Reuniones de Estudios Regionales



Energy efficiency impacts:

a multiregional application to Portugal

Luís Cruza (lmgcruz@fe.uc.pt), João-Pedro Ferreirab (joao.ferreira@fe.uc.pt),

Pedro Ramosa (pnramos@fe.uc.pt), Eduardo Barataa (ebarata@fe.uc.pt)

a Faculty of Economics / CeBER (Centre for Business and Economics Research), University of Coimbra.

b Faculty of Economics, University of Coimbra; GOVCOPP (Research Unit on Governance, Competitiveness and Public Policies).

Subject area: 08. Sustainability, natural resources and environment

Abstract:

This work aims to contribute for the development of a modelling framework to account for the (inter)regional energy-economy-environment interactions resulting from energy efficiency gains. Accordingly, we propose an empirical application using an environmentally extended multi-regional Input-Output (IO) model for Portugal. This is a ‘closed’ rectangular tri-regional IO model, dividing the NUTS II Centro Region into Coast Centro Region and Interior Centro Region and adding, as a third region, the Rest of Portugal, derived from the MULTI2C (multi-sectoral multi-regional Coimbra) model. The MULTI2C framework has a high level of detail concerning both products and the industries that produce them. Further, it is recognized that electricity production and distribution encompasses a set of activities with different technologies, either by main energy source or by region. To account for this, we consider the division of the original electricity industry into one industry of “electricity distribution” and 9 industries of “electricity production” (differentiating several renewable and non-renewable sources). Next, we extend this modelling approach to assess the energy, socio-economic and environmental impacts resulting from energy efficiency scenarios.

Preliminary results reveal that the energy efficiency gains, in the majority of the scenarios, would generate a (small) net contractionary effect on the Portuguese economy, following the electricity production reduction, which is not fully compensated by the second-order expansionary effects. However, effects are being counterweighed by social and environmental improvements.

Keywords: Electric power industry; Energy efficiency; Energy use; Energy-related CO2 emissions; Input-Output analysis

JEL codes: C67; Q56; R11

1. Introduction

Energy efficiency is at the heart of the European Union (EU) 2020 Strategy for smart, sustainable and inclusive growth (COM/2010/2020). The initial purpose of this work is to build an Input-Output (IO) model to analyse the interactions between the Centro Region (a NUT II located in mainland Portugal, occupying the central part of its territory) and the remaining areas of the country. This NUTS II region is characterised by a strong asymmetry, where a less developed region (the interior part), still relying on exports of primary products, either abroad or to the rest of the country, coexists with a more industrialised, populated and developed region (the coastline). Hence, as shown in Figure 1, the Centro Region is split into the interior part of the territory and the coastal region, according to the aggregation of NUTS III (2002 classification) that follows: the Coast Centro (C) region includes Baixo Vouga, Baixo Mondego, Pinhal Litoral, Médio Tejo and Oeste; while the Interior Centro (I) region embraces Pinhal Interior Norte, Dão-Lafões, Beira Interior Norte, Serra da Estrela, Cova da Beira, Pinhal Interior Sul and Beira Interior Sul. The model also considers the remaining Portuguese NUTS III regions all merged into one additional region, termed Rest of Portugal (RP) (thereby exhausting the Portuguese economy).

[pic]

Figure 1. Map of Coast Centro, Interior Centro and Rest of Portugal regions

Next, the tri-regional Coast Centro – Interior Centro – Rest of Portugal (C-I-RP) IO model is explored to assess the impacts, in the three regions, resulting from energy efficiency scenarios. For this it is critical to extend the modelling framework, in order to account for energy-economy-environment interactions. Indeed, the ultimate aim of this research is to extend the assessment of the impacts resulting from energy efficiency gains to the environmental dimension.

The rest of this paper is structured as follows. Section 2 starts by presenting the structure of the multi-sector tri-regional C-I-RP IO model, followed by a discussion on the core aspects of its extension for the consideration of energy and environmental issues. This section presents both the methodological choices and the data sets used for the tri-regional Portuguese empirical model. The main results of the scenarios considered are to be presented and discussed in Section 3. Section 4 concludes.

2. Methodology and data

The tri-regional C-I-RP IO model proposed in this work is an application of the MULTI2C (multi-sectoral multi-regional Coimbra model) framework, using 2010 data. MULTI2C is a general flexible approach, developed by a group of researchers from the University of Coimbra (Portugal) that allows for the construction of multi-regional IO models for different geographic configurations and empirical applications. The MULTI2C framework uses top-down non-survey methods to regionalize I/O tables (for the 30 Portuguese NUTS III regions), using detailed information provided by the Portuguese National and Regional Accounts, together with other detailed statistical information at the regional level from Statistics Portugal (INE) (population census, households expenditure survey, agricultural census and national forestry survey) (further details on MULTI2C can be seen in Ramos et al., 2015).

2.1. The structure of the tri-regional C-I-RP IO model

The proposed multi-sector tri-regional C-I-RP IO model has a configuration as outlined in Figure 2. This structure is based on a set of characteristics and hypotheses, which leads to the classification of this modelling approach as a “closed rectangular tri-regional IO model” (Miller and Blair, 2009: Chapter 5), using domestic flows at basic prices. The approach has a great level of detail concerning both the products (or groups of products) included and the industries that produce them.

[pic]

Figure 2 - Structure of the multi-sector tri-regional C-I-RP closed IO model

Rows corresponding to products (431 products × 3 regions) describe their different destinations, which include: the intermediate consumption (IC) in each region (naturally, e.g. a product produced in Coast can be inter-regionally exported and used as intermediate consumption in Interior or Rest of Portugal); the final consumption of the different types of households in the three regions; and other destinations in the “Other Final Demand”.

Columns corresponding to industries describe their technologies in absolute values, i.e., each product intermediate consumption in each industry, according to the origin’s region (C, I or RP); the intermediate inputs internationally imported; the (non-deductible) taxes less subsidies falling upon the purchased inputs (in order to assure that each industry IC is expressed at purchaser prices); the income generated in each industry and in each region, i.e., the gross value added (GVA), whether it is directly distributed to households living mainly from their labour income, or distributed to some other institutional sector.

i. Closed model

The model considers, in the three regions, different household’s types, according with their main source of income, namely: labour earnings, capital income, real estate income, pensions and other social transfers.

The model is “closed” regarding the consumption of households that live mainly from labour income (employees or self-employed workers). The income generated in each region contributes only for the consumption of households living in the same region; commuting and other periodical or seasonal migrations between C, I and RP (that are relatively small between these regions) were not considered. Consumption of other household’s types (the non-labour income dependent ones) is considered exogenous, and therefore considered as part of the Other Final Demand. Besides the consumption of the non-labour income dependent households, the Other Final Demand includes the consumption expenditures of general government and non-profit institutions, the investment, the consumption of non-residents in Portugal that visit the regions and, finally, other international exports of goods and services.

ii. Regional production technologies and the electric power industry specificities

This rectangular tri-regional IO model admits that each industry has its own technology, identically to the production of all its primary or secondary products (secondary products represent only a residual value in total industry production).

Moreover, in general, it is assumed that each industry has the same production technology in all the regions, i.e. each input has the same weight in the intermediate consumption regardless the production place.

At the high disaggregated level we conducted our work the equal production technology assumption in the three regions leads to negligible errors. But this is not the most accurate for all the cases, because it does not take into account the disparities of activities within some industries and/or regional differences. E.g., the original information provided by the Portuguese National Accounts Supply and Use Tables (INE, 2012a) considers one column vector of technologies for the electricity industry. However, the electricity power industry is composed by different activities such as production and distribution. Further, for the purposes of this research, it is particularly relevant to highlight that electricity can be generated by a wide variety of sources (renewable and non-renewable), encompassing a set of activities with different production technologies. Thus, as the regions possess different structures of electricity production and distribution, first and higher-order effects of a shock spread-out differently. In short, the disaggregation of the original electricity industry considers one industry of electricity distribution and 9 industries of electricity production according to the following sources: wind, geothermal, hydro, photovoltaic, coal, fuel oil, natural gas, diesel and cogeneration. Each of these sub-industries has its own production technology. The total electricity sector, that merges the 10 sub-industries, produces two distinct products (in each region): produced electricity and distributed electricity. Technical details on the procedures used to derive the electricity production structure for the different regions, in the Portuguese case, can be found in Ramos et al. (2015).

A second exception we looked at was the case of the refined petroleum industry, where a single column technical vector would hide the regional disparity of its input structure. Accordingly, we assume diverse production technologies for the different regions, based on Regional Accounts’ information (INE, 2012b), and taking into account our knowledge of the refineries’ actual location in Portugal.

Summing up, this model considers that the 431 products included in the MULTI2C approach are produced by the 134 industries in the three different regions or are being internationally imported, i.e., part of these products are produced outside the Portuguese territory. It is important to mention that the portion of Figure 2 inside the black bold border - a square matrix of dimension 1698 (431 products, 134 industries and 1 extra row relating to income of households that live mainly from their labour earnings, for each one of the three regions) - is the core of the IO framework implemented. Indeed, one departs from this core to compute the inverse matrix, which comprises a set of multipliers that measure impacts of exogenous final demand changes on products and industries production (Miller and Blair, 2009). Also, this inverse matrix includes the impacts on the income of the households that live mainly from their labour earnings, caused by those shocks.

2.2. The extension of the IO modelling framework to account for energy-economy-environment interactions

Then, through an environmental extension of our tri-regional IO model, we apply the IO technique to the structural analysis of energy requirements and CO2 emissions by economies, relating this pollution with the use of fossil fuels. For this, it is used a satellite account approach regarding the primary energy consumption by industry, from which the changes in oil and derivatives, natural gas and coal consumption are evaluated in terms of tons of oil equivalent (toe) and then estimated the corresponding CO2 emissions. The methodology used follows closely the approach proposed by Cruz and Barata (2012), adapting it to the multi-regional context.

Generally, the data required for the estimation of primary energy consumption is not directly available in the appropriate, or consistent, form. This subsection shortly presents the assumptions and estimations required in order to correlate the different data sources, with the final aim of obtaining suitable estimations of the physical quantities of the three primary fuels (natural gas, oil and coal) used by each sector in the Portuguese regions.

The estimation of primary fuels consumption (in physical terms) by each of the 134 industries considered in the IO tables made available in INE (2012a), takes advantage of the 2010 “Energy Balance” statistics (DGEG, 2013). The values for the total consumption of coal, oil and natural gas (expressed in tonnes of oil equivalent (toe)), from the 2010 “Energy Balance” (DGEG,2013), were considered as credible totals of Portuguese domestic energy use (by type of fuel) and it was from these that was derived the sectoral use of these three primary energy sources. Then, from these national figures, several procedures where applied to estimate the regional consumption of the primary fuels (in physical terms). As a rule, the regionalization process was developed considering the structure of intermediate consumption for each sector that consumes each of the primary fuels in each of the NUTS III regions (INE, 2012b). But some additional steps and further information was required to accomplish the task.

First, the estimation of coal consumption per region, for “351011 - electricity produced by coal” takes into account information on the quantities of coal consumption by the thermoelectric power plants located in the different regions. This information is available on the reports published by the companies that manage these power plants (EDP Produção, 2011; Tejo Energia/Pegop, 2011).

Second, it was considered more specific information regarding natural gas consumption by two particular sectors: “351017 - Electricity produced by natural gas” and “2303 – Manufacture of cement, lime and plaster”. The natural gas regional consumption by the electricity sector was estimated taking into account the information on the quantity of this fuel consumption by the thermoelectric power plants (available on the reports published by companies that manage these power plants: EDP Produção, 2013; Turbogás/Portugen, 2011; Tejo Energia/Pegop, 2011). On the other hand, the natural gas regional consumption by the cement sector was estimated taking into account the reports published by the companies that manage the cement plants (Secil, 2010a, 2010b, 2010c).

Figure 3 presents the decomposition of electricity production by source in each of the three regions.

|Coast Centro region (C) |Interior Centro region (I) |Rest of Portugal (RP) |

|[pic] |[pic] |[pic] |

Figure 3. Electricity production structure by source (% distribution)

Figure 3 shows that we could not assume equal production technology in the three regions, for the electricity industry as a whole, when in fact the production sources are very distinct. Indeed, while in the Coast Centro region tree-quarters (75.1%) of electricity generation is assured by non-renewables (50.1% for natural gas), in the Interior Centro region electricity generation is almost exclusively (96.5%) produced with the use of renewable sources (66.3% wind and 30.3% hydro); further, the Rest of Portugal presents an almost balanced share of renewables (51.8% in total, 12.5% from wind and 38.3% from hydro) and non-renewables (48.2%). Further, concerning renewables, it is noticeable that the Interior Centro region guarantees 37.4% and 9.5%, whereas the Coast Centro region produces 14.5% and 8.7%, of the national electricity generation from wind and hydro, respectively. On the other hand, regarding non-renewables, the Coast Centro region is responsible for 50.6%, 21.8%, 17.7% and 15.4% of the national electricity generation using natural gas, cogeneration, coal and fuel oil, respectively.

2.3. The energy efficiency scenarios

Next, we extend the tri-regional C-I-RP modelling approach to assess the energy, socio-economic and environmental impacts resulting from energy efficiency scenarios, departing from the assumption of a hypothetical 1% reduction in electricity consumption by all industries and household types.

3. Results and Discussion

Preliminary results reveal that the energy efficiency gains, in the majority of the scenarios, would generate a (small) net contractionary effect on the Portuguese economy, following the electricity production reduction, which is not fully compensated by the second-order expansionary effects. However, effects are being counterweighed by social and environmental improvements.

Work in progress.

4. Conclusions

This work proposes a closed rectangular tri-regional IO model, concerning the Coast Centro, the Interior Centro and the Rest of Portugal regions. Recognizing that electricity production and distribution encompasses a set of activities with different technologies, either by main energy source or by region, we consider the division of the original electricity industry into one industry of “electricity distribution” and 9 industries of “electricity production”.

Next, we use this modelling approach to assess the socio-economic and environmental impacts, in the Coast Centro, Interior Centro and nationwide, resulting from energy efficiency scenarios. Preliminary results reveal that the economic negative effects of the energy efficiency gains are counterweighed by social and environmental improvements.

References

Cruz, L.; Barata, E. (2012) “Hybrid IO Analysis of CO2 Emissions: An Application to the Portuguese Economy”. In Llop, M. (Ed.) Air Pollution: Economic Modelling and Control Policies, Ch. 5, p. 65-96, Environmental Sciences Series, Bentham EBooks.

DGEG (2013) Energy Balance 2010, Directorate General for Energy and Geology, Lisbon, Portugal.

EDP Produção (2011) Environmental Statement 2010 – Sines Thermoelectric Power Plant (Portuguese only). Directorate of Thermal Production - EDP Gestão da Produção de Energia, S. A., Sines, Portugal.

EDP Produção (2013) Environmental Statement 2012 – Lares Thermoelectric Power Plant (Portuguese only). Directorate of Thermal Production - EDP Gestão da Produção de Energia, S. A., Figueira da Foz, Portugal.

INE (2012a) Portuguese National Accounts 2010 (Base 2006), Statistics Portugal – National Accounts, Lisbon, Portugal.

INE (2012b) Portuguese Regional Accounts 2010 (Base 2006), Statistics Portugal – Regional Accounts, Lisbon, Portugal.

Miller, R.; Blair, P. (2009) Input-Output Analysis – Foundations and Extensions, 2nd Ed., Cambridge University Press, Cambridge, UK.

Ramos, P.; Cruz, L.; Barata, E.; Parreiral, A.; Ferreira, J-P. (2015) “A Bi-Regional (Rectangular) Input-Output Model for Portugal: Centro and Rest of the Country”. In Godinho, P.; Dias, J. (Eds.), Assessment Methodologies: energy, mobility and other real world applications, Ch. 12., p. 265-285. Coimbra: Imprensa da Universidade de Coimbra.

Secil (2010a) Environmental Statement 2010 – Secil-Outão Plant (Portuguese only), Secil – Companhia Geral de Cal e Cimento S.A., Setubal, Portugal.

Secil (2010b) Environmental Statement 2010 – Macieira-Liz Plant (Portuguese only), Secil – Companhia Geral de Cal e Cimento S.A., Maceira, Portugal.

Secil (2010c) Environmental Statement 2010 – Cibra-Pataias Plant (Portuguese only), Secil – Companhia Geral de Cal e Cimento S.A., Maceira, Portugal.

Tejo Energia/Pegop (2011) Environmental Statement 2011 – Pego Thermoelectric Power Plant (Portuguese only), Tejo Energia - Produção e Distribuição de Energia Elétrica, S.A., and PEGOP - Energia Elétrica, S.A., Abrantes, Portugal.

Turbogás/Portugen (2011) Environmental Summary 2010 - Tapada do Outeiro Combined Cycle Power Plant, Turbogás - Produtora Energética S.A. and Portugen - Energia, S.A., Gondomar, Portugal.

-----------------------

Coast

Centro

Interior

Centro

Rest of Portugal

Rest of Portugal

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download