Households’ impact onGHG and SO2 emissions in Aragon
Duarte, R.; Rebahi, S.; Sánchez Chóliz, J.
Abstract
Households, through its demand in the economy, are potentially a major actor to reduce atmospheric emissions. The environmental impact of private demand includes direct and indirect emissions. This analysis focuses on atmospheric pollution caused by emissions of greenhouse gases and sulphur dioxide. Using an Applied General Equilibrium Model we analyze, through simulations, the environmental impact of some measures of the Action Plan of the Spanish Strategy on Energy Saving and Efficiency (PAE4 2005-2012) related to changes in consumer behaviour. Specifically, we analyze the impact of electricity savings and the promotion of public transport versus private car use under different assumptions on savings and public transport efficiency in Aragón, a region situated in the northeast of Spain.
- Introduction
Actually, the result of development and technological progress has led to various forms of pollution. Atmospheric pollution is one of the most important environmental problems thatgrowing fast. This analysis focuses on atmospheric pollution caused by GHG (greenhouse gases) and sulphur dioxide (SO2) emissions. GHGs are the main causes of climate change. By adapting the Kyoto Protocol, countries agreed to reduce global emissions of GHG by 8% between 2008 and 2012, in reference to 1990 levels. In Spain, CO2emissions have increased at around 50% in 2004 and 2005 compared to 1990 levels. On the other hand, SO2emissions have a local impact and come mainly from burning fossil fuels. This type of emision is a good indicator of local pollution. These emissions are partially responsible for acid rain and pollution in urban centres. Reducing these emissions is one of the main objectives of European Community Environmental Policy, having established emission ceilings of this substance for each member state (Directive 2001/81/EC).
Different states organize their contribution to environmental improvement through national strategies, which are reflected at regional level. In Spain, the Autonomous Communities develop their strategies to combat climate change, including technological proposals, production and changes in lifestyle habits. In this paper we focus on Spain and specifically in the Community of Aragon[1].In this geographical context, the Aragones Climate Change Strategy and Clean Energy (ACCSCE) is the contribution of this region to objectives established in the Spanish Strategy of Energy Saving and Efficiency (PAE4 2005-2012) in order to achieve GHG emissions reductions in sectors not affected by the plan of trade emission rights. The aim of the ACCSCE is the reduction of GHG emissions in Aragon by 1.3 Mt of CO2eq (0256Mt of CO2eq/year) during the period 2008-2012. Moreover, the Spanish Plan for Reducing Emissions from Large Combustion Plants, whose time horizon covers the period 2008-2015, proposes to reduce SO2 emissions by 80% in respect to 2001 levels.
Concretely, the objective of this study is the evaluation of the impact of household consumption on GHG and SO2emissionsin Aragon using an Applied GeneralEquilibrium Model (AGEM). The inclusion of SO2 emissions in the analysis helps us to approach the regional impact of Aragones households because, unlike GHG emissions, this gas is characterized by its local effect. According to this objective, we generate representative scenarios of changes in consumption patterns, consistent with measures proposed by ACCSCE, by analyzing the impact of electric saving in the domestic sector and the partial substitution of private car use by public transports.
Our first hypothesis consists in consider that the responsibility of atmospheric emissions are not associated only with who produces them, but also to end users of products. Munksgaard etal., (2001). Under this criterion, we consideras households indirect emissions, emissions generated in the production processes to meet theprivate demand. Following this finalist assumption, we can also associateemissions of economic activities to different components of final demand, ie, exports, public expenditures and investment. Moreover, emissions generated from heating, cooking, use of car ... etc. are considered as householddirect emissionsand depend on the amount of energy used (fuel, gas, coal ... etc.). Using the input-output model, this approach was addressed in several studies to calculate the environmental impacts, Ferng (2002), and Giljum Hubacek (2003), Resosudarmo (2003), McDonald and Patterson (2004), among others. In the case of Spain, these models have also been widely used to study the environmental impact. A review of the literature can be seen in Sánchez-Chóliz et al. (2007).
The second hypothesis refers to “Rebound effect” or “Backfire effects”' (Brookes 1990; Herring, 1999; Birol andKeppler, 2000, Saunders, 2000; Schipper, 2000). The Rebound effect is usually discussed in connection with “energy-saving”. The energy-saving essentially implies a lower energy bill, which can be viewed as a reduction of the real price of energy services. For example, if petrol costs less pertransport unit, private car use may increase, this can affect the initial energy-saving potential.Furthermore, lower energy costs increase real income, which leads to an increase in consumption of other goods. This in turn offsets the emission reductions from the initial energy saving. A third effect may be denoted general equilibrium effects, since changes in aggregate consumption patterns may lead to structural change and changes in relative prices. Taken together, these effects can be denoted the rebound effect; see Brannlundet al (2007). In this context, the Applied General Equilibrium Model (AGEM) has been increasingly used in empirical analysis of changes in demand structure and energy efficiency improvements, Anson and Turner (2009). The study of environmental rebound effects is a relatively new area; Sorrell (2007) has identified only eight studies that have used AGEM to evaluate economic and environmental impacts of rebound effects resulting from energy efficiency improvements.
The rest of the paper is organized as follows. Section 2 presents the sources of information used to obtain a Social Accounting Matrix (SAM) for the Aragonese economy and emissions accounts, and the main assumptions used to the specification of the AGEM, calibrated to the previous SAM. In the third section, we describe the baseline scenario including the pollution structure of Aragonese economy. Section 4; resume the scenarios performance and the discussion of results. Section 5 closes the paper with a summary of the main conclusions and possible extensions of the study.
- Data and methodology
Broadly, the study of the impact of electric saving and promotion of public transport is done by generating and evaluating different scenarios using an AGEM, built and calibrated to the Aragonese economy of 2005. The AGEM used is based on the standard model developed by “International Food Policy Research Institute” (IFPRI). For its calibration we used a Social Accounting Matrix for Aragon with reference to 2005 (SAMA-05), obtained by the authors from the most recent Input-Output Framework (IOFA-05) available, Perez and Parra (2009).
2.1.The Social Accounting Matrix for Aragon (SAMA-05)
The structure used to register the different flows of SAMA-05 is a square matrix that covers 30 economic activities, two factors of production (labour and capital), Households, No-profit institutions serving households (NPISH), Societies, Government and Foreign Sector. The government account includes public administrations (AAPP), value added tax (VAT) and taxes net of subsidies product. Since this SAM represents a regional economy, the foreign account includespurchases and sales realized with the Rest of the World (ROW), European Union (EU) and other commercial exchanges with the Rest of Spain (ROS).
Basically, SAMA-05 is an extension of IOFA-05 that incorporates income flows produced between institutions. This sub-matrix of income among institutions is a major contribution of all SAM, because it provides a disaggregated view of the circular flow of income and describes how production process influences income generation and determines consumption, savings, investment and financing requirements of institutional sectors. The structure of this sub-matrix, followed to obtain SAMA-05, can be seen in Mainar (2010).
Faced with a lack of information at the regional level, the estimation of this sub-matrix started with a first estimation al the national level using data of National Accounts of Spain, INE (2005a), the Financial Accounts of the Spanish Economy 2005, Bank of Spain (2005), and government accounts 2005, IGAE (2005). Therefore, GRAS method is applied using data of Aragon Regional Accounts, INE (2005b). Once the sub-matrix of income is estimated, we proceed to include tax paid by institutions registered in the IOFA-05 (VAT and taxes net of subsidies product). Savings[2] account and Capacity/ Requirement of financing of the Aragonese economy are determined by matching resources and uses of the several accounts.
2.2.Emissions account
Emissions analyzed in the present paper refer to greenhouse gas (GHG) and sulphur dioxide (SO2). According to the guidelines of the Kyoto Protocol, greenhouse gases are: carbon dioxide (CO2), methane (CH4), nitrogen monoxide (N2O), sulphur hexafluoride (SF6) hydrofluorocarbons (HFCs) and perfluorocarbons (PFCs). The construction of emissions account follows the NAMEA system, De Haan et al (1994), where definitions and accounting rules are compatibles with concepts of the System of National Accounts (ESA-95). To construct the GHG emissions account, we started from CORINAIR- Inventory of Aragón 2005(DGA, 2006a), witch has a different system than the NAMEA one, and we apply some adjustments following the methodology developed in Vetrella Tudini (2004). Since the six greenhouses have a different Global Warming Potential (GWP), we consider total GHG emissions in Kilotons of equivalentcarbon dioxide (Ktn of CO2eq), using factors published in IPCC (2007). The account of SO2 emissions is obtained from the Inventory of emissions to the atmosphere in the region of Aragón (DGA, 2006b).
2.3.The Applied General Equilibrium Model
In our analysis, we construct an IFPRI-AGEM for the Aragones economy. Several empirical studies have based on this type of model; see eg Banerjee et al. (2006). Additionally to the database (SAMA-05), the calibration procedure requires the specification of certain elasticity. A detailed description of IFPRI model can be found in Löfgren et al. (2002). In what follow, we describe briefly the main assumptions made for the construction of the AGEM.
Productive activities maximize benefits by minimizing costs subject to a given level of production. The production technology is modelled as a combination of intermediate consumption (inputs) and value add (Labour and Capital). We assume a productionfunction of fixed coefficients without possible technological changes.
The import is regulated by the Armington function, wherethe aggregate consumption demand (Armington aggregate) is decomposed into a domestic component and import Armington (1969). The flexibility of substitution between both components is expressed by Armington elasticity, which takes the value frequently used σq = 0.8 for all economic activities. Total production is distributed between exports and domestic sales. The transformation between the two sales is expressed by an elasticity of transformation function (CET), assumed equal to σt = 1.6 as in other models IFPRI.
Households receive income from production factors and transfers from government and other institutions. Household expenditure equals to disposable income after tax cuts, savings and transfers to other institutions. All households pay taxes, realize transfers and save according to fixed percentages of their income. Private consumption per product results from utility maximization under a linear expenditure system (LES). In this function we have used expenditure elasticities estimated in Mainar (2010) for Spain.
Government revenues come from taxes paid by the productive activities, capital, households and transfers realized by institutions. These revenues are allocated entirely too public expenditure, public transfers and savings in proportion to the data of the year base. Similarly, investment is determined from the quantity of savings and it is allocated to Gross Capital, tax on capital and Capacity/Requirement of financingin proportion to the data of the year base.
The IFPRI model includes three macroeconomic balances: External balance, Government and Savings-Investment balances. These statements allow us to choose between several closures rules that can be consulted in detail in (Löfgren et al., 2002). Because the AGEM used here is for one period, it is assumed that government savings is flexible, which means that size of public sector (measured by public expenditure) is fixed in the short term, the exchange rate is endogenous and foreign savings is constant. In the savings-investment balance, gross capital is flexible while taxes on capital and Capacity/Requirement of financingare constant.
2.4.The emission model
The emission estimated in the AGEM takes into account households direct emissions and direct and indirect emissions of production activities.
ETOT= EDH + EDIA
WhereETOT are the total emissions, EDH household’s direct emissions and EDIA are direct and indirect emissions of all economic activities.
EDH emissions results from household energy consumption. These emissions are obtained as the product of the vector i = (ie) of emissions per unit of each type of energy by energy vector c = (ce), the index e refers to the type of energy product (“Coal”, “Refined oil” and “Gas”). Note that electric consumption is not included in these emissions because their emissions are attributed entirely to the electricity sector.
EDIA emissions are those that occur during the production process, including those relating to electric consumption. Through the input-output model, we can relate emissions of productive activities with monetary flows expressed in SAMA-05 according to the following relation, see Sanchez-Chóliz et al. (2007).
EA = d (I-A)-1s
Where d is a unit vector in emission (Ktn of CO2eqper monetary unit of output);(I-A) - 1, is the Leontief inverse matrix, s is the vector of final demand or a component of this demand, and finally, EA will be emissions of economic activities associated to s.
After the specification of the AGEM used and it calibration to the SAMA-05, we proceed to make changes in the structure of private consumption. In the AGEM, these changes will affect the actual prices of goods and services and determine the behaviour of economic agents, defining a new equilibrium. The comparison of the magnitudes of the new equilibrium with the baseline scenario allows us to quantify the impact of changes introduced.
- Description of the baseline scenario
The information contained in the accounts of emissions allows us to know the structure of atmospheric emissions in Aragon, which can be seen in Table A1 of the Annex. We observe that foureconomic activities cover more than 75% of GHG emissions. Emissions of “Electricity”, “Mineral and metal products” and “Paperand printing” are mainly CO2 emissions, while emissions of “Agriculture, forestry and aquaculture” are mostly N2O emissions and CH4. GHG emissions generated in the domestic sector totalled 1.581 Ktn of CO2eq, which represent 6.92% of emissions in Aragon. In the account of SO2emissions, it appears that the electricity sector represents over 82% and the direct responsibility of households is almost zero (587 Tons, 0.30%).
Using the input-output model, direct emissions of economic activities have been distributed between components of final demand. As shown in Table 1, emissions associated to export (embodied emission) are the most important, accounting 43.97% of GHG emissions and 43.87% of SO2 emissions generated in production processes. Emissions of this gas associated to private consumption represent 42.63%, while the weight of GHG emissions associated to this account stood at 33.34%. For the rest of final demand, emissions associated represent less than 16%, where investment accounts about 2/3 of this rest. This structure reveals the brunt of household’s indirect emissions compared to direct one.
Table 1. GHG emissions of SO2 associated to final demand
The pollutant structure (only economic activities) of each component of final demand by product group is presented in figures 1 and 2. Figure 1 shows that GHG emissions associated to household consumption are mainly due to “Energy products”, “Agriculture and Food” and “Services” (especially commercial and real estate services). The importance of these services is due to its volume in the consumption and not to its emission potential. In emissions attributable to exports, products of “Agriculture and Food” and “Industrial products” have more contribution, while “Energy products” and “Services” lose it. Finally, we see that emissions associated to government and NPISHs are mainly explained by the consumption of services (education and health) and those associated to investment are due mostly to “Construction”.The analysis of SO2emissions in the Figure 2 shows a structure quite different from the previous one. In fact, we see that “Energy products” have a greater contribution in emissions associated to private consumption and exports, while the weight of “Agriculture and Food” is greatly reduced. For the rest of the final demand, the structure of SO2emission is very similar to that of greenhouse gases. These differences are explained by emission intensity of each economic activity (emission per unit of output), since the composition of demand is identical in both graphs.
Figure 1. Structure of GHG emissions associated with final demand
Figure 2. Structure of SO2 emissions associated with final demand
- Performance of simulations
According to objectives stated above, we simulate two scenarioswhich research an approximation to the environmental impact of two specific measures of the ACCSCE.
4.1.Description of scenarios
Scenario 1: Electric saving in the domestic sector
This measure focuses on reducing electricity consumption of Aragones households by 10%. According to our estimation, this consumption represents 41.34% of household energy consumption and accounts at around 4901 Gwh (DGA, 2005). This saving can be achieved with a more responsible use of electricity by the replacement of appliances with low energy labels by Class A or higher as outlined in the Plan RENOVE of appliances of (ACCSCE). The simulation is performed under two assumptions: a). the savings obtained is allocated to the consumption of other products and services (rebound effect), this will generate an increase in consumption of other products of 0.2%. b). the savings obtained is allocated to increase private savings account, and the consumption of other products is constant. This implies a reduction in private consumption expenditure of 0.2%.
Scenario 2. Modality transport substitution
In the second scenario, we simulate a change in the transport modality of Aragones households, as proposed in the (ACCSCE). The ultimate objective of this measure is to improve urban mobility with greater involvement of most efficient modes of transport (especially public transports) and reducing the use of private car with low occupancy. This measure is introduced in the AGEM as a reduction in the cost of private car use and increasing spending on public transport by a percentage of the reduction applied. When lower is the percentage designated to “Transports and Communications”, more efficient is the alternative mode of public transport. As it known, costs that may be reduced when persons decide to reduce their car use, are associates with car maintenance, repairs, tires, loss of value of the vehicle (depreciation), fuel, insurance...etc. However, for simplicity, only costs of fuel are considered in this paper. Concretely, we assume a reduction of 10% in private consumption of “Refined petroleum products”,this account includes diesel and gasoline consumption.