A global, scope-based carbon footprint modeling for effective carbon reduction policies: Lessons from the Turkish manufacturing

https://doi.org/10.1016/j.spc.2015.05.005Get rights and content

Highlights

  • A global, scope-based carbon footprint of Turkish manufacturing sectors is analyzed.

  • A time-series multi-region input–output model is developed.

  • Only 5 out of 16 Turkish manufacturing sectors are dominated by scope 1 category.

  • Scope 3 category is responsible for nearly 56.5% of total carbon emissions of manufacturing sectors.

  • Electricity, gas and water supply sector has the largest contribution to supply chain emissions.

Abstract

The World Business Council for Sustainable Development (WBCSD) and the World Resource Institute (WRI) set the scope-based carbon footprint accounting standards in which all possible supply-chain related indirect greenhouse gas emissions are captured. Although this carbon footprint accounting standards are widely used in regional policy making, there is little effort in analyzing the scope-based carbon footprints of nations using a multi-region input–output (MRIO) analysis in order to consider the role of global trade. This research aims to advance the body of knowledge on carbon footprint analysis of the manufacturing sectors with a holistic approach combining the WBCSD & WRI’s scope-based carbon footprint accounting standards with a time series MRIO framework. To achieve this goal, a global scope-based carbon footprint analysis of the Turkish manufacturing sectors has been conducted as a case study. We employed a time series MRIO analysis by using the World Input–Output Database on the world’s 40 largest economies covering 1440 economic sectors. The results showed that electricity, gas and water supply was the most dominant sector in the supply chains of the Turkish industrial sectors with the largest carbon footprint. On average, indirect emissions of the Turkish manufacturing industry are found to be higher than direct emissions during the period from 2000 to 2009. The results of this analysis revealed that supply chain related indirect emissions (represented by scope 3) are responsible for nearly 56.5% total carbon emissions of sectors, which highlights the crucial role of supply chains on overall carbon footprint of sectors.

Introduction

The current environmental footprint of humanity has been growing in an unsustainable manner, which creates significant problems, one of which is the global climate change (Hoekstra and Wiedmann, 2014). Carbon dioxide (CO2), methane CH4) and nitrous oxide (N2O) are the most important greenhouse gases (GHGs) and responsible for approximately 64%, 17% and 6% of man-made global warming, respectively (European Commission, 2014). GHG emissions of industrial sectors are among the main contributors of the global climate change in the world and sustainable societies throughout the world cannot be realized without more efficient manufacturing activities that create goods and services while protecting the ecological equilibrium of the ecosystem’s goods and services (Iribarren et al., 2010, Smith and Ball, 2012). Based on a recent report published by the Intergovernmental Panel on Climate Change (IPCC, 2014), the industrial sectors consume around 19% of the total energy and contribute 30% of the total GHG emissions in 2010. Although there is a declining trend in the share of industry in global gross domestic product, GHG emissions of global industry increased from 10.4 Gt CO2-eqv to 15.5 Gt CO2-eqv during the period from 1990 to 2010 (IPCC, 2014). Therefore, environmentally friendly industrial facilities are critical for achieving a low-carbon economy, because a significant portion of GHG emissions comes from the manufacturing sectors (Hoffmann and Busch, 2008, Wang et al., 2013). In this regard, in order to stabilize the rising atmospheric emissions and mitigate the climate change impacts in the mid and long run, one of the most important policy areas that need urgent attention is the manufacturing industry.

Over the last decade, European countries have set aggressive targets regarding the minimization of GHG emissions towards a low-carbon economy (Čuček et al., 2012, Wright et al., 2011). For instance, the European Commission and a number of member states have developed adaption strategies to minimize the inevitable consequences of the climate change. For 2020, the EU has committed to cutting its GHG emissions by 20% compared to 1990 levels. Also, for 2050, EU leaders aim to reach the objective of reducing Europe’s GHG emissions by 80%–95% compared to the levels of 1990 (European Commission, 2011). In order to achieve these targets, EU member states identified the manufacturing sectors as one of the main policy areas to reduce GHG emissions most cost-effectively. In addition, Germany, which is the greatest emitter of GHG emissions in EU, has an aggressive GHG reduction target that goes well beyond the climate targets set by the EU (40% emission reduction target by 2020, compared to 1990 levels) (DG Climate Action, 2013). In parallel with the EU’s commitment on preventing climate change, the Turkish Ministry of Environment and Urban Planning has recently made the carbon footprint reporting mandatory for industrial sectors to comply with the EU’s emissions targets. According to guideline published in the Ministry’s official website, manufacturing sectors in Turkey must annually audit and report their carbon footprint starting from 2015 (Turkish Ministry of Environment and Urban Planning, 2014).

While there are solid actions for achieving a low-carbon economy, several methodological approaches have been proposed to estimate the GHG emissions related to production activities. In the literature, process-based life cycle assessment (P-LCA), input–output based LCA, and hybrid LCA (a combination of the P-LCA and input–output based LCA) are used extensively to quantify the environmental impacts of products or processes (Suh et al., 2004, Suh and Nakamura, 2007, Williams, 2004). P-LCA is capable of analyzing the environmental impacts of product life cycles with the so-called cradle to grave analysis including raw material extraction and processing, production, transportation, use, and end-of-life (Atilgan and Azapagic, 2014, Azapagic and Clift, 1999, Cuéllar-Franca and Azapagic, 2012, De Benedetto and Klemeš, 2009, Santoyo-Castelazo et al., 2011). However, when working with large-scaled systems such as industrial sectors, Input–Output (I–O) based models can be better approaches than the P-LCA and provide an economy-wide analysis (Dong et al., 2013, Egilmez et al., 2014, Guinee et al., 2010, Kucukvar et al., 2014a, Kucukvar et al., 2014b, Noori et al., 2015, Onat et al., 2014a, Onat et al., 2014b, Park et al., 2015, Rodríguez-Alloza et al., 2014, Wiedmann and Barrett, 2011). This is because P-LCA models involve the limited number of processes without tracing the entire supply chains of products, and the inclusion or exclusion of processes is decided on the basis of subjective choices, which create the so-called system boundary problem (Suh et al., 2004). Earlier studies on the direct and indirect GHG emissions of economic sectors also showed that P-LCA results suffer from significant truncation errors which can be order of 50% or higher (Feng et al., 2011, Kucukvar and Tatari, 2013, Larsen and Hertwich, 2009, Lenzen, 2000, Matthews et al., 2008).

In the literature, single region I–O models were used in order to capture the carbon footprints of large-systems at the economy-wide level (Mackie et al., 2015, Minx et al., 2009, Wiedmann, 2009a). For example, Joshi (1999) proposed an I–O based LCA model to compare the environmental performance of steel and plastic automobile tank production systems and selected GHG emissions among the environmental impact categories. Foran et al. (2005) presented a novel approach by integrating I–O analysis into the supply chains of 135 Australian sectors and chose carbon footprint as a proxy for environmental sustainability. Weber and Matthews (2008) estimated the GHG emissions of the US food production and concentrated on three key supply chain elements, i.e. food production, transportation, and final delivery. Wood and Dey (2009) focused on the direct and indirect carbon footprint of Australian sectors and identified key industry sectors in terms of economic turnover and carbon footprints. Acquaye et al. (2011) built a hybrid LCA model to complement the traditional LCA framework and estimated the indirect GHG emissions of biodiesel production in the EU. In another research, Egilmez et al. (2013) analyzed the US manufacturing sectors’ direct and indirect environmental impacts versus economic performance, where the supply chain sectors’ impact shares were found to be dominant. As a follow-up, Egilmez et al. (2014) developed a comprehensive I–O based model in order to calculate the environmental performance of the 33 U.S food manufacturing sectors including the direct and supply chain-related carbon emissions. Kucukvar et al. (2014a) used an I–O approach and analyzed the GHG emissions of the US final consumption categories including households, government consumption and investments, private fixed investment, and the export of goods and services. Overall, the results of these studies showed that total carbon emissions are driven by the manufacturing supply chains and all upstream supply chain players should be traced in order to prevent significant underestimations in the results.

Although single-region I–O models have been widely used in previous studies, Multi Region Input–Output (MRIO) models represent the state-of-the-art in the estimation of carbon footprint of production and consumption at global scale. In the past, the majority of the studies using I–O analysis are case studies of carbon footprint analysis for a single country for a single year (Hoekstra, 2010). These studies initially tried to estimate manufacturing related carbon emissions embodied in trade with the so-called domestic technology assumption. However, contemporarily, the majority of countries have open economies and they import goods and services from elsewhere. Therefore, the global carbon emissions that are embedded in imports are found in a multiple country’s I–O tables (Tukker and Dietzenbacher, 2013, Dietzenbacher et al., 2013). In this regard, MRIO models have become a widely discussed topic in the literature and they are used for regional policy making in carbon footprint analysis (Ewing et al., 2012, Hertwich and Peters, 2009, Lenzen et al., 2004, Lenzen et al., 2010). Currently, there are a number of initiatives aimed to compile large-scale global MRIOs such as Externality Data and Input–Output Tools for Policy Analysis (EXIOPOL), Global Resource Accounting Model (GRAM), Global Trade Analysis Project (GTAP), World Input–Output Database (WIOD), and EoRA (Andrew and Peters, 2013, Dietzenbacher et al., 2013, Lenzen et al., 2013, Peters et al., 2011a, Tukker et al., 2009, Bruckner et al., 2012).

Among the global MRIO databases, the WIOD has been used in numerous sustainability assessment studies. For example, Cansino et al. (2015) built a WIOD-based model in order to test the carbon footprint reduction policies of the Chinese government using a combined input–output based econometric projection approach. Timmer et al. (2015) used the WIOD database and showed its usefulness by analyzing the geographical and factorial distribution of value added in global automotive production. In other study, using a multivariate statistical analysis combined with environmentally-extended MRIO tables of the WIOD, Pascual-González et al. (2014) analyzed the environmental impact patterns of developed countries covering 69 indicators which are classified into 5 main categories: energy use, atmospheric emissions, material use, water use and land requirement. Arto et al. (2014) used the WIOD for the global carbon footprint analysis of nations and compared the results with the other MRIO database, GTAP-MRIO. Several studies also used different MRIO databases (e.g. EoRA, EXIOPOL, GRAM and GTAP) in order to capture the role of international trade for a holistic carbon footprint analysis. For example, carbon footprints of households (Ewing et al., 2012, Galli et al., 2012, Galli et al., 2013), consumption and production (Kovanda and Weinzettel, 2013, Steen-Olsen et al., 2012, Wiedmann, 2009b), international trade (Peters and Hertwich, 2008, Peters et al., 2011b, Su and Ang, 2011, Wiebe et al., 2012), and nations (Andrew et al., 2009, Hertwich and Peters, 2009, Wiedmann et al., 2010) were studied.

While MRIO models have been extensively used to estimate the global carbon footprint of production, the World Business Council for Sustainable Development (WBCSD) and the World Resource Institute (WRI) set an internationally-accepted carbon footprint accounting standard in which all possible supply-chain related indirect emissions are covered (Lin et al., 2013, Wiedmann and Barrett, 2011). According to a report published by the WBCSD and WRI, at least 80% of carbon emissions are produced in scope 3 capturing most indirect emissions in the total supply chain (WBCSD and WRI, 2009). Based on this carbon footprint accounting standard, GHG emissions are divided into three scopes. Scope 1 refers to on-site GHG emissions related to combustion of fossil fuels. Emissions resulted from purchased electricity, heat and/or steam are represented by scope 2, whereas scope 3 refers to upstream GHG emissions such as supplier emissions including indirect emissions from production of purchased materials, transportation, service inputs, etc. (WBCSD and WRI, 2004).

In the literature, only a handful of studies have used an I–O analysis for scope-based carbon footprint analysis. For example, Matthews et al. (2008) studied the scopes of carbon footprint emissions in the US sectors. The results indicated that only 26% of total GHG emissions are captured in the scope 1 and 2, and almost two-thirds of the total GHG emissions are captured under scope 3. Huang et al. (2009a) proposed a methodology in order to screen corporate carbon footprints and discussed the importance of using an I–O analysis. In another study, Huang et al. (2009b) highlighted the dominance of scope 3 carbon footprints in enterprise carbon footprint and concluded that on average more than 75% of an industry sector’s carbon footprint is attributed to scope 3 sources. Kucukvar and Tatari (2013) also develop an I–O model to capture the scope 1–3 emissions of the US construction sectors including residential, commercial and industrial buildings, heavy civil infrastructures and the repair of those sectors. In a recent work, Onat et al. (2014b) calculated the scope-based carbon footprint of US residential and commercial building sectors using a hybrid LCA approach. Similarly, the results of the aforementioned single-region I–O models showed that scope 3 emissions are mostly dominant and should be considered to prevent significant errors and incomplete results in carbon footprint analysis. However, none of the aforementioned scope-based studies analyzed the carbon footprint of the production sectors using a time series and MRIO approach considering the role of international trade for the world’s major economies. Therefore, in this paper, a global scope-based carbon footprint analysis of Turkish manufacturing is conducted as a case study. As a carbon footprint accounting framework, this paper presented the consumption-based approach in order to estimate emissions of GHGs embedded in the international trade of several countries and world regions. MRIO models are commonly chosen in consumption-based carbon footprint accounting studies since they are capable of capturing carbon footprint estimates at the multi-national level (Wiedmann, 2009b). Consumption-based accounting of GHG emissions differs from traditional, production-based inventories (limited to territorial emissions) because of imports and exports of goods and services that, either directly or indirectly, involve carbon emissions (Peters, 2008, Davis and Caldeira, 2010). According to Barrett et al. (2013), consumption-based carbon emissions are complementary to production-based inventories as they provide a complete picture of GHG emissions at the global level.

Overall, this paper aims to advance the body of knowledge on carbon footprint analysis of manufacturing sectors via a holistic approach combining the WBCSD & WRI’s scope-based carbon footprint framework with a time series MRIO framework and answer the following research questions:

(1) What are the direct and indirect GHG emissions of Turkish manufacturing sectors at the national and global scale?

Methodology: A MRIO framework is developed by using the WIOD on the world’s 40 largest economies covering 1440 economic sectors (Dietzenbacher et al., 2013). MRIO methodology has been used since Turkish is an open economy and imports the majority of goods and services from other countries such as Bulgaria, China, Germany, Iran, Korea, Russia, United States, United Kingdom, etc.

(2) What are the shares of onsite activities, electricity consumption, and upstream supply chain in the overall carbon footprint of Turkish manufacturing sectors considering their global trade links?

Methodology: Scope-based carbon footprint accounting standard set by the WBCSD and WRI (2009) is employed to understand the direct and indirect emissions based on electricity, heat and/or steam consumption and all other upstream supply chains. This framework is then linked to a MRIO framework by using the WIOD, which was funded by the European Commission under the 7th framework programme for research (Timmer, 2012, Timmer et al., 2015).

(3) What is the trend in scope 1–3 emissions for the Turkish manufacturing sectors since 2000?

Methodology: A time series MRIO framework is employed in order to analyze the effects of time on carbon footprints of the Turkish manufacturing sectors. The Analysis of Variance (ANOVA) techniques is also used to statistically validate whether the variation in total carbon footprints is significant or not (SPSS, 2013). For a time series analysis, the authors selected the 10-years period starting from 2000. Since there is no data available in the WIOD for sectoral GHG emissions after 2009, this research did not consider the carbon footprints of Turkish manufacturing sectors after 2009.

By answering the aforementioned research questions, this paper will help policymakers understand: (1) the major manufacturing sectors in Turkey with the largest direct and supply-chain related indirect GHG emissions, (2) the contribution of the categories of scope 1–3 for Turkish manufacturing sectors, (3) the geographical distribution of scope 3 emissions by country, and (4) the time-based variations of scope 1–3 emissions of each manufacturing sector.

Section snippets

MRIO modeling

MRIO modeling has a long history in the literature and has been extensively used for a regional policy making (Bon, 1984, Leontief, 1974, Miller and Blair, 2009, Peters and Hertwich, 2009, Rueda-Cantuche et al., 2009). MRIO models include trade flow matrices covering all countries or regions in the model. Hence, these matrices are able to trace the international supply chains of national economies and the global trade links among trading partners. A MRIO model typically involves national I–O

Direct and indirect carbon footprints

To get an overall understanding of the contributions of direct (onsite) manufacturing activities and supply-chain industries to the total carbon footprint impact, Egilmez et al.’s (2013) previously developed a supply chain decomposition analysis was performed. For instance, the agriculture, hunting, forestry and fishing (AHFF) industry uses inputs from the chemicals and chemical products (CCP) industry to produce its output. In this case, all the supply chain industries’ contribution to the

Conclusion and policy implications

This paper presented a scope-based carbon footprint analysis of the Turkish Manufacturing sectors considering the international trade-links between 2000 and 2009. Analysis results revealed the carbon hotspots of the manufacturing sectors and established a solid basis for future studies as well as for policy makers. Since carbon intensive hotspots of the supply chains are pointed out, the policy makers can develop more effective and sector-specific strategies to reduce these emissions. Results

References (114)

  • A. Galli et al.

    A Footprint Family extended MRIO model to support Europe’s transition to a One Planet Economy

    Sci. Total. Environ.

    (2013)
  • A. Galli et al.

    Integrating ecological, carbon and water footprint into a “footprint family” of indicators: definition and role in tracking human pressure on the planet

    Ecol. Indic.

    (2012)
  • D. Iribarren et al.

    Carbon footprint of canned mussels from a business-to-consumer approach. A starting point for mussel processors and policy makers

    Environ. Sci. Policy

    (2010)
  • J. Kovanda et al.

    The importance of raw material equivalents in economy-wide material flow accounting and its policy dimension

    Environ. Sci. Policy

    (2013)
  • M. Kucukvar et al.

    Sustainability assessment of U.S final consumption and investments: triple-bottom-line input–output analysis

    J. Clean. Prod.

    (2014)
  • H.N. Larsen et al.

    The case for consumption-based accounting of greenhouse gas emissions to promote local climate action

    Environ. Sci. Policy

    (2009)
  • J. Lin et al.

    Using hybrid method to evaluate carbon footprint of Xiamen City, China

    Energy Policy

    (2013)
  • N.C. Onat et al.

    Towards greening the US residential building stock: a system dynamics approach

    Build. Environ.

    (2014)
  • N.C. Onat et al.

    Scope-based carbon footprint analysis of US residential and commercial buildings: an input–output hybrid life cycle assessment approach

    Build. Environ.

    (2014)
  • Y.S. Park et al.

    A novel life cycle-based principal component analysis framework for eco-efficiency analysis: Case of the US manufacturing and transportation nexus

    J. Clean. Prod.

    (2015)
  • G.P. Peters

    From production-based to consumption-based national emission inventories

    Ecol. Econom.

    (2008)
  • E. Santoyo-Castelazo et al.

    Life cycle assessment of electricity generation in Mexico

    Energy

    (2011)
  • L. Smith et al.

    Steps towards sustainable manufacturing through modelling material, energy and waste flows

    Int. J. Prod. Econ.

    (2012)
  • B. Su et al.

    Multi-region input–output analysis of CO2 emissions embodied in trade. The feedback effects

    Ecol. Econ.

    (2011)
  • A. Tukker et al.

    Towards a global multi-regional environmentally extended input–output database

    Ecol. Econom.

    (2009)
  • A.A. Acquaye et al.

    Identification of ‘carbon hot-spots’ and quantification of GHG intensities in the biodiesel supply chain using hybrid LCA and structural path analysis

    Environ. Sci. Technol.

    (2011)
  • R.M. Andrew et al.

    A multi-region input–output table based on the global trade analysis project database (GTAP-MRIO)

    Econ. Syst. Res.

    (2013)
  • R. Andrew et al.

    Approximation and regional aggregation in multi-regional input–output analysis for national carbon footprint accounting

    Econ. Syst. Res.

    (2009)
  • I. Arto et al.

    Comparing the GTAP-MRIO and WIOD databases for carbon footprint analysis

    Econ. Syst. Res.

    (2014)
  • B. Atilgan et al.

    Life cycle environmental impacts of electricity from fossil fuels in Turkey

    J. Clean. Prod.

    (2014)
  • A. Azapagic et al.

    Life cycle assessment as a tool for improving process performance: a case study on boron products

    Int. J. Life Cycle Assess.

    (1999)
  • J. Barrett et al.

    Consumption-based GHG emission accounting: a UK case study

    Climate Policy

    (2013)
  • R. Bon

    Comparative stability analysis of multiregional input–output models: column, row, and Leontief–Strout gravity coefficient models

    Quart. J. Econ.

    (1984)
  • G.B. Bonan

    Forests and climate change: forcings, feedbacks, and the climate benefits of forests

    Science

    (2008)
  • S.J. Davis et al.

    Consumption-based accounting of CO2 emissions

    Proc. Natl. Acad. Sci.

    (2010)
  • DG Climate Action, 2013. Assessment of climate change policies in the context of the European Semester. Country Report:...
  • E. Dietzenbacher et al.

    Input–output analysis: the next 25~years

    Econ. Syst. Res.

    (2014)
  • E. Dietzenbacher et al.

    The construction of world input–output tables in the WIOD project

    Econ. Syst. Res.

    (2013)
  • Doğu Marmara Kalkınma Ajansı, T.C., 2013. Tekstil Sektörü Raporu. Available at:...
  • European Commission, 2011. Roadmap for a low carbon economy by 2050. Available at:...
  • European Commission, 2014. Causes of climate change. Available at:...
  • Energy Efficiency Association (EEA), 2014. 5th Region Incentives for investments on Energy Efficiency. Available at:...
  • EuroStat, 2008. Statistical classification of economic activities in the European Community....
  • EuroStat, 2010. Key indicators, manufacture of other non-metallic mineral products (NACE Division 23). Available at:...
  • K. Feng et al.

    Comparison of bottom-up and top-down approaches to calculating the water footprints of nations

    Econ. Syst. Res.

    (2011)
  • J.B. Guinee et al.

    Life cycle assessment: past, present, and future

    Environ. Sci. Technol.

    (2010)
  • E.G. Hertwich et al.

    Carbon footprint of nations: A global, trade-linked analysis

    Environ. Sci. Technol.

    (2009)
  • Hoekstra, R., 2010. Towards a Complete Overview of Peer-Reviewed Articles on Environmentally Input–Output analysis...
  • A.Y. Hoekstra et al.

    Humanity’s unsustainable environmental footprint

    Science

    (2014)
  • V.H. Hoffmann et al.

    Corporate carbon performance indicators

    J. Ind. Ecology

    (2008)
  • Cited by (67)

    View all citing articles on Scopus
    View full text