Heat pumps and tradable emission permits: On the carbon dioxide emissions of technologies that cross a tradable emission market boundary

Energy Conversion and Management 47 (2006) 3510–3518

www.elsevier.com/locate/enconman

Heat pumps and tradable emission permits: On thecarbon dioxide emissions of technologies that cross

a tradable emission market boundary

Marcus Eriksson *, Lennart Vamling

Heat and Power Technology, Department of Energy and Environment, Chalmers University of Technology, VOM,

Kemivogen 4, SE 41296 Goteborg, Sweden

Available online 19 April 2006

Abstract

Since January 2005, there is a system with tradable emission permits/allowances in the European Union. Currently,power producers and district heating plants are included in the system, but not the residential sector. In this analyticalstudy, it is discussed how a separation between a trading sector, in which power producers are participating and a non-trading residential sector affect carbon dioxide emissions consequences from heat pumps in households. It is concludedthat a replacement of heat pumps in the residential sector results in a leakage of emissions. The emission target in the trad-ing sector is partly achieved at the expense of increased emissions in the residential sector.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Heat pump; Carbon dioxide emission; Residential heating

1. Introduction

Space heating in households is one of the major energy users and also a major source of carbon dioxide(CO2) emissions in many societies [1]. For this reason, it is of interest to find measures in households thatcan contribute to less energy use and/or less carbon dioxide emissions. However, residential energy systemsare often connected to other energy systems. It might be district heating systems or/and the electric power gen-eration system. In such cases, measures in residential sector energy use, interact with connected energy sys-tems. Depending on the characteristics of the connected energy systems, different conclusions regarding theenergy savings and/or emission reduction potential of measures might be possible. In a Swedish study differentmeasures in a building is presented and compared [2]. In the study, the building is connected to a district heat-ing system, which through combined heat and power is also connected to the power grid. The conclusionsregarding CO2 emissions depend on how the measures and district heating system interact with the power grid.Another study also highlights the interaction between energy systems with respect to the CO2 consequences

0196-8904/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.enconman.2006.03.018

* Corresponding author. Tel.: +46 31 7723011; fax: +46 31 821928.E-mail address: [email protected] (M. Eriksson).

mailto:[email protected]

Nomenclature

CO2 carbon dioxideTEP tradable emission permit/allowanceM carbon dioxide emissions in marginal power plant (kg/year)H carbon dioxide emissions in household before changing residential energy system (kg/year)I carbon dioxide emissions in industry before introduction of tradable emission permits (kg/year)DH change in carbon dioxide emissions at household due to change in residential energy system

(kg/year)DM change in carbon dioxide emissions in marginal power plant (kg/year)DI change in carbon dioxide emissions in industry (kg/year)�D emission reduction to be achieved in the tradable emission permit system (kg/year)P* market equilibrium price of tradable emission permits (€/kg)MC marginal CO2 emission abatement cost (€/kg)T total CO2 emissions within system (kg/year)

Subscripts

HP situation with a heat pump in householdGas situation with a gas heater in householdI industryM power plant

Superscript

Sys total system

M. Eriksson, L. Vamling / Energy Conversion and Management 47 (2006) 3510–3518 3511

and costs of measures in households [3]. In an empirical study in Nova Scotia, it was concluded that the con-nection between the residential sector and the power sector is of importance for residential sector emissionstatistics [4]. The emission consequences of a decrease in the use of fossil fuels and an increase in electricityuse in the residential sector were in statistics assigned to the residential sector and power sector respectivelyand a new method on how to calculate residential sector emissions is suggested. However, no study knownto the authors, discusses CO2 consequences of residential energy measures in a context where the powerand/or district heating plants are included in a tradable emission permit system.

Since January 2005, there is a system with tradable emission allowances1 (TEP) in the European Union(EU) [5]. One important feature of the TEP system implemented in the EU is that, currently, not all partsof the energy sector are obliged to have emission permits for their CO2 emissions. The TEP system includes2

approximately 12,000 plants within the EU. These plants are thermal combustion plants, process industrieslike pulp and paper, steel and glass industries, district heating systems etc. The residential sector is not par-ticipating3 in the EU emission trading system and other policy instruments, like taxes, are applied on fuelsused for residential heating. With a technology like heat pumps, the heat is used in the residential sector, whichis not part of the TEP system, and power to the heat pump is produced in the power sector, which is a part ofthe TEP system. Heat pumps in households will thus provide a link between the trading sector and the resi-dential sector as illustrated in Fig. 1, or in other words, the heat pump will cross the boundary between thetrading and the non-trading sector. The above reasoning has the implication that due to measures in residen-tial energy system, emissions can be transferred, not only between the residential and the power sector, butalso between the emission trading sector and the non-trading sector.

1 In the rest of the study, the term tradable emission permit (TEP) is used.2 With participating and included in the TEP system means that actors are obliged to present as many permits to the regulatory

authority as they have released emissions during a specified time period.3 It is possible to participate on a voluntary basis.

Fig. 1. Heat pump linking the trading and the non-trading sector.

3512 M. Eriksson, L. Vamling / Energy Conversion and Management 47 (2006) 3510–3518

2. Objective

The objective of this study is to analyse the affect of an emissions trading system, in which the power sectoris included, but where the residential sector is not obliged to have emission permits, on carbon dioxide emis-sion consequences of heat pumps in households. The study starts with a discussion of emission permits. Then,it continues to discuss the carbon dioxide emissions consequences of technology choices in the household sec-tor in two cases; one when there is a tradable emission permit system and one reference case where there is noemission permit system. The technology choices considered are a gas heater and a heat pump. Finally, thestudy contains a discussion and conclusions about the findings of the previous parts.

3. Emission permits and electric power markets

The main idea of tradable emission permits is to achieve an emission target (in this case CO2 emissions) in acost efficient way. A regulatory authority issues emission permits to the actors that are participating in theTEP system. The sum of emission permits issued within the TEP system equals the desired emission targetin the TEP system. Cost efficiency is obtained by allowing trade with emission permits between the actorsin the system on a TEP market. Thus, the actors that are participating in the TEP system are not obligedto make a certain reduction but can choose whether to make emission abatements or buy emission permitson the TEP market. The choice for an actor between making emission abatements or buy permits dependson the marginal cost for emission abatements and the cost for buying permits. An actor in the TEP system,which can reduce emissions at a low cost, can thus sell excess permits to another actor for whom it is better, atthe market price of emission permits, to buy permits instead of making emission abatements.

The electric power and the emission permit markets are linked to each other since if a power plant increasesits production of power as a response to an increasing demand on the power market, either it has to makeemission abatements or demand more emission permits.

4. Methodology

The analysis is based on a cap- and trade emissions trading system in which the power sector, but not theresidential sector, is included.

In order to illustrate the affect of an emissions trading system when considering CO2 emission consequencesof heat pumps, a house where we can choose between a natural gas heater and a heat pump is analysed. Thestudy compares two different cases, a reference case when there is no TEP system and one case where a TEPsystem is introduced in which the power sector is included. For each of the two cases, the total4 CO2 emissions

4 Total equals CO2 emissions from the house, industry and power plant.


are calculated when a heat pump and a gas heater is used for heating the house. The society, in which thehouse is situated, consists for simplicity of only three actors, the house, a power plant and an industry.The house exchanges power with the power plant and, in the case with a TEP system, the power plantexchanges emission permits with the industry.

The aim here is not to compare a heat pump with a gas heater and no considerations are made on costs, thetechnologies are used to illustrate the CO2 consequences when the heat pump is crossing the CO2 emissioncap- and trade boundary. For an analysis of cost and environmental impact of different heating options,including both natural gas and heat pumps, see for example [6].

Since the link between residential energy measures and the TEP system, in the studied case, is the powergrid it is important to have an appropriate baseline of what happens on the power market due to measuresin the household. Several different methods dealing with how to account for changes on the power marketare discussed in [7,8]. This study only considers direct on-site, i.e., at the household, and direct off-siteemissions, i.e., in the power system. When considering direct off-site emissions, it is assumed that changeson the power market can be represented as changes in the operating margin power plant, i.e., the power plantthat is changing its production in response to an increased or decreased power demand. Indirect effects exceptthose on the TEP market, like economy–wide responses to changes in market prices of for example electricityand gas, are not considered.

Accordingly, the power plant can be seen as the marginal power plant on the power market and the indus-try can be seen as the marginal demander on the TEP market, that is the actor that will buy emission permitsinstead of making emission abatements for any decrease in TEP price.

5. CO2 consequences of technology choices in residential energy system

5.1. Reference case: without tradable emission permits

The reference case is used to represent a situation where there is no TEP system and is illustrated in Fig. 2.In the reference case illustrated in Fig. 2, CO2 emission abatement measures in the industry are independent

of emissions in the marginal power plant and emissions in the industry remain unchanged independently ofmeasures in the residential energy system.

Call emissions from the marginal power plant on the power market for M, emissions from the house for H

and emissions from the industry for I. When a heat pump is used in the house, emissions from the house arezero (eventual greenhouse gas emissions due to working fluid leakage are neglected), emissions from the powerplant MHP and emissions from the industry IHP. If a gas heater replaces the heat pump, there is an increase inemissions from the house with DHGas but a decrease of emissions from the power plant with DMGas since lesspower is produced. The increase DHGas depends on the efficiency of the gas heater and the specific emissions of

Fig. 2. Reference case without TEP system.

Fig. 3. Case where the power plant and the industry are included in a TEP system.


gas while the decrease DMGas depends on the efficiency of the heat pump, the fuel and the conversion efficiencyof the marginal power plant.

Total emissions within the system boundary when a heat pump is used are expressed in Eq. (1) and emis-sions when a gas heater is used are expressed in Eq. (2). Subscript HP denotes the situation with a heat pumpand subscript Gas the situation with a gas heater in the house.

T SysHP ¼ MHP þ IHP ð1Þ

T SysGas ¼ DHGas þMHP � DMGas þ IHP ð2Þ

Obviously, if DHGas > DMGas, a heat pump results in lower CO2 emissions than a gas heater.

5.2. Case with tradable emission permit system

The case where the industry and the power plant are included in a TEP system is illustrated in Fig. 3. Thehouse and the power plant act on the power market and the power plant and the industry act on the TEPmarket. Call the emissions from the marginal power plant, prior to the introduction of a TEP system, M

and emissions from the industry I. Assume that a regulatory authority has decided that emissions withinthe TEP system should decrease with D so that the emission target within the TEP market is I + M � D.The regulatory authority issues emission permits equal to the sum I + M � D within the TEP system so thatthe total supply of emission permits equals the emission target in the TEP system. The demand of emissionpermit is determined by the marginal abatement cost for actors in the TEP system. The market equilibriumprice P* of emission permits is assumed to be equal to the marginal abatement cost of the actor that is themarginal demander.

In order to achieve the CO2 emission target in the TEP system, the marginal power plant and the industryhas to reduce its emission with a total amount of D. The reduction in the TEP system can be written asDM + DI = �D where DM is emission reductions in the marginal power plant and DI is emission reductionsin the industry. However, nothing is said about how they distribute this reduction between each other, thatdistribution is determined by their marginal abatement costs. There are of course various ways to achievethe emission target for the actors within the TEP system, for example a fuel shift from coal to gas in the powerplant, efficiency measures like process integration in the industry or replacement of fossil fuels with bio fuels.Emissions from the marginal power plant can be reduced in two different ways:

• Emission abatement measures in the power plant, for example a fuel switch to a fuel with less CO2 emis-sions or improving the conversion efficiency.

• Production is reduced, for example because households demand less electric power.

Fig. 4. Schematic illustration of emission abatements and price in the TEP system with heat pump.


In order to separate effects from residential sector technology choices and CO2 abatement measures beingmade by actors within the TEP system, it is for clarity assumed that emission abatements measures in thepower plant are expensive. Consequently, it is assumed that emission reductions in the marginal power plantcan only be achieved through a decreased electric power production.

The emission target in the TEP system should be achieved independently of measures that is made in theresidential sector since the supply of emission permits is M + I � D. This condition is expressed in Eq. (3).

DMHP þ DIHP ¼ DMGas þ DIGas ¼ �D ð3Þ
Now, assume that there is a heat pump in the house and a TEP system is introduced. Prior to the introduction
of TEP, total emissions within the TEP system are I + M and the emissions within the total system are also I + M

since H = 0. Emission abatements that have to be achieved within the TEP market to reduce emission with�D,are made in the industry since it has the lowest emission abatement cost. Thus, DIHP = �D and DMHP = 0. Aschematic illustration of the situation on the TEP market when a heat pump is used in the house is shown ina supply-demand diagram in Fig. 4. The supply of emission permits within the TEP system is assumed that tobe inelastic to the TEP market price in a short-term perspective and is represented with a straight vertical line.

Emissions within the total system, after the introduction of TEP, when a heat pump is used in the house areexpressed in Eq. (4).

T SysHP ¼ 0þM þ I þ DMHP þ DIHP ¼ M þ I � D ð4Þ

Consider now a situation where the heat pump is replaced with a gas heater. Then, emissions from thehouse will increase with DHGas, but less electricity has to be produced in the marginal power plant and con-sequently emissions from the marginal power plant decrease with DMGas. Since the power plant now producesless power, the marginal power plant will also demand less emission permits and the replacement of the heatpump results in a shift in demand for emission permits by the marginal power plant. Assume thatjDMGasj < jDj, since in reality there are many plants within the TEP system. If jDMGasj < jDj, there is adecrease in emissions within the TEP system due to the replacement of the heat pump with the gas heaterbut this reduction is not enough to achieve the emission target in the TEP system. The condition apply thatDMGas + DIGas = �D and consequently there is also a reduction in the industry. The situation on the TEPmarket when a gas heater replaces the heat pump is illustrated in Fig. 5.

Fig. 5. Schematic illustration of emission abatements and price in the TEP system with gas.


Notice that in Fig. 5., the shift in demand of emission permits is due to a replacement of the heat pump witha gas heater and thus at the expense of increased emissions from the house, outside the TEP system. Totalemissions if the heat pump is replaced with the gas heater are expressed in Eq. (5). By comparing Eqs. (4)and (5) it is seen that total emissions are higher when the heat pump is replaced with a gas heater sinceDHGas > 0.

T SysGas ¼ DHGas þM þ I þ DMGas þ DIGas ¼ DHGas þM þ I � D ð5Þ

The result obtained in Eq. (5) concerning CO2 emissions when a gas heater replaces the heat pump can beseen as a leakage effect. A part of the emission reductions within the TEP system (through less power produc-tion) is achieved through increasing emission outside the TEP market (the gas heater replacing the heat pump).Notice that this result is independent of the magnitude of DMGas, which means that it is independent of thetype of power plant that constitutes the marginal power plant. The result is also independent of the efficiencyof the heat pump.

It should also be observed that the TEP price P* on the TEP market is different with the two technologies,as illustrated in Figs. 4 and 5. A gas heater in the house results in a lower TEP price compared to the case witha heat pump. This is not surprising since the TEP price is what adjusts to keep the relation DM + DI = �D onthe TEP market. The TEP price decrease is dependent on the marginal abatement cost in the industry and alsodependent on marginal power plant fuel and conversion efficiency as well as heat pump efficiency since it deter-mines the magnitude of DMGas and thus also DIGas.

6. Discussion

This study illustrates with an analytical example that exchange between actors within a TEP system andoutside the TEP system is important when considering CO2 consequences of technologies that cross an emis-sions trading system boundary. This situation occurs for example if power producers and district heating pro-ducers are obliged to have emission permits but not households for alternative heating technologies (nor anyactor upstream the household in the alternative heating fuel distribution chain).

Fig. 6. Illustration of policy instruments applied on technologies for residential heating.


The study shows that, in the case with a TEP system, replacing a heat pump with gas in the residential sec-tor results in a leakage. The emission target within the TEP system is partly achieved through increased emis-sions in the residential sector.

In the study, it is assumed that the emission target is fixed and independent of the TEP price. The emissiontarget within the TEP system is only fixed in a short-term perspective. If a heat pump is replacing a gas heater,emissions are transferred inside the TEP system but this is done at the expense of a higher TEP price. An inter-esting question is of course what the affect of an increased TEP price on the TEP market due to measures inthe residential sector means in a long-term perspective. It is possible that the political possibilities to have amore ambitious emission target within the TEP system in the future is affected by the cost to achieve a lessambitious target.

This study does not analyse which technologies that are preferred in the residential sector from an environ-mental and/or cost perspective. It is of course possible that a replacement of heat pumps with other technol-ogies in households is a cost efficient way of reducing emissions. However, considering the results of the study,it is important from a policy perspective to be aware of the interconnection between measures outside the TEPsystem and the TEP system. If a TEP system is introduced on the emissions caused by heat pumps (i.e., powerplants) and/or district heating plants and the residential sector is subject to other policy instruments, it resultsin a situation as illustrated in Fig. 6.

It is important that the policy instruments used in the residential sector and the TEP system interact in away so that cost efficiency is maintained in the TEP system while not result in unintended subsidies for certainresidential heating technologies.

7. Conclusions

The aim of this study was to discuss the affect of an emission trading system, in which power producers butnot households are participating, on the carbon dioxide emission consequences of heat pumps in households.In that context, the heat pump will link the emission trading sector with the non-trading residential sector. It isargued that in such situations, it is important to consider the interaction between the emission permit marketand measures in the residential sector. The interaction with the TEP market means that replacing a heat pumpwith a fossil fuel in the residential sector, results in a leakage of emissions since the emission target in the TEPsystem is partly achieved through increased emissions outside the TEP system.

Acknowledgement

This study has been carried out under the auspices of the Energy Systems Programme, financed by theSwedish Energy Agency.


References

[1] Balaras CA, Droutsa K, Dascalaki E, Kontoyiannidis S. Heating energy consumption and resulting environmental impact ofEuropean apartment buildings. Energ Build 2005;37:429–42.

[2] Rolfsman B. CO2 emission consequences of energy measures in buildings. Build Environ 2002;37(12):1421–30.[3] Joelsson A, Gustavsson L. A systems perspective on heat pumps and district heating in the residential sector. In: Thonon B, editor.

Heat transfer in components and systems for sustainable energy technologies. proceedings of the heat–SET 2005 conference, Grenoble,France: Edition-GRETh; 2005.

[4] Hughes L, Bohan K, Good J, Jafapur K. Calculating residential carbon dioxide emissions–a new approach. Energy Policy2005;33(14):1865–71.

[5] European parliament and the council. Directive 2003/87/EC of the European parliament and of the council-establishing a scheme forgreenhouse gas emission allowance trading within the community and amending council directive 96/61/EC. Official Journal of theEuropean Union 2003; L275/32.

[6] Kartha S, Lazarus M, Bosi M. Baseline recommendations for greenhouse gas mitigation projects in the electric power sector. EnergyPolicy 2004;32(4):545–66.

[7] Sjodin J, Gronkvist S. Emissions accounting for use and supply of electricity in the Nordic market. Energy Policy 2004;32(13):1555–64.[8] De Almeida AT, Lopes A, Carvalho A, Mariano J, Nunes C. Evaluation of fuel-switching opportunities in the residential sector. Energ

Build 2004;36(2):195–203.

Documents

Heat pumps and tradable emission permits: On the carbon dioxide emissions of technologies that cross a tradable emission market boundary