ‘‘agents’’ of the model are energy generation technologies, which are divided into three groups: existing, modernised and new plants. The first group represents typical Polish lignite or hard coal firing plants. They can be refurbished through the following options: simple modernisation, simple

modernisation with gas turbine and FBC boiler investment. New technologies include coal gasification, natural gas, oil, biomass, hydropower, solar power, etc. An interesting feature of the model is the fact that a relatively large mix of clean coal technologies and modification options are considered. It is assumed that these technologies can be regarded as an efficient option for the future development of the Polish energy sector— keeping in mind a high reliance on domestic coal reserves as the most probable alternative for energy generation. To assure flexibility in emissions reduction different efficient pollution control technologies are allowed. Post-combustion technologies represent wet scrubbers, spray dry scrubbers, sorbent injection processes (for SO2 control) and low NOx burners, selective catalytic reduction (SCR) and selective non-catalytic reduction (SNCR) (for NOx control). For computational reasons pre-combustion strategies, i.e. coal preparation and coal switching, were not employed. The demand side is represented by the main electricity and heat consumers, i.e. industry and construction, transport, agriculture, trade and services, individual

Private costs Type of fuel.

Fuel costs

Social welfare Fixed and variable costs of technol. prod.

Investment costs of technol. prod.

Investment costs of abatement technol.

Fixed and variable costs of abatement technol.

Balance of Import and export costs

Consumerss and producers ers surplus

External costs

Source of supply

Domestic

Technology efficiency

Import

Capacity of supply

Balance of fuels supplies

Modernization of technologiies

Production investments

Fuel consump. rate

Existing technologies

Energy balance of production

Fuels supplies

Balance of production investments

Capacity

Demand sectors

Electricity/heat at ratio

Energy production

Transport losses

Load periods

Demands ratio in load periods

Balance of production and demand for final energy

Import

Balance of production capacity

Consumers demand

Energy price

New technologies Availability factor

Abatement technologies

Environ. investments

Balance of environ. investments

Capacity

Balance of abatement capacity

Emissions reduction

Balance of emissions reduction

Symbols: Balances

Variables

Parameters

Costs component

Efficiency of abatement technologies

Fig. 1. Structure of the model.

Technology efficiency

Export Export

Emissions factor

Demand functions

Price Price elasticity

Income elasticity

Balance of emissions

Emissions

Unit external costs

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

consumers and export. The demand curves were estimated by appropriate price and income elasticity coefficients, both for electricity and heat markets. It is important to account that in the electricity and heat energy models the year consists usually of both peak and off-peak periods of energy demand in summer and winter. That is why the year is divided into four different load periods. Regarding the international trade of electricity it is assumed that electricity can be bought or sold at the global price according to the limited transmission capacity. The model framework presented in Fig. 1 indicates the main equations, variables, parameters and objective functions within the model. The most important equations are: balance of fuels supplies, balance of energy production and balance of demand for final energy. The remaining relationships regard technological constraints (balances of production investments and capacity, balances of environmental investments and capacity) and emissions equations (balances of emissions and capacity reduction). The major variables in the model refer to fuel supplies, electricity and heat production, consumers demand, energy prices, production capacity, environmental investments and emissions level. A full numerical specification of the model comprises technological, economical and environmental parameters. Relevant parameters, given exogenously, are associated with the capacity of fuel supplies and their prices, efficiency of energy technologies, price and income elasticities, emission rate of energy technologies, unit level of external costs, etc. The potential of fuels supplies and their prices were based on formal estimations published in domestic sources (Energy Policy Assumptions, 2000, 2002; GCSS, 2000), foreign documents (EC, 1999) and own estimations. The crucial assumption refers to coal data and revisions of the potential of geothermal energy use for heat production on a local scale in Poland. This is also the case for the capacities of other renewable sources. Fuel prices are assumed to increase slightly and stably (generally at 1% per year), except imported steam coal for which the price stabilisation is allowed (EC, 1999). In addition to that, as a result of the restructuring process of the Polish coal sector, domestic coal prices are not likely to change over the coming years. The cost components incorporated in the objective function, including fuel costs, investments, the fixed and variable cost of production and emissions abatement technologies, all create the private costs of energy production. The model is dynamic in the sense that the electricity market equilibrium is modelled for the mid-term horizon 2002–2020. Relevant production data is presented in Table 1. Energy technologies involve the typical production characteristics of existing coal fired power plants and

3413

projected to be implemented in Poland. Due to a strong opposition nuclear reactors are not allowed to operate for the next years. It was assumed that as a result of the modernisation of existing power plants their efficiency should increase in the range of 2–4%. Each of the technologies is characterised by different cogeneration and availability factors. First one is the ratio of electricity production to the total energy production (including heat) and second one is the production time at full capacity during 1 year. For Polish public power plants over 90% of total production is electricity. Public and industrial CHP plants and local heating boilers produce mainly heat, both for industrial processes and central heating. The depreciation rate of existing technologies was assumed to be 4% for public power and CHP plants and 6% for other units. A set of typical modernisation options for existing units is allowed with respect to the specific capacity range. Thus an increase in total capacity due to the employment of modernisation options cannot exceed 5% of their primary capacity. Such a range of capacity extension seems reasonable for the technological and location constraints of existing power plants. Transport losses are assessed as 10% of total energy production (EMA, 2002a, b). Fixed and variable costs of existing energy technologies are based on data published in the official documents (EMA, 2002a, b). The investment costs of new energy representative technologies, e.g. integrated gasification combined cycle (IGCC), steam and gas turbines (STAG), gas and wind turbines, biomass boilers, etc. are based on data published both in domestic (e.g. Jankowski, 1997; Kamrat and Augusiak, 1997; Radovic, 1997) and foreign sources (e.g. OECD/ IEA, 1997). Emissions factors associated with energy technologies were estimated with respect to unpublished emissions data set of Polish units combusting both steam coal and lignite. It was assumed that new coal technologies in principle are equipped with FGD and low NOx burners with 95% and 50% emissions reduction ratios, respectively. Damages related to energy technologies are derived from the ExternE estimations (EC, 1998). For the analytical reasons external costs estimates are attributed to the unit emissions of SO2, NOx, TSP and CO2, which means that externalities of non-fossil fuels technologies, e.g. wind turbines, nuclear and hydro plants are virtually underestimated. In addition not all relevant external effects of each phase of energy production, including those caused by the production and transport of fuels used, are tracked down. Moreover, ExternE estimations do not include damages related to forests and ecosystems and in case of other components (health, crops and material damages) are subject of serious controversy. Consequently data used in this research only in limited scope express external costs generated by energy technologies.

2275 2275 2275 2275

73

75

75

75

0.91

0.96

0.27

0.27

0.97 0.69 0.69 0.96 0.57 1 1 1

6500 6500 6500 3500 3500 6500 3500 3500

Public CHP plants Existing hard coal CHP public plants—life extension CHP hard coal public plants— simple modern CHP hard coal public plants— simple modern+gas turbine CHP hard coal public plants— FBC

0.91

6500

1

0.97

New hard coal public power 41 plants New lignite public power plants 41 New IGGC (hard coal) 47 New IGGC (lignite) 44 New gas turbines 35 New STAG power plants 55 New nuclear power plants 100 New hydro power plants 100 New wind turbines 100

6128

41

0.96

1989

6128

42

0.97

0.97

0.91

0.96

0.91

0.91

0

0

0

24946

0 0 0 0 0 0 0 0

0

2186

0

0

0

9752

0

0

0

19082

Cogeneration Capacity factor installed (MW)

72

Existing hydro power plants

6128

4055

41

38

4055

42

5600

4055

39

35.5

3600

Availability factor (h)

36

Efficiency (%)

5740

713

410

0

5300 5500 6000 1200 2870 9000 8200 6500

4800

0

5740

713

410

0

5740

713

410

0

100

100

100

100

120 120 120 75 120 150 86 100

120

86

200

200

200

200

150

150

150

150

Investment Fixed costscosts (zl/kW) electr. (zl/ kW)

50

50

50

50

15 15 15 15 15 0 0 0

15

0

20

20

20

20

20

20

20

20

3.0

3.0

3.0

3.0

3.0 3.0 3.0 0.2 0.2 3.0 3.0 3.0

3.0

3.0

3.0

3.0

3.0

3.0

3.0

3.0

3.0

3.0

Fixed costs- Variable heat (zl/kW) costs-electr. (zl/kW)

1.9

1.9

1.9

1.9

1.9 1.9 1.9 0.1 0.1 0 0 0

1.9

0

1.9

1.9

1.9

1.9

1.9

1.9

1.9

1.9

Variable costs-heat (zl/kW)

70.0

380.5

475.3

475.3

160.0 4.3 6.5 0.9 0.9 0 0 0

80.0

0

158.0

531.5

664.0

664.0

65.0

345.9

432.0

432.0

Emissions factor SO2 (g/GJ)

75.0

144.1

174.4

174.4

74.0 89.0 67.9 27.8 27.8 0 0 0

90.0

0

50.0

115.0

138.0

138.0

76.0

147.0

178.0

178.0

Emissions factor NOx (g/GJ)

61.4

49.1

61.4

61.4

36.2 38.2 32.6 0 0 0 0 0

42.0

0

36.2

29.0

36.2

36.2

42.0

33.6

42.0

42.0

Emissions factor TSP (g/GJ)

94.0

86.2

94.0

94.0

103.0 84.5 92.5 55.0 55.0 0 0 0

94.0

0

103.0

93.4

103.0

103.0

94.0

86.2

94.0

94.0

Emissions factor CO2 (g/GJ)

3414

Existing lignite public power plants—life extension Lignite public power plants— simple modern Lignite public power plants— simple modern+gas turbine Lignite public power plants— FBC

Public power pants Existing hard coal public power plants—life extension Hard coal public power plants— simple modern Hard coal public power plants— simple modern+gas turbine Hard coal public power plants— FBC

Technology

Table 1 Power plants characteristics

ARTICLE IN PRESS

M. Kudelko / Energy Policy 34 (2006) 3409–3422

hard coal heat plants fuel oil heat plants gas heat plants biomass heat plants geothermal heat plants solar heat collectors

83 55 35 100 100 100

4000 4000 4000 4000 4000 4000

3500 3500 0 0 0 0 0 0

0 0

0

0

0

0

0.96 0.96

0.135

0.135

0.135

0.96

0.135

0.135

0.96 0.57 0.96

0.27

Conversion factor: 1 zl ¼ 4.1h. Source: Own estimations based on Polish statistical sources.

New New New New New New

55 35

3000

3500 3500

35 55

plants— 80

3500

48

3000

3500

45

plants— 80

3500

35

3000

3500

67

plants— 80

3500

67

1737

3500

3500 3500 3500

35 55 45

65

3500

75

plants— 75.8

Existing fuel oil heat plants Existing gas heat plants

Municipal heat plants Existing hard coal heat life extension Existing hard coal heat simple modern Existing hard coal heat gas boilers Existing hard coal heat FBC

Industry CHP plants Existing hard coal industry CHP plants—life extension Existing hard coal industry CHP plants—simple modern Existing hard coal industry CHP plants—simple modern+gas turbine Existing gas industry CHP plants Existing fuel oil industry CHP plants New hard coal industry CHP plants New CHP gas turbines New CHP fuel oil turbines

New CHP hard coal public plants New gas turbines New STAG power plants New fuel oil turbines

0 0 0 0 0 0

0 0

0

0

0

35274

0 0

0

0

308

0

0

17350

0 0 0

0

4920 4100 1200 4000 4000 20500

0 0

5740

713

410

0

1200 4100

4920

0

0

713

410

0

1200 2870 4100

4920

0 0 0 0 0 0

0 0

0

0

0

0

70 70

70

70

75

250

250

250

75 70 70

70

100 120 100 100 100 100

50 50

40

40

40

40

30 30

30

30

30

30

30

30

32 32 32

50

0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.0

0.0

0.0

0.0

0.0

0.2 1.5

3.0

1.5

0.2

3.0

3.0

3.0

0.2 0.2 1.5

3.0

3.0 1.0 0.1 3.0 3.0 3.0

1.0 0.1

5.0

5.0

5.0

5.0

0.1 1.0

1.9

1.0

0.1

5.0

5.0

5.0

0.1 0.1 1.0

1.9

230.0 135.7 0.9 1.5 0 0

135.7 0.9

70.0

853.1

1066.7

1066.7

0.9 135.7

230.0

137.8

0.9

677.9

847.5

847.5

0.9 0.9 137.8

230.0

87.2 48.9 27.8 64.8 0 0

48.9 27.8

75.0

430.2

532.2

532.2

27.8 48.9

87.2

49.0

27.8

216.6

265.2

265.2

27.8 27.8 49.0

87.2

42.4 57.1 0 3.7 0 0

57.1 0

61.4

674.4

843.7

843.7

0 57.1

42.4

57.1

0

336.0

420.0

420.0

0 0 57.1

42.4

94.0 74.1 55.0 4.4 0 0

74.1 55.0

94.0

86.2

94.0

94.0

55.0 74.1

94.0

74.1

55.0

86.2

94.0

94.0

55.0 55.0 74.1

94.0

ARTICLE IN PRESS

M. Kudelko / Energy Policy 34 (2006) 3409–3422 3415

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

3416

Figures used in the model represent the average level of external costs aggregated for 15 EU countries. The appropriate volumes of damages caused by air emissions have been assessed at 600 Euro/Mg for SO2, 5000 Euro/ Mg for NOx and 13,000 Euro/Mg for TSP. In order to adjust these values to the Polish economic conditions they were converted based on the PPP ratio equals 0.38. Unfortunately for global CO2 damages, a high degree of uncertainty is observed. The range of global warming damage in the ExternE project differs from 3.8 to 139 USD/Mg for CO2 (Krewitt, 2002). Other studies (for example Frankhauser, 1994) estimates these damages at a lower level of 20 USD/MgC, which means 5.5 USD/ Mg for CO2. At present there are no single estimates for externalities that reflect the best understanding of this category of damages. For that study rather the low value of 8 USD/Mg CO2 was used. The driving force of the model is the growth rate of the country’s economy. An annual growth rate in income of 3% is assumed over the time horizon, which is a little bit below the GDP growth rate estimations (Energy Policy Assumptions, 2002). The growth of the demand in the domestic energy markets is determined by the income elasticity, which is assumed to be 0.7 for electricity and 0.5 for heat (Andersson and Haden, 1997). The lack of reliable data makes it difficult to determine the actual price reactions. Therefore, relevant price elasticities are roughly set at the level of
0.25 for electricity and
0.3 for heat consumption (Kudelko and Suwa"a, 1998; Bates et al., 1994; Bose and Shukla, 1999). The sensitivity analysis reveals (Figs. 2 and 3) that these uncertain parameters have a crucial influence on model’s results. The relevant computations were carried out for three key parameters: price elasticity of demand for energy, level of external costs and income growth. Their range changed from 80% to 120% of the original values. The change of the objective function indicates the ‘‘economic’’ reaction of the energy system and total emissions of CO2 provides information on technological (structural) adjustments. 450000 400000 350000 300000 250000 200000 70

80

90

Price elasticity

100

110

120

130

Income growth

External costs

Fig. 2. The sensitivity analysis—the change of objective function, millions zl.

1050000 1000000 950000 900000 850000 800000 70

80

90 Price elasticity

100

110

120

130

Income growth

External costs

Fig. 3. The sensitivity analysis—the change of CO2 emissions, 000 Mg.

Price elasticity of demand for energy seems to be the decisive parameter on model’s results since it directly influences the value of consumers’ surplus. It also means that its incorrect estimation could bring serious structural changes in energy production, measured by the CO2 emissions indicator (Fig. 3). This is also the case of income growth, but the direction of changes is opposite and the intensity of relation is lower. Furthermore very important for relevance of the results is right estimation of external costs. The level of uncertainty in that case is very high and even their relatively small variation showed here could influence the energy structure in Poland. The range of CO2 emissions as a main indicator of structural changes varies here only by about 10% of the total, but more serious effects are probable if greenhouse gas effects would be greater than assumed in this work. Such attempt has been done by doubling the value of CO2 damages assumed in this study. As a result the structural changes are substantial and low carbon technologies are most preferable. What is more limited possibilities of fuels supplies enforces also rapid growth of nuclear power plants as the only and the cheapest way to satisfy the rising demand for energy. Of course in that case the welfare negative effects are also essential. The computations were carried out in constant prices of 2002. Thus the model was calibrated to the 2002 level of electricity and heat production and costs with 5-year time intervals. A discount rate of 10% is used in the model. The main reason for such a high value is the fact that the Polish economy is still perceived as relatively uncertain and unstable. Scenarios applied in the model reflect possible results in terms of the objective function used, demand specification, emissions of main air pollutants and volume of external costs estimates. The category and value of social welfare is the main feature that differs the solutions obtained for each scenario. The key elements for scenario creation are summarised in Table 2.

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

3417

Table 2 Scenarios of the model

Type of model Objective function

Demand Emissions External costs

Variant 1

Linear Minimisation of energy production costs Determined Limited Not considered

Variant 2 Scenario 1

Scenario 2

Scenario 3

Non-linear Maximisation of consumers’ and producers’ surpluses Calculated Calculated Not considered

Non-linear Maximisation of consumers’ and producers’ surpluses less external costs Calculated Calculated Partly considered (only SO2, NOx, and TSP)

Non-linear Consumers’ and producers’ surpluses less external costs

The reference scenario—variant 1—is characterised by the objective function in the form of minimisation of discounted cost of energy production. In this costeffective scenario demand for electricity and heat is determined and emissions limits are exogenously given. The reference solution has one important advantage. This case is mainly intended to demonstrate the optimal least-cost structure of energy production according to the environmental programs approved by the government. It must be stressed that the results of the reference scenario are more or less consistent with the official forecasts of the Polish energy policy. Given this the results can be compared with the more socially oriented policies embodied in the next scenarios of variant 2. In contrast to the reference scenario demand for electricity and heat in variant 2 is indigenously determined on the base of maximisation of the social welfare criterion. The level of emissions is computed depending on the intensity and structure of energy production. Three scenarios included in variant 2 differ according to the range of externalities considered in the decision-making process. The presented model was run on the GAMS software package (Brook et al., 1992). Two types of solvers were used for computations. The first solver—CPLEX—was used to solve the linear programming task for variant 1 and the second one—CONOPT—to solve the quadratic programming (variant 2). A standard run of all scenarios takes approximately 45 min to solve.

4. Results The results exemplify different solutions in terms of energy production structure, costs of electricity and heat production, major technological and environmental investments and private and social welfare. Results of the model reveal that the type of criterion used in the calculations has a significant influence on energy production in Poland. It is not surprising that the energy production structure in the cost-effective alloca-

calCulated Calculated Fully considered (SO2, NOx, CO2 and TSP)

1400 1200 Production [PJ]

Specification

1000 800 600 400 200 0 2002

2003

2004

2005

Hard coal

Lignite

Renewables

Other

2010

2015

2020

Natural gas

Fig. 4. Structure of total energy production—variant 1.

tion scenario should be dominated by relatively cheap energy conversion technologies that are generally based on domestic solid fuels—hard coal and lignite (Fig. 4). Existing coal fired power plants and their modernisation options are economically competitive alternative to cover a slightly increasing demand for electricity. As a result of diminishing domestic coal reserves, stricter emissions standards, the possible increase of gas use by industry and municipal heat plants and a promotion of renewable sources, a slight fall in hard coal use should be observed after 2005. But in fact hard coal and lignite seem to be still preferred as cheap energy sources for public power and CHP plants. As demand for electricity grows, new natural gas-fired plants and renewables will come into operation after 2015. Given this, a relatively high rise in the cost of electricity and heat production (namely 30% and 66% in 2020, respectively) is expected. The energy costs are likely to change also due to environmental investments needed after 2005. Most preferred and competitive future investments options seem to be modernisations of existing power plants. Old coal fired boilers will be gradually replaced by new ones. More than 20% of the total installed generation capacity (which equals 108 GW) is predicted to be replaced in the Polish energy system since 2010. Due to

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

3418 Table 3 Summary of variant 1 Specification

2002

2003

2004

2005

2010

2015

2020

Production, PJ: Electricity Heat Total Unit cost of electricity production (zl/MWh) Unit cost of heat production (zl/GJ) New installed generation capacity (MW) New installed abatement capacity (MW)a

513 572 1085 112.7 20.6 0 0

518 567 1085 113.5 21.2 3600 0

518 559 1077 114.7 21.7 3420 0

525 554 1079 116.5 22.3 4185 12,145

540 552 1092 120.8 25.3 20,806 1790

656 551 1207 135.6 31.8 37,555 26,905

762 551 1313 146.4 34.3 23,084 6344

a

New installed abatement capacity expressed in MW means the level of investments related to the heat power of existing and new energy boilers.

1400

Production [PJ]

1200 1000 800 600 400 200 0 2002

2003

2004

2005

Hard coal

Lignite

Renewables

Other

2010

2015

2020

Natural gas

Fig. 5. Structure of total energy production—variant 2 scenario 1.

stricter environmental standards (LCP Directive) high progress in abatement investments is also expected after 2005 and 2015. The high rate of use of solid fuels gives relatively high emissions of air pollutants. Therefore, the level of external costs in this scenario is at the same order of magnitude as private costs of energy production. Table 3 shows the estimated level of electricity and heat production, costs of energy production and level of investments in variant 1. A change in the objective function of the model forces serious technological and economic adjustments. In variant 2 and scenario 1, which is based on the maximisation of private welfare, these changes are relatively small, compared to the results of variant 1 (see Fig. 5). We can observe some similarities as regards the level and structure of energy production. This seems reasonable since the model was calibrated for two types of objective functions. Nevertheless few changes are expected mainly because of different economic assumptions adopted for both variants. The crucial point is that the demand forecasts published in the official governmental documents and used in variant 1 were rather overestimated and are now subject to updating. Unlike this rather stable and slow economic growth was considered in variant 2, which leads to slight discrepan-

cies between both variants. Given this the energy production is assumed to be more stable after 2015 in scenario 1 than in variant 1. It practically preserves a dominant position of coal technologies and makes impossible the broaden use of renewables in energy generation. Renewables seem to be still an uncompetitive option for developing the Polish energy sector, at least for the predicted level of energy demand. In addition to that, the cost of electricity production is much higher than in variant 1. It rises from 120 zl/MWh in 2002 to 228 zl/MWh in 2020. It must be stressed that in the model the market equilibrium is reached by the marginal energy producer (technology). This implies in turn that the energy price has to be equal to the marginal cost of energy production. It explicitly clarifies the main reason of that sharp price growth. However, it must be carefully considered and the final price of energy does not have to reflect the market clearing rule. It is possible to apply different solutions and the regulator of the energy system is responsible for setting up the appropriate price system. The possible price setting options are the marginal (as it set in this scenario) and average price rule. The range of new installed generation capacity is similar to the previous variant, but some changes can be observed in case of new abatement capacity, which here are not necessary. Values of the main variables of scenario 1 are presented in Table 4. The inclusion of external costs in the companies’ economic decisions affects the substantial changes in the production level and its structure. The greater the range of externalities, the more extensive adjustments are expected (Figs. 6 and 7). For both scenarios we can observe a significant drop in the energy production compared to variant 1 and scenario 1. The decline of electricity and heat production in scenario 2 is approximately equal to 23% in 2002 and 17% in 2020, respectively. In scenario 3, where global warming externalities are quantified, the foreseen reduction of the energy production is even greater and equals 29% in 2002 and 17% in 2020. Except for such a huge decrease of production, which is in fact a market response for the

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

3419

Table 4 Summary of variant 2, scenario 1 Specification

2002

2003

2004

2005

2010

2015

2020

Production, PJ: Electricity Heat Total Unit cost of electricity production (zl/MWh) Unit cost of heat production (zl/GJ) New installed generation capacity (MW) New installed abatement capacity (MW)

515 573 1088 119.6 22.8 0 0

505 550 1055 140.3 27.0 1480 0

507 536 1043 147.4 30.0 3227 0

517 536 1053 148.7 31.2 4810 0

536 523 1059 180.7 38.8 21,081 0

566 539 1105 204.6 42.4 23,009 0

593 550 1143 227.8 46.7 19,959 0

1400

Production [PJ]

1200 1000 800 600 400 200 0 2002

2003

2004

2005

Hard coal

Lignite

Renewables

Other

2010

2015

2020

Natural gas

Fig. 6. Structure of total energy production—variant 2 scenario 2.

1400

Production [PJ]

1200 1000 800 600 400 200 0 2002

2003

2004

2005

Hard coal

Lignite

Renewables

Other

2010

2015

2020

Natural gas

Fig. 7. Structure of total energy production—variant 2 scenario 3.

reduced demand for electricity and heat due to the higher prices, significant changes are seen in the structure of the energy production. A dominant position of solid fuels is expected to decrease in favour of gas and renewables. The use of hard coal, the primary energy fuel in the Polish energy balance, is declining by about 30–50% in comparison with variant 1. The major reason that coal is squeezed out by other fuels is its high level of generated external costs. However, which seems to be reasonable, this is not the case for lignite which is

expected to be a very competitive source of energy production. In that case the cost of electricity production is about 30% below the cost of electricity generated in hard coal-fired plants. Furthermore these plants are generally equipped with new FGD and DENOX installations which means a relatively low level of externalities produced. In addition to that the low volume of external costs generated by gas technologies and renewables leads to their growing use. Hence, it is economically reasonable that gas should be perceived as the main fuel source for electricity generation in 2020. For both scenarios renewables are virtually used at the maximum capacity. Public power plants, CHP plants and new small heating plants are expected to dominate existing industry and municipal heating plants, mainly due to a relatively lower negative influence on the environment. However, various technological and market conditions not incorporated in the model make it difficult to determine the right level of energy produced by industry and municipal heating plants. Nevertheless, the results of the model suggest that energy supplies for the industrial processes could be assured by public power plants and local heating plants using geothermal, biomass and wind energy. It is worth stressing that a comparatively high level of energy in Poland comes from public CHP plants. That is why high efficiency and low production costs create good conditions for competitive future use of CHP plants, even taking into account their high reliance on hard coal supplies. All abatement options are economically efficient with respect to the unit external costs estimates employed in the model. Therefore, the high level of externalities causes that all abatement technologies should be applied as much as possible. However, in case of CO2, the best way to lower the global warming effects is to foster the use of alternative energy technologies or reduce the level of production. The external costs included in the decision-making process lead to the sharp growth of energy prices. The predicted rise in energy prices is about twice as high compared with their present level. Consequently, the lower demand and common use of energy technologies

ARTICLE IN PRESS 3420

M. Kudelko / Energy Policy 34 (2006) 3409–3422

The theory of the welfare economy and the results of the model suggest that from a social point of view the internalisation of external costs by the energy prices increase renders this scenario the most desired among others. The summary results of the model are illustrated in Table 7 where the volumes of private and social welfare in discounted terms are shown. Private welfare in the micro-economic sense is defined as a sum of consumers’ and producers’ surpluses. Consumers’ surplus measures the difference between the actual price

that have a less negative impact on the environment substantially reduce the level of emissions. For SO2 and NOx the predicted emissions reduction is about 2–6 times below their present levels. The technological constraints are surely the main barrier for the relatively low decline of CO2 emissions that equals approximately 20–30%, depending on the scenario. The predicted level of electricity and heat production, energy production costs and level of investments in scenarios 2 and 3 are presented in Tables 5 and 6. Table 5 Summary of variant 2, scenario 2 Specification

2002

2003

2004

2005

2010

2015

2020

Production, PJ: Electricity Heat Total Unit cost of electricity production (zl/MWh) Unit cost of heat production (zl/GJ) New installed generation capacity (MW) New installed abatement capacity (MW)

485 349 834 148.1 52.7 0 119,834

490 363 853 154.2 52.2 2918 3109

488 369 857 165.7 52.2 2695 2666

492 380 872 172.5 52.2 3336 5812

523 428 951 193.6 51.4 20,166 39,133

588 426 1014 183.4 57.4 19,878 31,359

643 444 1087 180.3 60.8 19,789 24,585

Table 6 Summary of variant 2, scenario 3 Specification

2002

2003

2004

2005

2010

2015

2020

Production, PJ: Electricity Heat Total Unit cost of electricity production (l/MWh) Unit cost of heat production (zl/GJ) New installed generation capacity (MW) New installed abatement capacity (MW)

448 324 772 183.6 56.1 0 108,534

457 337 794 186.2 55.6 3091 3464

462 347 809 190.8 55.2 3457 3441

461 360 821 202.7 54.7 5234 2899

512 398 910 205.0 55.5 22,715 13,550

558 442 1000 211.6 55.3 18,237 21,431

605 480 1085 215.9 55.9 17,151 28,610

Table 7 Private and social welfare [billion zl] Specification

Variants Variant 1

Variant 2 Scenario 1

Scenario 2

Scenario 3

Consumers’ surplus Producers’ surplus Private costs External costs, including: SO2 NOx CO2 TSP

— — 358 265 97 33 63 72

547 109 311 285 115 33 64 73

478 121 318 139 44 15 54 26

442 134 325 114 37 12 48 17

Private welfare Social welfare

— —

656 371

599 460

576 462

ARTICLE IN PRESS M. Kudelko / Energy Policy 34 (2006) 3409–3422

received and the price at which producers are willing to supply energy. Producers’ surplus equals the private profits of energy producers. Social welfare is calculated as the difference between private welfare and external costs generated by the main air pollutants. Private costs are derived from fixed, variable and investment cost of energy and abatement technologies. Variant 1 with least-cost approach is also presented. The point of reference is scenario 1 where the prices of energy producers reflect only their generation and abatement costs. From the social point of view this case should not be preferred, although private welfare is the highest among the considered scenarios and equals 656 billion zl. This is also the case of the smallest private costs which equal 311 billion zl. Even though the total electricity and heat consumption in scenario 1 is the highest, which also affects the highest value of consumers’ surplus, the structure of energy production and the level of emissions result in a high level of externalities coming from air pollutants (285 billion zl). Consequently, social welfare amounts to 371 billion zl and it is smaller than in other scenarios. SO2 emission is predicted to be the most harmful with 115 billion zl of damages in total. As expected, social welfare increases significantly due to the internalisation of external costs. Moreover, the greater range of externalities considered the greater the social welfare improvement. Given this and considering scenario 2, where global warming damages are not included in the objective function, we can see that social welfare growth is considerable and equals 24% (460 billion zl). It is very similar to the results of scenario 3 where externalities caused by all pollutants are considered. The reduced demand for electricity and heat is the source of the distributional impacts and change of private welfare. The higher prices of energy enforce the loss of the consumers’ surplus that is smaller with respect to the previous scenario. The losses are predicted at 8% for scenario 2 and 12% for scenario 3. However, the economic condition of producers, expressed in terms of private profits, is improving mainly for the higher prices of energy. The significant change in the energy production structure, the high level of abatement technologies and reduced demand for electricity and heat lead to a sharp reduction of air pollution. Consequently, the total level of externalities for both scenarios decreases by about 51% and 60%, respectively.

5. Conclusions Different modelling technologies, i.e. optimisation, macro-economic (general equilibrium simulation) and system dynamics have been developed and extensively used for the analysis of different kinds of energy–eco-

3421

nomy–environment policies. In this paper the partial equilibrium model for the Polish energy sector has been presented. The analytical method applied here allows the verification of the mid-term development of the Polish energy sector on the basis of the social welfare criterion with external costs included. This research seems to be the first such extensive attempt to examine the impact of external costs generated by air emissions on the whole energy sector in Poland. The model presented in this paper has many interesting features that make it useful as a reliable planning tool. It comprises the current general performance of electricity and heat technologies used both by the public power plant sector and small industry and heat producers in Poland. All data used in the model was calibrated to prepare a good starting point for further research. Hence, the mix of energy technologies and demand forecast in the reference scenario are consistent with the most probable scenarios of development of the Polish energy policy. Since they do not, as a rule, cover the influence of external costs on the future structure of fuels used, this research can be a good example of assessing the alternatives of domestic energy system developments. However, it must be emphasised that the presented model has some limitations that refer to the type of approach applied for these simulations. The best tools to assess the mid-term macro-economic effects are CGE models with their explicit inclusion of interlinkages between the energy sector and the rest of the economy. Low experience with this kind of simulations and—what is most important—the lack of reliable data for model calibration, are the crucial barriers to employing such an approach for Poland as yet. A general conclusion that can be drawn from this study is that the internalisation of external costs in the decision-making process can improve social welfare measured on the energy market. Since external costs of air emissions are very high the substantial changes in the structure of energy production could be observed. First of all, from a social point of view, the extensive use of solid fuels in Poland should be reduced. In practice the possible strategies to achieve a social welfare improvement are the implementation of low emissions energy technologies, the greater use of abatement technologies and finally the fall in energy production. All of these should be extensively used depending on the scope of externalities considered. All these measures working together substantially decrease the emissions of all air pollutants. The demand for electricity should be met to a large extent by gas and renewables with relatively low emissions coefficients. The position of traditional old coal fired power plants should diminish in favour of new clean coal technologies. However, the public electricity sector is likely to raise the level of electricity and heat production in contrast to the industry power and heat plants. The cost-effective strategy to limit air emissions

ARTICLE IN PRESS 3422

M. Kudelko / Energy Policy 34 (2006) 3409–3422

is the common use of abatement technologies. According to the results of the model, all coal technologies should be equipped with FGD and DENOX installations. All these measures could improve the total social welfare by 24% compared with the expected mid-term development of the Polish energy sector. Given this it seems reasonable to promote these processes in many ways. Increasing attempts at the implementation of the EU environmental directives and instruments in Poland make this policy possible in the nearest future. These conclusions should be carefully considered since it is very difficult to determine the exact implications of possible interventions on the energy market. It must be mentioned here that despite sophisticated techniques employed in that field, a high degree of uncertainties influences the reliability of mid- or longterm projections. This applies both to the different models’ assumptions and exogenous parameters. As the sensitivity analysis revealed, the technological and economical assumptions implemented in the model might be crucial for the results. If we modify key parameters we will get essential changes in the volumes of the objective functions and major variables, i.e. the energy prices and the structure of energy production. Price and income elasticities, values of externalities caused by air pollutants, future fuel prices, the growth rate of the economy, etc. appear to be crucial for this analysis. A more precise evaluation of these parameters could make the research more accurate and useful. References Andersson, B., Haden, E., 1997. Power production and the price of electricity: an analysis of phase-out of Swedish nuclear power. Energy Policy 25 (13). Bates, R., Cofa"a, J., Toman, M., 1994. Alternative policies for the control of air pollution in Poland. World Bank Environment Paper No. 7, Washington. Berger, C., Haurie, A., 1987. Modelling long-range energy technology choices: the MARKAL approach. Technical Paper, GERAD, Montreal. Bigano, A., Proost, S., Van Rompuy, J., 2000. Alternative environmental regulation schemes for the Belgian power generation sector. Environmental & Resource Economics 16 (2). Bose, R.K., Shukla, M., 1999. Elasticies of electricity demand in India. Energy Policy 27 (3). Brook, A., Kendrick, D., Meeraus, A., 1992. GAMS Users’ Guide, Release 2.54. The Scientific Press, San Francisco. Burniaux, J.M., Martin, J.P., Nicoletti, G., 1992. GREEN—a multi sector, multi region dynamic general equilibrium model for quantifying the costs of curbing CO2. Working Paper 104. OECD Department of Economics and Statistics, Paris. CSO (Central Statistical Office), 2001. Environment 2001. Information and Statistical Papers, Warsaw. Energy Information Centre (EIC), 2002. Primary Energy Balances for Poland, Warsaw. Energy Policy Assumptions for Poland until 2020, 2000. Governmental document approved by the Council of Ministers, Warsaw.

Energy Policy Assumptions until 2020, 2002. Revision accepted by the Council of Ministers, Warsaw. European Commission (EC), 1998. Newsletter of the EC study on the externalities of energy. http://externe.jrc.es/nletter6.html. European Commission (EC), 1999. European Union Energy Outlook. Finon, D., 1974. Optimisation model for the French energy sector. Energy Policy 2. Frankhauser, S., 1994. The social costs of greenhouse gas emissions: an expected value approach. The Energy Journal 15 (2). Hille, E., van Leeuwen, N.H., 1993. The Nederland–Poland Task Force on Integrated Energy and Environmental Planning: A computable General Equilibrium Modelling Approach. ECN, Patten. ICF’s Integrated Planning Model, 1999. Reference Manual, Version 7.3 for Personal Computers. ICF Resources Incorporated, Washington. Jankowski, B., 1997. Modelling of the Polish energy system with respect to stabilization of CO2 emissions. IPPT PAN, Warsaw. Jankowski, B., Suwa"a, W., Gaj, H., Kude"ko, M., Kamin´ski, J., 2002. Global Climate Initiatives and Countries in Transition. Case Study for Poland. MEERI Publisher, Krakow. Jorgenson, D.W., Wilcoxen, P.J., 1990. Global change, energy prices and US economic growth. Paper prepared for the Energy Price Hearing, US Department of Energy, Washington. Kamrat, W., Augusiak, A., 1997. Energy cogeneration effectiveness of modern technologies. Symposiums and Conferences 28. MEERI Publisher, Krakow. Kiuila, O., 2003. General equilibrium modeling of sulfur regulations in Poland. International Journal of Environment and Pollution 19 (4) Interscience Enterprises Limited, Geneva, Switzerland. Krewitt, W., 2002. External costs of energy—do the answers match the questions? Looking back at 10 years of ExternE. Energy Policy 30 (10). Kudelko, M., Suwa"a, W., 1998. The Possibilities and Effects of Implementation of Fuel Taxes in Poland. MEERI Publisher, Krakow. Manne, A.S., Richels, R.G., 1990. Global CO2 emissions reductions— the impact of rising energy costs. The Energy Journal 12. Morita, T., Kainuma, M., Harasawa, H., 1996. A guide to the AIM/ ENDUSE Model—technology selection program with linear programming. AIM Interim Paper, National Institute For Environmental Studies, Tsukuba, Japan. OECD/IEA, 1997. Energy Technologies for the 21 Century. Paris. Pandey, R., 2001. Energy policy modelling: agenda for developing countries. Energy Policy 30 (2). Radovic, U., 1997. The competitiveness of energy from renewable sources—present state and prospects. Symposiums and Conferences 28. MEERI Publisher, Krakow. Raskin, P., 1986. LEAP: a Description of the Long-range Energy Alternatives Planning System. The Beijer Institute, Sweden. Rose, A., Benavides, J., Lim, D., Frias, O., 1996. Global warming policy, energy, and the Chinese economy. Resource and Energy Economies 18 (1). The Energy Market Agency (EMA), 2002a. Heat bulletin, Warsaw. The Energy Market Agency (EMA), 2002b. Energy statistics, Warsaw. The Governmental Centre of Strategic Studies (GCSS), 2000. Poland 2025—the Long-term Sustainable Development Strategy. Warsaw. UNECE, 1979. Convention on Long-range Transboundary Air Pollution. UNFCCC (The United Nations Framework Convention on Climate Change), 1992. Report of the Intergovernmental Negotiating Committee for a Framework Convention on Climate Change, New York.