Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analyzing Canada’s ecological footprint embodied in international trade : a unidirectional multi‐regional… Kuki, Yu 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_fall_kuki_yu.pdf [ 1.1MB ]
Metadata
JSON: 24-1.0071891.json
JSON-LD: 24-1.0071891-ld.json
RDF/XML (Pretty): 24-1.0071891-rdf.xml
RDF/JSON: 24-1.0071891-rdf.json
Turtle: 24-1.0071891-turtle.txt
N-Triples: 24-1.0071891-rdf-ntriples.txt
Original Record: 24-1.0071891-source.json
Full Text
24-1.0071891-fulltext.txt
Citation
24-1.0071891.ris

Full Text

Analyzing Canada’s Ecological Footprint Embodied in International Trade:A Unidirectional Multi-Regional Input-Output ApproachbyYU KUKIB.A. Hokkaido University, 2009A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMaster of Arts in PlanninginThe Faculty of Graduate Studies THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)June, 2011©Yu Kuki 2011 ii AbstractThe  ‘Ecological  Footprint’  (EF)  of  a  specified  population  is  a  comprehensivesustainability index that estimates the ‘bio-capacity‘ (hectares of global averageproductivity)  required  to  produce  the  resources  consumed  by  that  population  andassimilate its carbon emissions. The greater the population’s material consumption andwaste production, the larger its eco-footprint (EF).The standardized method for Ecological Footprint Analysis (EFA) is maintained andregularly updated by the Global Footprint Network (GFN), a non-profit organization inCalifornia. In recent years, various EF analysts have experimented with weddingInput-Output (I-O) analysis to the standard method. I-O based models are potentiallysuperior for estimating the trade portion of the footprint because: (1) they account forcountry-specific technological efficiencies when estimating the trade component of eco-footprints (rather than world-average techno-efficiency); (2) they account for theservice-related consumption which is absent from the existing method; and (3) theyprovide more detail on the origins of the imports.This thesis contributes to I-O based ecological footprint estimates. I develop aunidirectional trade-inclusive multi-regional input-output (MRIO) model for Canada using2005 data. The results show that Canada relies for about 25% of its consumption-relatedresource needs on bio-capacity imported from other countries, compared to 44% using theGFN approach. Over 60% of Canada’s import-embodied footprint comes from the U.S. andChina. Food-related sectors including agriculture were the largest contributors to Canada’sfootprint  overseas.  Overall,  my MRIO model  yields  a  larger  EF for  Canada (9.77 gha)  thanthe GFN standard method (7.33 gha). This difference is explained by the fact that the GFNstandard method overestimates the footprint of exports for Canada (which presumably hasproduction efficiencies that are higher than world-average) and hence leading to anunderestimate of the footprint of consumption. Therefore, I conclude that while the MRIOapproach is possibly more accurate, the important finding is that the two methods mutuallyreaffirms the fact that Canadians on average use four to five times more bio-capacitycompared to their “fair share”. I discuss several policy implications of my analysis from anenvironmental, economic and social perspective using an interregional analytic framework. iii Table of Contents Abstract ........................................................................................................................................................................... ii Table of Contents ...................................................................................................................................................... iii List of Tables ................................................................................................................................................................. v List of Figures.............................................................................................................................................................. vi List of Acronyms ....................................................................................................................................................... vii Acknowledgements .............................................................................................................................................. viii Chapter 1: Introduction ......................................................................................................................................... 11.1 Purpose of this Thesis .............................................................................................................................. 11.2 Problem Statement and Rationale of the Study ........................................................................... 31.2.1 Human Prosperity and Ecosystems Decline ......................................................................... 31.2.2 Economic Growth or “Uneconomic” Growth? ...................................................................... 41.2.3 Accounting for the Earth ................................................................................................................ 71.2.4 Thinking Sustainability in an Interconnected World........................................................ 91.2.5 Calculating Embodied Resource Use in Trade .................................................................. 11 Chapter 2: Methods ............................................................................................................................................... 132.1 Review of the Existing NFA Method ................................................................................................ 132.1.1 Bio-capacity and Ecological Footprint Calculations ....................................................... 132.1.2 Shortcomings of the Existing Method ................................................................................... 182.2 Concept and Theory of the Input-Output Based Method ..................................................... 192.2.1 Conceptual Framework ................................................................................................................ 192.2.2 Brief History of the Input-Output Analysis......................................................................... 222.2.3 Input-Output Tables ...................................................................................................................... 222.2.4 Theory of I-O Based Ecological Footprint Calculation .................................................. 262.2.5 Extending the I-O Analysis to Estimate Ecological Footprints .................................. 282.3 Multi-Regional Input-Output Model ............................................................................................... 312.3.1 Three MRIO Model Scenarios.................................................................................................... 312.3.2 The Model Used in this Thesis .................................................................................................. 35 Chapter 3: Constructing the Unidirectional Trade MRIO Model ............................................... 363.1 Structure of the Model........................................................................................................................... 363.2 Data Sources............................................................................................................................................... 383.2.1 Summary ............................................................................................................................................. 383.2.2 Input-Output Tables ...................................................................................................................... 393.2.3 Bilateral Trade Database (BTD) ............................................................................................... 423.2.4 National Footprint Account (NFA) .......................................................................................... 443.2.5 Other data ........................................................................................................................................... 443.3 Assumptions and Limitations ............................................................................................................ 443.3.1 Base Year Difference Between I-O Tables and BTD ........................................................ 443.3.2 Approximations Using Proxies ................................................................................................. 453.3.3 Sector Aggregation ......................................................................................................................... 45 Chapter 4: Results .................................................................................................................................................. 464.1 Summary........................................................................................................................................................... 464.2 Ecological Footprint of Imports (EFI) ................................................................................................. 474.2.1 By Trading Partner Countries ........................................................................................................ 474.2.2 By Industrial Sectors .......................................................................................................................... 524.3 Ecological Footprint of Consumption (EFC) ..................................................................................... 54 iv 4.4 Ecological Footprint of Exports (EFE) ................................................................................................. 56 Chapter 5: Discussion and Conclusion ...................................................................................................... 585.1 Discussion ................................................................................................................................................... 585.1.1 Comparison with Existing NFA Results ................................................................................ 585.1.2 Summary on the Strengths and Weaknesses of I-O Based EFA ................................ 605.1.3 Policy Implications ......................................................................................................................... 625.2 Summary and Conclusion .................................................................................................................... 665.3 Future Research Agendas .................................................................................................................... 67 Bibliography .............................................................................................................................................................. 69 Appendix A: Major Assumptions and Limitations of the I-O Analysis ................................... 77 Appendix B: Example of Calculating Ecological Footprint Using I-O Analysis.................. 80 Appendix C: Sensitivity Analysis for RoW Category.......................................................................... 86 Appendix D: Exchange Rate Table ................................................................................................................ 89 vList of TablesTABLE 1:Pros and Cons of Using MIOTs and PIOTs for Calculation of the EF .............................. 24TABLE 2:Hypothetical 3 sector I-O Table (Million $) ............................................................................... 25TABLE 3: List of Countries and Sector Disaggregation ........................................................................... 39TABLE 4: List of Countries and their I-O Table Base Year ...................................................................... 41TABLE 5: Sector Classification of I-O Tables and BTD and Concordance with ISIC Rev.3 ...... 43TABLE 6: EFI of Canada by Trading Partner Country (Unit: gha)....................................................... 47TABLE 7: Country Share in Each Industrial Sector ................................................................................... 49TABLE 8: EFI of Canada by Industrial Sectors (Unit: gha) ..................................................................... 52TABLE 9: EFC of Canada by Industrial Sectors (Unit: gha) .................................................................... 54TABLE 10: EFE of Canada by Industrial Sectors (Unit: gha).................................................................. 56TABLE 11: Comparison of NFA approach and MRIO approach (Year: 2005) ............................... 58TABLE 12: Strengths and Weaknesses of I-O Based EFA........................................................................ 60TABLE 13: Research and Policy Questions that can be Answered Using EEI-O Analysis ....... 63TABLE 14: Hypothetical 3 sector I-O Table of Country A (Unit: Million $) .................................... 80TABLE 15: Ecological Footprint of Production (EFP) Data of Country A ....................................... 80TABLE 16: CO2 Emissions by Industrial Sector and their Share ........................................................ 81TABLE 17: Allocation of EFP to its Respective Sector .............................................................................. 81TABLE 18: Direct Footprint Intensity Matrix Calculation (Unit: gha/million $)......................... 82TABLE 19: Ecological Footprint of Domestic Consumption of Country A (Unit: gha) ............. 85TABLE 20: Scenario 1- Proxy = China (2005) .............................................................................................. 86TABLE 21: Scenario 2 - Proxy = Indonesia (2005) .................................................................................... 87TABLE 22: Scenario 3- Proxy= U.S.A (2005) ................................................................................................ 87TABLE 23: US dollar per Local Currency by Year (1997-2005) .......................................................... 89 vi List of FiguresFIGURE 1: Contrasting Worldviews (adapted from Rees, 1995)........................................................... 6FIGURE 2: Global Total Merchandise Trade ................................................................................................. 10FIGURE 3: Schematic of the Yield Factor and Equivalence Factor ..................................................... 15FIGURE 4: Existing NFA Method ........................................................................................................................ 19FIGURE 5: Input-Output Based Method (A Multi-Region Case) ......................................................... 20FIGURE 6: A schematic representation of 3 trade scenarios for a 5-region MRIO Model ...... 31FIGURE 7: Schematic of the Domestic Technology Assumption ........................................................ 32FIGURE 8: Schematic of the Unidirectional Trade MRIO Model. ........................................................ 33FIGURE 9: Schematic of Multidirectional Trade MRIO Model. ............................................................ 34FIGURE 10: Schematic of Canada’s Unidirectional Trade MRIO Model .......................................... 36FIGURE 11: Summary of Model Results (Year: 2005, Focal Country: Canada) ............................ 46FIGURE 12: EFI of Canada by Trading Partner Country Share ............................................................ 48FIGURE 13: EFI of Canada by Industrial Sector Share.............................................................................. 53FIGURE 14: EFC of Canada by Sector Share................................................................................................... 55FIGURE 15: EFE of Canada by Industrial Sector Share ............................................................................ 57FIGURE 16: Percentage of RoW to Total EFI ................................................................................................. 88 vii List of AcronymsEF Ecological FootprintEFA Ecological Footprint AnalysisEFC Ecological Footprint of ConsumptionEFE Ecological Footprint of ExportsEFI Ecological Footprint of ImportsEFP Ecological Footprint of ProductionGFN Global Footprint NetworkGHG Greenhouse GasIEA International Energy AgencyI-O Analysis Input-Output AnalysisMRIO Model Multi-regional Input-Output ModelNFA National Footprint AccountsOECD Organisation for Economic Co-operation and Development viii AcknowledgementsThis thesis would have not completed without the help of the following people. First I would like to thank my supervisor Dr. William Rees at the University of BritishColumbia for his insightful advice and for his passion for sustainability. Without hispioneering work on the ecological footprint, I would have never landed on this researchtopic  nor  had  been  inspired  to  study  at  this  school.  I  was  also  very  fortunate  to  have  awonderful second reader, Dr. Maged Senbel, who gave me constructive feedback andhelpful comments. I cannot thank more to my mentor Katsunori Iha at the Global Footprint Network (GFN) inCalifornia. During my internship at GFN, he taught me all the technical knowledge and thebasics of the model despite his busy schedule. The idea of this thesis also evolved throughmy discussions with him. This thesis would have not taken off without his assistance. Itruly respect his dedication and generosity to help others. I also thank the otherresearchers  and  staff  at  GFN,  especially  David  Moore  and  Brad  Ewing  for  providing  mewith essential data. Much appreciation goes to Ms. Kirsten Wiebe at the Institute of Economic StructuresResearch (GWS mbH) in Germany for providing me important guidance regarding thespecifics of the modeling procedures. Special  thanks  goes  to  Ms.  Farah  Kassab  for  her  continuous  encouragements,  helpfulcomments and inspiring ideas throughout the process. I had the fortune of being surrounded by many talented and motivated colleagues at theSchool of Community and Regional Planning (SCARP). I thank each and every one of themfor making my life at SCARP so enjoyable and fruitful. Last but certainly not least, I would like to thank my family back home in Japan for theirendless support. Without their help, my graduate studies would have not been possible. 1Chapter 1: Introduction 1.1 Purpose of this ThesisEcological Footprint Analysis (EFA) is a sustainability indicator that tracks humanpressures on the planet. It estimates the amount of biologically productive land and marinearea required to produce the resources that an individual, population, or activity consumes,and to absorb the waste it generates, given prevailing technology and resourcemanagement  practices  (Global  Footprint  Network,  2010a).  EFA  is  a  popular  way  tovisualize sustainability in terms of how much “bio-capacity” is required to sustain a certainpopulation and lifestyle. Among other things, this approach challenged the commonperception of cities as geographically discrete and contained places (Rees, 1992). Cities, inecological terms, “occupy” land areas orders of magnitude larger than is contained by theirjurisdictional boundaries. In other words, cities and their residents depend totally on landlocated “elsewhere” to support their consumption of goods and services (Rees, 2010a). Theprinciple  mechanism  by  which  cities  (and  many  whole  nations)  are  currently  able  tosurvive is by importing goods and services from abroad. Indeed, the ecological footprints ofnations, cities and individuals are scattered all around the globe (Kissinger & Rees, 2009a).Some key environmental policies, such as the Kyoto Protocol for GHG emissionsreduction, are based on a producer-responsibility principle (Turner, Lenzen, & Wiedmann,2007). However, producer-responsibility places disproportionate weight on producers fortheir environmental impacts by not holding consumers responsible for their choices(Lenzen, Murray, Sack, & Wiedmann, 2007; J Munksgaard, 2001). One unique feature of theEFA is that it supports a consumer-responsibility principle by enabling people to accountfor  their  resource  use  and  waste  discharges  (carbon  sink  requirements).  However,  asmentioned above, imported goods and services now constitute a significant portion of whatwe consume. Therefore, a good understanding of the resources “embodied” in imports (i.e.,the ecological footprint of trade) is becoming a key factor in making accurate estimates ofconsumption-related ecological footprints (Kissinger & Rees, 2009a; 2010; Kitzes et al.,2009; Turner, Lenzen, & Wiedmann, 2007).The current standardized ecological footprint method and national ecological footprintestimates  are  carried  out  by  the  Global  Footprint  Network  (GFN)  in  Oakland,  California.GFN annually publishes the National Footprint Accounts (NFA) for 241 countries,territories and regions using over 5,000 data points per country (Global Footprint Network, 22010a). However, there are two important omissions in the existing approach (Wiedmann,Lenzen, Turner, Minx, & Barrett, 2007): (1) Footprints embodied in traded goods are estimated using world-averageextraction rates (i.e. conversion factors). This means that computers made in theUnited States are assumed to require the same quantity of resources and energyas a computer made in China, Japan, India, etc. In reality, different countries havedifferent economic structures, technological levels, energy sources and otherfactors that result in differing quantities of inputs per unit output. (2) Trade  in  services  are  not  included  in  the  existing  approach,  resulting  in  anunderestimate of the ecological footprint of trade. The purpose of this thesis is to develop a multi-regional input-output (MRIO) model asan alternative method for estimating the ecological footprint embodied in imports. Ianalyze the ecological footprint embodied in imported goods and services to Canada in theyear 2005 (for which the most current input-output tables are available) using OECDpublished data. By taking into account the fact that every nation’s economy is differentlystructured (and therefore has different production efficiencies) this study will contribute tothe understanding of a more accurate measure of Canada’s ecological footprint.Furthermore, I hope not only to contribute to methodological refinement but also tounderstanding the complex web of inter-dependence among countries established throughtrade and its implications for the environment and geopolitics.This  research  is  being  conducted  as  part  of  a  collaborative  effort  with  the  GlobalFootprint Network. 31.2 Problem Statement and Rationale of the Study 1.2.1 Human Prosperity and Ecosystems DeclineIt is now probable that human beings are facing the biggest challenge in theircivilization’s history - degradation of several critical dimensions of its life-support systems.According to the United Nation’s Millennium Ecosystem Assessment (MEA), 60% (15 out of24) of the ecosystem services assessed in the study were being degraded or usedunsustainably. This includes fresh water, capture fisheries, air and water purification, andthe  regulation  of  regional  and  local  climate,  natural  hazards  and  pests  (MEA  2005).  Forexample, 5% to possibly 25% of global freshwater use exceeds long-term accessiblesupplies and is now met through either engineered water transfers or overdraft of groundwater supplies (MEA 2005). Flows of reactive (biologically available) nitrogen andphosphate increased by two- and four-fold respectively since 1960, helping to cause majorocean  dead  zones  in  over  400  locations  worldwide  (MEA  2005;  Diaz  et  al.  2008).  Since1750, the atmospheric concentration of carbon dioxide has increased about 40% (from280ppm to about 390ppm in 2010).Human population grew very slowly for most of human history. However, the world'spopulation quadrupled in  the 20th century to  reach 6 billion in  late  1999 and by 2006 ithad reached 6.7 billion (UNFPA, 2011). A mixture of lower mortality, improved nutritionalconditions, increased food supply, urbanization and other socio-economic factors allcontributed to this rapid population growth. Accompanying this population trend is alsothe rapid increase in material and energy consumption. Techno-industrial society isconstantly producing new products and innovative technologies to satisfy the increasingwants of the wealthier population and the expanding demand of emerging countries. Thuswhile the biophysical world is experiencing mass degradation, the wealthiest segments ofhuman society are enjoying unprecedented prosperity particularly in the last 150 to 200years since the beginning of industrial revolution. World energy consumption increased by80% during the period of 1973-2008 (IEA, 2010) despite increased efficiency fromtechnological innovation. Efficiency is being overwhelmed by exploding demand andpopulation growth (UN Department of Economic and Social Affairs, 2010).Although the world population growth rate peaked in the 1960s (US Census Bureau,2011), it is still growing at an average of 1% every year and global GDP may grow three- tosix-fold by 2050 (MEA, 2005). Most of this projected growth is expected to take place in the 4so-called developing countries which are already experiencing conflicts resulting from foodshortages, water scarcity and a changing climate (UNEP, 2011). Given the current unequaldistribution of wealth and our dysfunctional relationship with the biophysical world, manysuggest that our modern civilization is on an unsustainable path, both in social andenvironment terms (Kissinger & Rees, 2009a; Rees, 2010b). 1.2.2 Economic Growth or “Uneconomic” Growth?The term “sustainable development” was popularized by the Brundtland Report,published by the United Nations World Commission on Environment and Development (theBrundtland Commission). This report defines sustainable development as “developmentthat meets the needs of the present without compromising the ability of future generationsto meet their own needs” (WCED, 1987). In chapter 2 section 1 “The Concept of SustainableDevelopment”, the Brundtland Report also mentions that: “Meeting essential needs depend in part on achieving full growth potential, and sustainable development clearly requires economic growth in places where such needs are not being met. Elsewhere, it can be consistent with economic growth, provided the content of growth reflects the broad principles of sustainability and non-exploitation of others.” According to this passage, achieving sustainable (economic) development requireseconomic growth. However, one cannot resist suggesting that growth differs from development. Economic development implies qualitative improvement or progress, changethat achieves a set of goals that society agrees to be good (Pearce, Markandya, & Barbier,1989). This can be translated to increasing human wellbeing which may include, but is notlimited to, increasing real income per capita. On the other hand, economic growth is moreuncontroversially defined as increase in real GDP per capita, and GDP increase hashistorically been coupled to an increase in “stuff” or material possessions.However, studies like MEA join many others in suggesting that recent net gains ineconomic development have been achieved at growing costs in the form of ecosystemdegradation and unequal distribution of wealth (MEA 2005). Herman Daly, one of thefounders of the field of ecological economics, suggests we may have entered a phase of 5“uneconomic growth”; growth that destroys more wealth than it creates. Daly argues thatconventional macroeconomics fails to incorporate the notion of optimization - the “when tostop rule” (H. E. Daly, 1999). In microeconomics, there is a point in which firms andhouseholds should stop their activities. This is the point where marginal benefit equalsmarginal cost. Any activity beyond this point becomes “uneconomical”; that is, the cost ofcontinuing the activity outweighs its benefits. Why does this same rule not apply when itcomes to the entire economy? Daly and others argue that this flaw is a result of thepre-analytic vision (or ‘myth’) upon which neo-liberal economic thinking is based (H. E.Daly, 1999; Rees, 1995; 2002). Pre-analytic visions are perceptual frameworks that shape(often unconsciously) how an analyst approaches an issue. Neo-liberal economics andecological economics start from very different pre-analytic visions of the economy –ecosystem relationship.Neo-liberal economics treats the economy as an independent, self-regulating,self-sustaining entity that is not seriously constrained by the environment (Rees, 1995;2002). This vision is reflected in the vocabulary of traditional economics through such termas “externality” and “environment” which both imply that nature is somehow outside of thehuman domain (see Figure 1a). In this worldview, the “environment” is simply an externalsource of resources and a waste sink for the human society.Ecological economics on the other hand, recognizes that the economy is a growingsub-system of a larger ecosystem, which is finite and materially closed. Therefore,biophysical and thermodynamic laws ultimately determine and limit how human activities(the economy) ought to be operated and governed (see Figure 1b). This worldview analysesthe interactions between humans and nature holistically and using systems thinking. 6A. Neo-liberal Economics Worldview B. Ecological Economics Worldview FIGURE 1: Contrasting Worldviews (adapted from Rees, 1995) The former worldview has dominated and guided the economic policies of most majorgovernments and mainstream international agencies at least since the late 1970s (Rees,2002). Since mainstream economics recognizes no material limits, the focus of economicdevelopment has always been on economic growth, with little concern for ecologicaldegradation. That is why the Brundtland Report, too, saw economic growth as a solution tounsustainability, and not as a cause. Solar Energy Heat Loss Finite Ecosystem Recycling Growing Economic Subsystem Resources Wastes Infinite “Environment” Wastes Resources Growing Economy 7Studies in the economics of happiness, urban planning and many other disciplinesconfirms our intuitive notion that good environmental quality (clean air, water, access towildlife, etc.) is a major contributing factor to human wellbeing (Ng, 1993; Rehdanz &Maddison, 2005). Not only is preserving natural capital and hence environmental qualitygood  for  human  wellbeing,  it  is  now  also  necessary  for  economic  prosperity  in  an  agewhere natural capital is becoming the limiting factor of productivity (H. E. Daly, 1994). Thus,agendas for sustainable development must seek to address the problems we face withoutreaching  for  “solutions”  (e.g.  economic  growth)  that  exacerbate  the  problem.  On  a  finiteplanet with 7 billion people, expansion is no longer the solution. In Herman Daly’s words:“sustainable development is development without growth – that is, qualitativeimprovement in the ability to satisfy wants without a quantitative increase in materialthroughput beyond environmental carrying capacity” (H. E. Daly & Farley, 2004). 1.2.3 Accounting for the EarthAlthough we might be the first generations to face it on a global scale, (un)sustainabilityis not a new problem (Rees, 2002). Multiple civilizations in the past have collapsed becauseof over-exhaustion of resources, overpopulation and other reasons connected todestruction of supportive ecosystems. In his book Collapse: How Societies Choose to Fail or Succeed, Jared Diamond analyses examples such as the Mayans and the Easter Islanderswhose societies collapsed due at least in part to environmental degradation (Diamond,2005). While there are pessimists who believe that the current global society is headed tothe same destiny as these failed historical civilizations, others are optimistic. One of thestrongest proponents of the “no-limits to growth” argument was the late professor JulianSimon, who wrote in a report: “We have in our hands now--actually, in our libraries--thetechnology to feed, clothe, and supply energy to an ever-growing population for the next 7billion years.” (Simon, 1995)Extreme pessimism and optimism are equally damaging to societies’ efforts toimplement effective solutions. Both are mere states of mind. What we need is a realisticdiscussion based on empirical facts. A good starting point is to keep an accounting recordthat informs us how well (or not) we are performing on the “sustainability” scale. We arevery conscious of balancing our household, corporate and national budgets, and hencedeveloped sophisticated financial accounting tools. However, we are utterly ignorant when 8it comes to balancing the earth’s finite budget (i.e. natural resource base). Financialbankruptcy is trivial compared to ecological bankruptcy, which in the absence of anappropriate accounting method, we may not be able to foresee and avoid.Ecological Footprint Analysis (EFA) is one analytic tool that incorporates the idea ofphysical accounting. Many environmental indicators track impacts of economic activities,but not many succeed in capturing the comprehensive picture better and simpler than theEFA. EFA was invented during the 1980s by William E. Rees (1992) and further developedby Mathis Wackernagel and others at the University of British Columbia in Vancouver,Canada (Wackernagel & Rees, 1996). The essence of the concept is simple: it compares thelevel  of  human  consumption  (the  demand  side)  and  the  available  bio-capacity  of  thebiosphere (the supply side). The ecological footprint is the amount of biologicallyproductive land and marine area required to produce all the resources that an individual,population, or activity consumes, and to absorb the waste it generates, given prevailingtechnology and resource management practices (Global Footprint Network, 2010a). Intheory, the aggregated ecological footprint of all individuals on earth should be no largerthan the bio-productive land and water area of the world. As of 2007 (the most recentavailable data), however, total global ecological footprint exceeds the bio-productive landand sea area of the world by about 50% (WWF, 2010). This means that humans currentlyconsume renewable natural resources at a 50% higher rate than nature can regenerate.This state of “overshoot” can exist, at least temporarily, as humans deplete accumulatednatural capital (stocks) through unsustainable rate of harvest and extraction. In order forsociety to sustain its activity, one of the important criteria is to keep adequate physicalstocks of natural capital intact and constant on a per capita basis – we are not meeting thiscriterion. The Ecological Footprint provides policy-makers a clear metric of what actionsneed to be taken in order to address the issue. In the absence of such metric, policy-makerstend to engage in an ideological debate over the “affordability of sustainability” (GlobalFootprint Network, 2009). The concept of ecological footprint is widely used inmunicipalities and nations around the world to measure and track record of their impacton the environment. Countries like France are also considering incorporating the ecologicalfootprint concept into their measure of progress (Stiglitz & Sen, 2009). 91.2.4 Thinking Sustainability in an Interconnected WorldThere  are  two  types  of  EFA:  the  ecological  footprint  of  production  (EFP) and theecological  footprint  of  consumption  (EFC). EFP estimates the resource demands of allproduction that happens within a territorial/jurisdictional boundary of the population inquestion 1 . The EFC, on the other hand, estimates the resource demand created byconsumption activities of a specified population. This includes the demand incorporated inimported goods. In an increasingly globalizing world, more and more people consumegoods  and  services  produced  outside  their  country.  In  fact,  this  is  the  essence  ofglobalization: high mobility of people and money allowing countries to specialize inproducing what they have competitive advantage over others and importing the rest (Rees,2010b). Trade has become a major mechanism by which much of the human populationsupports its needs (Kissinger & Rees, 2009a). Of course, trade precedes even the earliestforms of civilization. However, because of mobility issues, the type of commodities and thedistances it could travel have historically been limited. Consequently, populations and theirconsumption were more or less constrained by the bio-capacity of their accessible habitats.In recent decades, because of abundant cheap energy, it has become both physically (withimproved transportation networks) and systematically (with trade liberalization) easier totrade virtually anything with anyone in the world. This trend is generally increasing (2009saw a decrease in trade due to the higher energy cost and the financial downturn - seeFigure 2). Thus, the ecological footprint of countries are now scattered across the globe in acomplex web (Kissinger & Rees, 2009a; Rees, 2010b). This current model of globaleconomic activity poses several important implications for achieving sustainability,especially from a governance perspective. 1 Every year, the Global Footprint Network (GFN) calculates the National Footprint Account (NFA) for240 countries, territories and regions. 10 FIGURE 2: Global Total Merchandise Trade First, the spatial separation between human population and the source of resourcesthey consume creates a psychological disconnect between their action and impacts. Formost of human history, people supported themselves mainly on resources and assimilativecapacities provided by the local ecosystem (Kissinger & Rees, 2009a). There was always anegative feedback mechanism in work, where degradation of ecosystems immediatelyaffected the livelihoods of the local population. In today’s world, however, most consumersare unaware of the impacts of their consumption on productive ecosystems on the otherside of the planet (Kissinger & Rees, 2009a; Rees, 2010b).Second,  the  spatial  and  psychological  disconnect  can  lead  to  false  notions  that  acountry is “decoupling” their economic growth (GDP growth) from environmental impacts.This may happen in industrialized countries that have a high proportion of their economyin  service  sectors.  GDP  is  an  aggregate  measure  of  all  production  that  happens  within  acountry. As economic structures shift towards non-manufacturing sectors, theenvironmental impacts that are associated with “domestic production” declines. However,in reality most such countries achieve an apparently low-impact economy by merelyoutsourcing their “dirty” sectors to other countries.Lastly, in today’s such globalized economy, the responsibility of environmental impactsassociated with the consumption of goods and services must be a shared between the 02 46 810 1214 1618 A m ou n t ( T ri ll io n  U S $) Year ExportsImports Data Source: World Trade Organization (WTO) 11 producers and the consumers. Traditionally, environmental policies have taken aproduction accounting principle. For example, the Kyoto protocol on GHG emissionreduction is based on a single spatial scale – nations. Although the Clean DevelopmentMechanism (CDM)2 allows transcending those scales, the flipside (i.e. increased emissionsin developing countries to satisfy demand in developed countries) are not counted for. Itbecomes especially troublesome if developed nations can shift their “dirty” sectors tocountries exempt from the protocol – a concept known as “carbon leakage”. This conceptapplies to other types of resource use and waste emissions as well. In essence,environmental policies “need new assessment tools and management tools that can fullycapture the scale of human economic activities and ecological consequences” (Kissinger &Rees, 2009a). 1.2.5 Calculating Embodied Resource Use in TradeFor global sustainability governance to properly adjust to the scale of economic activity,there first needs to be an understanding of the environmental impacts that are embodied intrade goods and services. In the absence of such understanding, it is impossible to developany policy that incorporates consumer-responsibility principles. In other words, thereneeds to be a consumption-based accounting (CBA) method that accurately measuresindirect environmental impacts so that consumer countries become aware of the negativeexternalities of their consumption activities. Unless we create such negative feedbackmechanisms, consumers remain blind to the ecological degradation that happen inproducing countries.Material flows analysis (MFA) which forms a sub-discipline of industrial ecology, hasroots in substance flow analysis (SFA, the tracking of individual substances through societyand the environment) which originally emerged from several separate initiatives (Suh,2009).  On  one  hand,  SFA  was  developed  as  a  tool  in  toxic  substances  policy  to  trace  thesources and destinations of problematic materials in the economy (Suh, 2009). On theother hand, SFA was part of the grand nutrient cycle studies conducted in ecology and earthsciences  (Suh,  2009).  In  more  recent  years,  more  comprehensive  MFA  studies  have  been 2 The CDM allows emission-reduction projects in developing countries to earn certified emissionreduction (CER) credits, each equivalent to one tonne of CO2. These CERs can be traded and sold, andused by industrialized countries to meet a part of their emission reduction targets under the KyotoProtocol (UNFCC, http://cdm.unfccc.int/about/index.html). 12 done on the embodied materials, pollutants, water and other physical measures ininternational trade (Ackerman, Ishikawa, & Suga, 2007; Bicknell, 1998; Chapagain &Hoekstra, 2007; Davis & Caldeira, 2010; Ghertner & Fripp, 2007; Giljum & Eisenmenger,2004; Giljum, Lutz, & Jungnitz, 2008; Suh, 2009; Turner, Lenzen, & Wiedmann, 2007;Wiedmann, 2009a; 2009b). These studies contribute to the understanding of complexmaterial flows in the economy through industrial linkages and trade.A substantial number of these researches employ a multi-regional input-output (MRIO)framework that combines physical data with monetary input-output tables and trade data.In this context, we can improve estimates of the trade component of the ecological footprintusing the MRIO framework (Bagliani, Galli, Niccolucci, & Marchettini, 2008; Bicknell, 1998;Kitzes et al., 2009; Wiedmann, 2009b; 2009c). This research is also part of such largemovements towards integrating the EFA and MRIO framework, using Canada as a casestudy country.The  following  chapters  will  outline  the  details  of  the  existing  method,  the  MRIOapproach and the specifics of the model. 13 Chapter 2: Methods 2.1 Review of the Existing NFA Method 2.1.1 Bio-capacity and Ecological Footprint CalculationsAs ecological footprint assessment became popular, various researchers used slightlydifferent methods and approaches for their estimates (Bicknell, 1998; Ferng, 2001;Mcdonald & Patterson, 2004). In 2003, Dr. Mathis Wackernagel started a California-basednon-profit organization called the Global Footprint Network (GFN) partially in response tothe need for a standardized EFA method. This would allow consistent monitoring and validcomparison of the results of ecological footprint studies in and of different countriesaround the world. GFN currently produces the annual National Footprint Accounts (NFA)for 241 countries, territories and regions (Global Footprint Network, 2010a).Calculating ecological footprints and bio-capacities involves disaggregating complexeconomic activities and understanding their biophysical implications. Both EFA estimatesand bio-capacity estimates require enormous amounts of data and involve manyuncertainties. Thus, like all other analytic tools, EFA starts from several key assumptions3.Each assumption is biased to underestimate human ecological impacts and overestimatebio-capacity so that studies produce conservative rather than inflated results. For instance,total human demand is underestimated because of the exclusion of freshwaterconsumption, soil erosion, greenhouse gas (GHG)  emissions other than CO2 as  well  asimpacts for which no regenerative capacity exists (e.g. pollution in terms of wastegeneration, toxicity, eutrophication, etc.) (Global Footprint Network, 2010b) On the supplyside, bio-capacity is overestimated because sustainable use is assumed. It is not possible toestimate accurately the rates of resource depletion in ways that can be handled in the EFAcalculation (Global Footprint Network, 2010b).The NFA includes six land use types: cropland, grazing land, fishing grounds, forest fortimber and fuel wood (forest land), forests for carbon uptake land (carbon footprint) andbuilt-up land4.  There is a demand on all land use types (ecological footprint), as well as asupply of each (bio-capacity) (Global Footprint Network, 2010a).  Both the ecologicalfootprint and the bio-capacity are converted to a common index – the global hectare (gha) 3 See Wackernagel et al., 2002 (page 1) for the six fundamental assumptions.4 For further details on each land use type and their respective calculation methodologies, refer toGlobal Footprint Network, 2010b (pages 8-11.) 14 using conversion factors to facilitate comparison among different EF configurations. Thefollowing sections describe the bio-capacity and ecological footprint calculation proceduresin more detail. Calculation of Bio-capacityA country’s bio-capacity (BC) (i.e. resource supply) for any land use type is calculatedas follows: ?? ? ? ? ?? ? ???                           (1) Where A is the bio-productive land area of a given country, YF and EQF are yield andequivalence factors, respectively.Yield factor and equivalence factor are both coefficients multiplied for the purpose ofnormalizing productivities across different countries and land uses. The yield factor is theratio of national average to world average yields (YP/YW). It differs by country, land use typeand year (Global Footprint Network, 2010b). It is used to correct for the differences inproductivity to allow comparison across different nations (see figure 3). The equivalencefactor  converts  the  average  productivity  of  the  respective  land  use  type  into  theirequivalent global average bio-productivity, thus allowing comparison across different landuse types (Global Footprint Network, 2010b).Equivalence factors and yield factors together translate normal hectares into globalhectares (gha), which is a hectare with world average productivity. For example, if countryA  has  X  hectare  of  arable  land  that  is  twice  as  productive  as  a  world  average  hectare  ofarable land, X hectares is translated to 2X global hectares. Conversely, if country A’s Xhectare of arable land has half the productivity of a world average hectare, X hectarestranslate to 0.5X global hectares. This allows fair comparison of bio-capacity and ecologicalfootprint size across different geographies and populations. 15 FIGURE 3: Schematic of the Yield Factor and Equivalence Factor Calculation of Ecological FootprintOn the other hand, ecological footprint (i.e. resource demand) is defined as the totalbio-productive land and water area required on a continuous basis to produce resourcesand absorb the waste of a specified population (Rees, 2006) In NFA, it is calculated bycountries. Ecological footprint tracks the annual “flow” of natural resources, rather than the“stock”. In its simplest form, the ecological footprint of a particular product (e.g. wheat) canbe expressed as the following equation: ??????? = ????????????                        (2) Where EF is the ecological footprint, D is the annual consumptions of a product (tonnes)and Y is  the  annual  yield  of  the  same  product  (tonnes/hectare).  The  aggregate  of  all  theecological footprint of different consumed products become the total ecological footprint. Yield Factor (YF) (Normalize across countries) Cropland Grazing Land Fishing Grounds Forest Land CarbonFootprint Country A Built-up Land Cropland Grazing Land Fishing Grounds Forest Land CarbonFootprint Country B Built-up Land Cropland Grazing Land Fishing Grounds Forest Land CarbonFootprint Country C Built-up Land … .. … … … .. … .. … … … .. … .. … … … .. Equivalence Factor (EQF) (Normalize across land use types) 16 This is called the ecological footprint of consumption, or in short, EFC, and is the mostcommonly reported form of the ecological footprint.However, equation (2) is a simplified formula because in reality, product-levelconsumption data is not directly available. In order to resolve this problem, thestandardized NFA method estimates the EFC by separately calculating different componentsof the consumption footprint using the following equation: ??? ? ??? ? ??? ? ???                       (3) Where EFP is the ecological footprint of production; EFI is the ecological footprint ofimports; and EFE is the ecological footprint of exports. Ecological Footprint of ProductionThe ecological footprint of production or EFP in short, is the total footprint associatedwith primary harvest and CO2 emissions that happen within a producing country’sgeographical boundary. This includes land required to produce all the primary production(cropland, grazing land, forest land and fishing grounds), land required to sequester all theCO2 emissions (carbon footprint) and land required to support all the infrastructure needsand hydropower (built-up land) (Global Footprint Network, 2010a).  EFP is expressed as: ??? = ??? ? ?? ? ???                           (4) Where P is the annual product harvested or CO2 emitted; YN is the national average yield for P; YF is the yield factor and EQF is the equivalence factor (Global Footprint Network, 2008). Ecological Footprint of Traded Products (“Embodied” footprints)Unlike the EFP which tracks only primary production5, the ecological footprint ofimports (EFI)  and  the  ecological  footprint  of  exports  (EFE)  require  the  calculation  ofecological footprints embodied in manufactured or “derived” products. In other words, in 5 Ecological  footprint  is  tallied  at  the  point  of  primary  harvest  or  carbon  emission. (Global FootprintNetwork, 2010b page 4) 17 EFP calculations,  one  only  has  to  consider  the  production  footprint  of  wheat  (a  primaryproduct), and not bread (a derived product), because calculating footprints for bread wouldlead to double counting. On the hand, traded products “embody” inputs from the countrywhere production took place. For example, unless we know how much wheat (and energy,and all other inputs) went into producing a loaf of bread, one cannot accurately assign afootprint to a derived product. For this, one needs to know the ratio of primary product perunit of derived product – a yield of derived product. In NFA, yield of derived products (YD) iscalculated by the following equation: ? ?? = ?? ? ?????                         (5) Where YP is the yield of the primary product and EXTRD is the world-average extraction rateof the derived product. Often, EXTRD is simply the mass ratio of derived product to primaryinput required. In the NFA calculations, footprints embodied in traded derived product arecalculated by multiplying the reported volume of product between nations by the footprintintensity (Global Footprint Network, 2010b): ??? (??????) = ??? ? ?? ? ???                                              = ? ?? ? ????? ? ?? ?? ? ??? =  ? ? ??? ?? ? ?????                                                (?) Where ??? ?? ? ??????   is the footprint intensity. 18 2.1.2 Shortcomings of the Existing MethodWhile the existing NFA method is a practical way of computing the ecological footprints,there are several possible improvements (Global Footprint Network, 2010a; Wiedmann,Lenzen, Turner, Minx, & John Barrett, 2007). One area that is gaining the most attention isthe calculation of footprints embodied in international trade, as trade has increasinglybecome  a  large  component  of  consumption  activities.  The  main  shortfalls  of  the  currentmethod with regards to the calculation of embodied footprints include: l The use of world-average extraction rates for manufactured and derived goodsoverestimates the ecological footprint of exports for countries with above-averageproduction efficiency (i.e. countries that are able to produce a unit of derived productwith less primary product input). On the other hand, it underestimates the ecologicalfootprint of exports for countries with below-average production efficiency6. l The omission of ecological footprints that are caused directly or indirectly by theimports and exports of services can bias the footprints of countries that are engaged inthe transaction. For example, carbon emissions are caused directly and indirectly bybanking services. If these footprints are not accounted for in the footprint of imports, itcould under-estimate the footprint of the importing country. Moreover, the current NFA method is only concerned with total imports and thus lacksthe means to examine the breakdown of where and how the  imported  products  areproduced (Wiedmann, Lenzen, Turner, Minx, & John Barrett, 2007).One strength of the standard ecological footprint is its focus on consumption. However,with the existing calculation method not accurately accounting for the trade component, itcannot fully achieve this goal (Wiedmann, 2009b). Therefore, country-specific extractionrates and service-related footprints need to be incorporated in the calculation for a moreaccurate estimation of embodied footprints. The next section outlines the input-output(I-O) approach as an alternative calculation framework that may resolves this problem. 6 Note that this is the production efficiency of derived products, not primary products. The productionefficiency  of  primary  products  also  differs  by  country  according  to  the  land  productivity,  but  they  arenormalized across countries and land use using the equivalence and yield factors. (Explained in thehectaresà global hectares conversion explanation) 19 2.2 Concept and Theory of the Input-Output Based Method 2.2.1 Conceptual FrameworkFigure 4 and 5 illustrate the conceptual differences of two methods for EFA calculation:(1) the existing NFA method and (2) the Input-Output (I-O) method. The latter I-O basedmethod, has gained much attention in recent years as the appropriate method forcalculating embodied footprints (Bicknell, 1998; Turner, Lenzen, & Wiedmann, 2007;Wiedmann, 2009b; Wiedmann, Lenzen, Turner, Minx, & John Barrett, 2007).As explained in the earlier section, the existing NFA method computes the ecologicalfootprint of consumption (EFC) by the following two calculation steps: (1) Calculate EFP, EFIand EFE separately, and (2) Add and subtract each component using the following equation:EFC = EFP + EFI - EFE. FIGURE 4: Existing NFA Method Source: Modified from Global Footprint Network, 2010a The I-O method uses a different approach. It sees each economy as an input-outputsystem where domestic inputs (EFP) and foreign inputs (EFI) are allocated to eitherdomestic consumption (EFC) or foreign consumption (EFE) (see figure 5). The actual I-O EFC EFE= +EFP -EFI Consumption Country A Economic System Production (Harvest) Domestic Bio-capacity (Direct Demand) CO2 uptake Global Bio-capacity (Indirect Demand) Exports Imports Global Bio-capacity (Direct & Indirect Demand) 20 analysis is based on money flows, but it can be extended to estimate ecological footprintflows using money flows as an indirect measure. In a multi-regional input-output model(which is explained in the last section of the chapter), country A’s imports are linked to thetrading partners’ exports.  Because of this,  country A’s EFI can be ultimately traced back tothe trading partner country’s EFP. Thus, instead of separately calculating EFP, EFI and EFE(as is the case with the NFA method), I-O method is solely focused on analyzing how andwhere the EFP is allocated. Trade is seen as an exchange of EFP. FIGURE 5: Input-Output Based Method (A Multi-Region Case)Orange Bubble: EFP, Green Bubble: EFI, Red Bubble: EFC, Blue Bubble: EFE ? ? ?? ? ? Country B Country C Country D Country X Country A Economic System 1.a 1.b 2.a 2.b (1) EFP (2) EFI 1.a+2.a + HCF EFC 1.b+2.b EFE Household Carbon Footprint (HCF) 21 Thus, for any given country A, the ecological footprint flow can be broken down to: (1) The ecological footprint associated with domestic production in country A that isembodied within:  (1.a) Products consumed in country A (1.b) Products exported from country A to other countries (2) The ecological footprint associated with production (in other countries) for exportsto country A that is embodied within:  (2.a) Products consumed in country A (2.b) Products re-exported from country A to other countries In this way it is possible to define EFP, EFI, EFC and EFE as: l EFP: (1) l EFI: (2) (= sum of trade partner countries’ EFE to country A) l EFC: (1.a) + (2.a) + Household Carbon Footprint (HCF) l EFE: (1.b) + (2.b) Therefore, it is possible to estimate a country’s EFC by summing the following: (1) theportion  of  EFP which  is  consumed  domestically  (1.a);  (2)  the  portion  of  EFI which isconsumed domestically (2.a); and (3) household carbon footprint. Household carbonfootprint must be added directly to the EFC category, because I-O analysis only captures the“indirect” footprints of consumption. Direct household CO2 emissions (heating, electricityuse, driving, etc.) are a significant portion of total emissions.This calculation is possible by using the I-O analysis. The following sections outline thefundamental  logic  and  procedures  of  the  I-O  based  calculation  method  which  forms  thebasis of the MRIO modeling presented in chapter 3. 22 2.2.2 Brief History of the Input-Output AnalysisInput-Output (I-O) analysis was originally developed in 1925 by Wassily Leontief toanalyze the money value of the complex inter-industrial exchange of goods and servicesthat happen within a national economy (Leontief, 1966; Richardson, 1972). I-O analysis isbased around a set of sectorally disaggregated economic accounts (called I-O tables) whereinputs to each industrial sector, and the subsequent uses of the output of those sectors, areseparately identified (Wiedmann, Lenzen, Turner, Minx, & John Barrett, 2007). The primaryfunction of I-O analysis is to quantify the monetary interdependence of different sectors. Itis commonly used in economic impact studies in public policy (C. Davis, 1990).However, this powerful yet simple structure of the I-O model allowed the developmentof many extended and applied uses of the I-O framework in different disciplines. One of themajor applied uses was developed among industrial ecologists and ecological economistswho realized its potential for analyzing the inter-industrial flow of materials by usingmonetary  figures  as  an  indirect  measure  or  by  using  actual  physical  accounts  (Suh  &Kagawa, 2005). Most of our so called “environmental problems” are closely connected withhow economies extract, use and dispose physical material. Understanding the structure ofthe economy that governs material and energy flows between producing industries andconsuming households is critical for regulating and governing environmental problems(Suh & Kagawa, 2005). 2.2.3 Input-Output Tables Monetary and Physical Input-Output Tables (MIOTs and PIOTs)I-O analysis is performed using a data matrix called an I-O table. It is important to drawthe distinctions between I-O tables and I-O analysis, because although I-O tables providethe data set  required to  perform the I-O analysis,  it  is  not  an operational  model  in  and ofitself (Fletcher, 1989).Traditional I-O tables were created as a monetary accounting database whichdisaggregates economic activities into different inter-sectorial monetary exchanges (seeTable 2). These I-O tables are specifically called monetary I-O tables (MIOTs) to distinguishthem from physical I-O tables (PIOTs) which were developed much more recently. PIOTswere developed in the 1990s by statistical offices of some European countries to show thephysical structure of the economy and to provide the basis for studying the 23 economic-environment relationship (Hubacek & Giljum, 2003). PIOTs emulate the MIOTs’structure and principles but are expressed in physical units (in tonnes) instead of monetaryunits (in values). The following identities hold true for the MIOTs and PIOTs, respectively: MIOT: Total Output = Total Value of Input of Goods and Services + Value Added PIOT: Total Output + Residuals (waste and emissions) = Input of Raw Materials As the equation reveals, PIOTs are based on the material balance principle in accordancewith the law of conservation of mass and therefore better resemble the physical realities ofthe economy (Hubacek & Giljum, 2003). For this reason, it is better suited for being used incombination with other physical accounting data such as the ecological footprint. Hubacekand Giljum (2003) concluded in their study, which quantified embodied ecological footprintusing both monetary and physical multipliers, that PIOT-based I-O analysis provided betterresults for two reasons: (a) The most material-intensive sectors are also the sectors with the highest ecologicalfootprint (b) Physical I-O analysis illustrates land appropriation in relation to the material flows ofeach of the sectors, which is more appropriate from the point of view ofenvironmental pressures than the monetary flows of a MIOT. However, the major limitation of the PIOT approach is data availability and consistency.Therefore, there has only been one study that calculated ecological footprints based onPIOTs (Hubacek & Giljum, 2003). Table 1 summarizes the key pros and cons of MIOTs andPIOTs. 24 TABLE 1:Pros and Cons of Using MIOTs and PIOTs for Calculation of the Ecological Footprint Pros Cons MIOTs l Abundant data l Better consistency with other data l Standardized sector aggregation l Monetary data do not always reflect physical realities PIOTs l Better accounting for physical reality l Limited data availability l Inconsistent sector aggregation For this reason, most I-O-based ecological footprint calculation studies to this day havebeen conducted using a hybrid approach that combines physical data with MIOT-basedmonetary multipliers (Bicknell, 1998; Wiedmann, 2009b; Wiedmann, Lenzen, Turner, &John Barrett, 2006; Wiedmann, Lenzen, Turner, Minx, & John Barrett, 2007; Wiedmann,Minx, J Barrett, & Wackernagel, 2006).Accepting the limitations and advantages of the MIOT, this research will use MIOTs.Thus, the following sections will outline the MIOT-based I-O analysis procedures. The Structure of MIOTsMIOTs are constructed using purchase and sales transaction data between differentsectors within a given period (usually the calendar year). As the name “input-output”suggests,  there  are  two  ways  to  interpret  the  table;  looking  at  it  as  a  table  of  salesdistribution or as a table of purchase patterns. Each sector in the model is represented asboth a  seller  and a  purchaser.  Each sector  will  buy its  inputs  from and sell  its  outputs  to,each  of  the  other  sectors  (C.  Davis,  1990).  Table  2  is  an  example  of  a  hypothetical  MIOTwith  a  simple  economy  consisting  of  only  three  sectors.  Each  component  of  the  table  iscolor  coded to  clarify  key concepts.  In  a  real  MIOT,  sectors  are obviously  divided in  muchgreater detail, depending on the purpose of the study and the availability of data. 25 TABLE 2:Hypothetical 3 sector I-O Table (Million $) Year X Sector 1 Sector 2 Sector 3 Final Demand Net Exports *1 Total Output Sector 1 10 5 5 25 20 65 Sector 2 20 30 25 15 -5 85 Sector 3 5 10 10 50 0 75 Value Added 30 40 35 Total Input 65 85 75 *1 Net Exports (NX) = Exports - Imports The rows reveal the outputs of a sector and its sales distribution. For example, sector 1produced, in total, 65 million dollars of output in year X. Of the 65 million dollars, 10million was sold to the same sector (sector 1), 5 million to sector 2 and 5 million to sector 3.This totals to 20 million dollars of sales in the intermediate demand (color coded blue).Intermediate demand refers to transactions of goods which will undergo further processingbefore it is sold as a final good (C. Davis, 1990). In addition to intermediate demand, 25million dollars were sold to the final demand sector (color coded orange), which includeshousehold purchases, government purchases and capital accumulation. The last 20 milliondollars is the net exports (color coded green), which is the value of exports minus the valueof imports. In a real situation, imported goods and services will often have its own I-O tableseparate from domestic production. However, for an aggregated I-O table, moststandardized I-O tables will have net exports added to the row.On the other hand, the columns reveal the inputs of a sector and its purchase patterns.For example, sector 1 spent, in total, 65 million dollars for inputs in year X. Of the 65 milliondollars, 10 million was purchased from the same sector (sector1), 20 million from sector 2and 5 million from sector 3. The total, 35 million dollars, is the cost of inputs that went intothe  production  (color  coded  blue).  The  rest  of  the  30  million  dollars  is  the  value  addedcomponent of inputs (color coded red) which is composed principally of wages, rents,profits, interests and taxes.I-O tables usually consist of “domestic” and “import” tables. The former includestransaction data for domestically produced goods and services, and the latter includestransaction of imported goods and services. Unless specifically noted, I-O tables usuallyrefer to the “total” I-O table, which is the sum of both the domestic and the import table. 26 2.2.4 Theory of I-O Based Ecological Footprint CalculationThe basic equation of the input-output model is (Leontief, 1966)7: ? = (? ? ?)???                         (7) Where x is the total output (input); (I-A)-1 is called the Leontief Inverse matrix; and f is  afinal demand vector. Each of these elements is explained in detail below. The quantity of the output of sector i absorbed by sector j per unit of j’s total output “j”is described by the symbol aij and is called the input coefficient or the technical coefficient ofproduct of sector i into sector j (Leontief, 1966). ??? = ??? ??? (8) For example, using the example of Table 2, the technical coefficient of product of sector 1into sector 1 is 10/65 = 0.15. If we were to do this for sector 2 (20/65 = 0.31) and sector 3(5/65 = 0.08), we get a column vector: ? = ??? ???? ?? ?? ?? ?                              (9) A is called the activity vector of sector 1. It is the “recipe” of sector 1’s output, because eachinput coefficient expresses the percentage of input required from different sectors toproduce a  unit  of  output  in  sector  1 (Leontief, 1966). When activity vectors of all sectorsare compiled into a matrix, it is expressed as an N x N matrix (N number of elements on thecolumn and row. “N” refers to the number of sectors in the I-O table) as follows: 7 See Appendix A and B for detailed methodology of the input-output model and major assumptions. 27 ? = ???? ??? ???????? ??? ????? ? ?    ? ??? ??? ???? ?                     (10) Using the Table 2 example: ? = ??? ?? ?? ?? ?? ???? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?                    (11) On the other hand, from equation (8), the following equation is derived for each elements: ??? ? ??? ? ?? (12) Expressed in a matrix, this becomes: ? ? ??                              (13) Where X = intermediate input vector, A= input coefficient vector and x= total output vector.Using the example of Table 2, equation (13) can be expressed as: ? ?? ? ? ?? ?? ?? ? ?? ?? ? = ??? ?? ?? ?? ?? ???? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?× ????? ?? ?            (14) Since total output is the sum of intermediate demand plus final demand, the following istrue: ? ? ?? ?                            (15) Where f = final demand vector (the final demand column of Table 2)From (13) and (15), the following formula can be derived: ? ? ?? ? ?                           (16) 28 When solved for x: ? = (? ? ?)???                         (17) This is the basic formula of the I-O analysis. (I-A)-1 is  a N × N constant multiplier calledthe Leontief Inverse matrix describing the direct and indirect input required in each sectorto satisfy a unit of final demand. Every economic unit (city, region, country, etc.) has itsunique economic structure and thus a unique Leontief Inverse matrix. By substituting fwith any final demand vectors (e.g., total domestic consumption, household consumption,government purchase, exports), it is possible to allocate the amount of output produced (orinput required) for different demand sectors. 2.2.5 Extending the I-O Analysis to Estimate Ecological FootprintsThis same I-O model can be extended to estimate the ecological footprint ofconsumption  through  the  conversion  of  monetary  figures  into  ecological  footprints  bymultiplying the intensity (The ratio of ecological footprint required per unit of monetaryoutput. Explained in detail below). The results can be interpreted in the exact same way asthe monetary model: the indirect ecological footprint required to fulfill the demand for oneunit of final demand. The calculation process involves the following steps (Wiedmann, Minx,J Barrett, & Mathis Wackernagel, 2006): Step 1: Allocate EFP to industrial sectors in the I-O tableStep 2: Calculate the Leontief Inverse Matrix ( (I-A)-1 )Step 3: Calculate the direct intensity matrix (EFdir)Step 4: Calculate the total intensity matrix (EFtot)Step 5: Multiply final consumption vectors to the total intensity matrix Step 1: Allocate EFP to industrial sectors in the I-O tableAllocation of the ecological footprint of production data to its respective sectors in theI-O table provides the basis of the I-O calculation. This process disaggregates the ecologicalfootprint to its related sectors in the economy. Most land types in the NFA including cropland,  grazing  land,  forest  lands  and  fishing  grounds  are  allocated  to  one  or  two  primary 29 industry sectors (e.g., agriculture, fishing), depending on the level of sector disaggregationused in the I-O table. The carbon footprint is proportionately allocated to each sector of theeconomy weighted based on the CO2 emissions data from each sector. Step 2: Calculate the Leontief Inverse Matrix ( (I-A)-1 )The Leontief Inverse matrix, as described before, is the key multiplier derived frominput coefficients extracted from I-O tables. This requires the preparation of inputcoefficients of each sector and an identity matrix with the same number of elements usedin the I-O table (e.g., if the I-O table is a 50 × 50 matrix, the identity matrix is also 50 × 50). Step 3: Calculate the direct intensity matrix (EFdir)The direct intensity matrix connects the monetary data with the ecological footprint. Itexpresses  the  ecological  footprints  that  are directly associated with the productionactivities of industrial sectors per unit monetary value of their product output (Wiedmann,Minx, J Barrett, & Mathis Wackernagel, 2006). In other words EFdir is the ratio between theecological footprint of physical inputs and monetary outputs which acts as a converterbetween the two units of measure. For example, the EFdir of sector j is expressed as: ??? ??? = ??? (???) ?? ($)?                        (18) Where EFj is the ecological footprint associated with the production activities of sector j (inglobal hectares) and xj is the total output of sector j (in monetary terms). EFdir is calculatedfor all sectors and for all land types (Hubacek & Giljum, 2003). Step 4: Calculate the total intensity matrix (EFtot)Multiplying the EFdir and the Leontief Inverse matrix results in the total intensity matrix EFtot which expresses the total (direct and indirect) ecological footprint of industrialactivities arising through the industrial supply chain to provide one unit of final demand(Wiedmann, Minx, J Barrett, & Mathis Wackernagel, 2006). ????? ? ??????? ? ?)??                        (19) 30 Step 5: Multiply final consumption vectors to the total intensity matrixOne  can  multiply  any  final  consumption  vectors  (household  consumption,  governmentspending or exports) by the total intensity matrix to estimate the total ecological footprintattributed to the respective consumption category. For example, in the case of estimatingecological footprints embodied in exports, the following equation is used: ??? ? ????? ? ??                           (20) Where EFE is the ecological footprint of exports and ex is the export demand vector. EFtotconverts monetary units to ecological footprints (in global hectares) and also calculates theindirect ecological footprints required to produce ex (in other words, the “embodied”ecological footprints). An example of this calculation step is given in Appendix B. 31 2.3 Multi-Regional Input-Output Model 2.3.1 Three MRIO Model ScenariosMulti-regional input-output (MRIO) models build on the basic single-country I-Oframework and extend it to a multi-region case. There are several different scenarios (or“levels of detail”) of MRIO models that treat the imported products differently. The maindifference  between  the  models  is  how  accurately  one  follows  the  complex  thread  ofinternational trade flows. There are three different levels of MRIO models known in theliterature, diagrammatically represented in figure 7. These are: (a) domestic technologyassumption (DTA); (b) unidirectional trade MRIO model (also called “linked single-regionmodel”); and (c) multidirectional trade MRIO model (also called “full MRIO model”)(Andrew, Peters, & Lennox, 2009a; Wiedmann, Lenzen, Turner, & John Barrett, 2006)8. FIGURE 6: A schematic representation of three trade scenarios for a five-region MRIO model.Arrows represent trade flows with the circular arrow representing the DTA.(Modified and adapted from Lenzen, Pade, & Munksgaard, 2004) The  country  from  whose  perspective  the  analysis  is  carried  out  is  referred  to  as  the“focal” country (country A in figure 7 and Canada in the actual model) and all other tradingpartner countries are referred to as the “non-focal” countries (country B,C,D,E in figure 7)(Andrew, Peters, & Lennox, 2009a). 8 For a full literature review of different MRIO modeling, refer to Wiedmann, Lenzen, Turner, & Barrett,2006. (a)Domestic technologyAssumption (DTA) A (c)MultidirectionalTrade MRIO Model B C D E A (b)UnidirectionalTrade MRIO Model B C D E A 32 (a) Domestic Technology Assumption (DTA)While the single-country I-O model omits the effect of trade all together, the domestictechnology assumption (DTA) is a slight improvement by assuming that imported goodsand services are produced using domestic technology. For example, a computer imported tocountry A is assumed to be produced using country A’s technology level (see figure 8). Thisrequires no foreign data, which greatly saves on resources but at the expense of accuracy9(Andrew,  Peters,  &  Lennox,  2009a).  Moreover,  this  model  does  not  improve  the  existingNFA calculation method because it does not reflect the differences in technological levels ofthe trading partners. FIGURE 7: Schematic of the Domestic Technology AssumptionCircular arrow represents the DTA. (b)Unidirectional Trade MRIO ModelThe unidirectional trade MRIO model (or “linked single-region model”) exogenouslylinks different national I-O tables using the bilateral trade data (Wiedmann, Lenzen, Turner,& John Barrett, 2006). This model assumes that only the “1st tier” of trade-linkages is active.That is, the focal country trades with non-focal countries, but the non-focal countries do nottrade with each other (Andrew, Peters, & Lennox, 2009a) (see figure 9). This approachcaptures only the last stage of an international supply chain of imports (Wiedmann, Lenzen,Turner,  &  John  Barrett,  2006).  For  example,  a  computer  produced  in  country  C  may  be 9 According to Andrew et al. (2009), DTA over- or underestimated the carbon footprints of countries byanywhere between 41-336% depending on the focal country (see table 1,2 and 3 of Andrew, Peters, &Lennox, 2009b). Country A EFP EFI EFC EFE 33 exported to country A through country B. In which case, with the unidirectional trade MRIOmodel, it will be assumed that the production happened in country B, rather than incountry C. FIGURE 8: Schematic of the Unidirectional Trade MRIO Model.Blue arrows represent trade flows with the circular arrow representing the DTA. Country B Country C Country D Country E Country A EFP EFI EFC EFE 34 (c) Multidirectional Trade MRIO ModelThe multidirectional trade MRIO model (or “full MRIO model”) endogenously combinesdomestic technical coefficient matrices with import matrices from multiple countries orregions into one large coefficient matrix10, thus capturing trade supply chains between alltrading partners (Wiedmann, Lenzen, Turner, & John Barrett, 2006). This means, forexample, that a computer exported to country A from country C through country B will beproperly assigned the technological level of the country of origin (in this case, country C)when calculating the embodied resource and energy use. FIGURE 9: Schematic of Multidirectional Trade MRIO Model.Blue arrows represent trade flows.Orange Bubble: EFP, Green Bubble: EFI, Red Bubble: EFC, Blue Bubble: EFE 10 See Andrew, Peters, & Lennox, 2009a (pages 329-330) for detailed mathematical explanations. Country C Country E Country B Country D Country A 35 2.3.2 The Model Used in this ThesisWhile  the  multidirectional  trade  model  is  the  ideal  model  for  capturing  the  mostaccurate picture of the world trade flows, it requires large amounts of data which are oftennot available (E. G. Hertwich & Peters, 2010). Therefore in most cases analysts have usedsome form of approximation to a full MRIO model (Andrew, Peters, & Lennox, 2009a; E. G.Hertwich & Peters, 2010). In their study of embodied carbon footprints using differenttrade scenarios, Lenzen et al. (2004) and Andrew et al. (2009) have both found that aunidirectional trade MRIO model gives a good approximation of the multidirectional tradeMRIO model – i.e. it does not introduce significant errors11. The unidirectional trade MRIOmodel reduces the data requirement considerably, which also allows calculation innon-specialist software such as Excel (Andrew, Peters, & Lennox, 2009b). Thus, I chose theunidirectional trade MRIO model as appropriate model for this thesis. 11 Lenzen et al (2004) demonstrate that in the case of a 5-region MRIO model analyzing the embodiedcarbon footprints to Demark, a unidirectional model gave very similar results to a full MRIO model(about 1-2% difference) (see table 7 of their paper). Similarly, Andrew et al. (2009) found that aunidirectional trade MRIO model with multiple scenarios using different number of regions gave similarresults to the full MRIO model (about 1-9% difference) (see table 4 of their paper). 36 Chapter 3: Constructing the Unidirectional Trade MRIO Model 3.1Structure of the ModelThe objective of the model is to estimate Canada’s ecological footprint of consumption(EFC) by estimating the ecological footprint of imports (EFI)  using  I-O  analysis  andaccounting for the different technological levels of countries. This is illustrated in figure 10. FIGURE 10: Schematic of Canada’s Unidirectional Trade MRIO Model For estimating each EFI and EFC, the calculations involve the following steps: Step 1: Calculating EFIEcological  footprint  embodied  in  imports  from  country  B  (non-focal)  to  country  A(focal) is expressed as: ??? ????? ? ??)??????                                                            (?) Where ?????? is the direct ecological footprint intensity of country B; ?? ? ??)?? is the ? ? ? ? ? ? Country B Country C Country D Country X Canada 1.a 1.b 2.a 2.b (1) EFP (2) EFI 1.a+2.a + HCF EFC 1.b+2.b EFE Household Carbon Footprint (HCF) Direct HH emissions 37 Leontief Inverse matrix of country B; and ???? is the bilateral trade data from country B tocountry A. Since the ecological footprint embodied in all imported products to country A isa sum of the embodied footprints from each of the trading partners, EFI is expressed as: ??? =????????? ? ??)??????? ?                                              (?) Where w is the number of countries. Step 2: Calculating EFCEFC is derived by summing (2.a) and (1.a) (see formulas below) and carbon footprint ofdirect household emissions that are added separately as non-tradable footprints. This canbe expressed as: ??? ? ? ??? ? ? ??? ? ? ???? ??????                                                 (?) Where ????  is the ecological footprint of consumption of country A; ????  is A’secological footprint of consumption from consumed imported products (2.a); ???? is A’secological footprint of consumption from consumed domestic products (1.a); and ??????????is the carbon footprint of direct household emissions. Step 2.1 EFC from Imported Products (2.a)To calculate 2.a, which is the estimated embodied footprints in the portion of importedproducts that is consumed domestically, one must allocate the EFI to  the  import  table  ofcountry A. I-O tables consist of domestic and import tables (see chapter 2, page 16). Thus,the technical coefficient matrix A can also be decomposed as: ?? ? ??? ? ???                                                                 (?) Where ?? is the total technical coefficient matrix of country A; ??? is the domestictechnical coefficient of country A; and ??? is the import technical coefficient of country A. 38 Thus, in mathematical terms this is expressed as: ??? = ???? =???????(? ? ???)????? ?                                        (?) Where ?????? is the direct ecological footprint intensity of country B; w is  the number ofcountries; (? ? ???)?? is  the  Leontief  Inverse  matrix  of  imported  products  of  country  A;and ??is the domestic final demand vector of country A. Step 2.2 EFC from Domestically Produced Products (1.a)Calculating the ecological footprint of consumption from domestic products (1.a) isstraightforward. Using the domestic I-O table and EFP data, it is expressed as: ?? ? ? ??? ? ? ???????? ? ???)????                                               (?) Where ?????? is the direct ecological footprint intensity of country A; (? ? ???)?? is theLeontief Inverse matrix of domestic products of country A; and ??is the domestic finaldemand vector of country A. 3.2 Data Sources 3.2.1 SummaryThe core data used for the construction of the unidirectional trade MRIO model are: (1)OECD Input-Output Tables for Canada (both domestic and import I-O tables) and its tradingpartners (total I-O tables); (2) OECD Bilateral Trade Data (BTD) between Canada and itstrading partners; and (3) NFA Ecological Footprint of Production (EFP) data for allcountries.  Most  data  used  in  this  model  are  similar  to  those  used  in  the  CO2 emissionsmodel developed by Ahmad & Wyckoff (2003), Nakano et al. (2009) and the GlobalResource Accounting Model (GRAM) developed by Giljum et al. (2008). Summary ofcountries and sectors in the model are presented in Table 3. 39 TABLE 3: List of Countries and Sector Disaggregation List of Countries List of Sectors Focal Country Sector DescriptionCanada 1 Agriculture, hunting, forestry and fishing2 Mining and quarrying Non-focal Countries 3 Food products, beverages and tobacco1 Argentina 4 Textiles, textile products, leather and footwear2 Australia 5 Wood and products of wood and cork3 Austria 6 Pulp, paper, paper products, printing and publishing4 Belgium 7 Coke, refined petroleum products and nuclear fuel5 Brazil 8 Chemicals excluding pharmaceuticals6 China 9 Pharmaceuticals7 Denmark 10 Rubber and plastics products8 Finland 11 Other non-metallic mineral products9 France 12 Iron & steel10 Germany 13 Non-ferrous metals11 Greece 14 Fabricated metal products, except machinery and equipment12 India 15 Machinery and equipment, n.e.c.13 Indonesia 16 Office, accounting and computing machinery14 Ireland 17 Electrical machinery and apparatus, n.e.c.15 Israel 18 Radio, television and communication equipment16 Italy 19 Medical, precision and optical instruments17 Japan 20 Motor vehicles, trailers and semi-trailers18 South Korea 21 Building & repairing of ships and boats19 Mexico 22 Aircraft and spacecraft20 Netherlands 23 Railroad equipment and transport equipment n.e.c.21 New Zealand 24 Manufacturing n.e.c; recycling (include Furniture)22 Norway 25 Electricity, water and gas supply23 Poland 26 Services24 Portugal25 Russia26 South Africa27 Spain28 Sweden29 Switzerland30 Turkey31 United Kingdom32 United States of America33 Rest of the World (RoW) 3.2.2 Input-Output TablesMost economies in the world have national statistical institutes (NSIs) responsible forcollecting and producing important economic data. I-O tables are published more or less ona regular basis by many NSIs around the world. However, these tables differ in data quality,sector disaggregation, currencies and base year, which make consistent MRIO modelingdifficult. Most preferably, these datasets are provided by one source using the sameassumptions  for  data  harmonization  procedures  (Giljum,  Lutz,  &  Jungnitz,  2008).  The 40 Global Trade Analysis Project (GTAP)12 and the Organisation for Economic Co-operationand Development (OECD)13 are the two main institutions that present such internationaldatasets of harmonized I-O tables.GTAP offers the most extensive data for MRIO modeling, covering 113 countries/regionsand 57 sectors (GTAP, 2010). Many trade embodiment studies use the GTAP in their model(S. J. Davis & Caldeira, 2010; Muñoz & Steininger, 2010; Peters & E. Hertwich, 2007). GTAPdatabase is constructed by voluntary data contributions from individuals, institutions andnations. While this mechanism allows extensive coverage, there are questions concerningthe consistency and transparency of data source and their harmonization process (Andrew,Peters, & Lennox, 2009b; Giljum, Lutz, & Jungnitz, 2008).OECD, on the other hand, offers I-O tables for 44 countries and 48 sectors covering theyears 1995, 2000 and 2005 or nearest years under the Structural Analysis (STAN) database(STAN Input-Output Tables, 2010). The latest sets of I-O tables (around year 2005) includeall OECD members except for Iceland and 11 non-OECD major economies includingArgentina, Brazil, China, India, Indonesia, Romania, Russia, South Africa, Taiwan, Thailandand Vietnam. OECD I-O tables are compiled by first requesting NSIs to provide data inaccordance with a harmonized industry structure based on the International StandardIndustrial Classification of all Economic Activities (ISIC). ISIC Revision 3 provides the basisfor the latest version of the dataset (Yamano & Ahmad, 2006)14. Since ISIC reporting is notmandatory,  most  countries  have  chosen  to  deliver  data  using  their  own  industrialclassification systems. Thus, the second step of compilation involves harmonizing each datato  conform  to  the  OECD  system15. Because harmonization is undertaken by only oneinstitution, the OECD I-O tables are considered more reliable and transparent than GTAP(Giljum, Lutz, & Jungnitz, 2008). Also, despite its lower number of country coverage, thegeographies represented in the OECD I-O tables cover about 66% of world population, and 12 GTAP (Global Trade Analysis Project) is a global network of researchers and policy makersconducting quantitative analysis of international policy issues. The GTAP project is coordinated by theCenter for Global Trade Analysis, Purdue University, USA.(https://www.gtap.agecon.purdue.edu/default.asp)13 OECD (Organisation for Economic Co-operation and Development) is an international economicorganization of 34 countries founded in 1961 to stimulate economic progress, world trade, and providea platform for solutions to common problems.(http://www.oecd.org/home/0,2987,en_2649_201185_1_1_1_1_1,00.html)See Wixted, Yamano, & Webb (2006)} for more detail on the OECD I-O tables. 14 See http://unstats.un.org/unsd/cr/registry/regcst.asp?Cl=17 for ISIC Rev.3 description15 See Yamano & Ahmad (2006) for detailed harmonization procedures. 41 90% of world GDP (Yamano & Ahmad, 2006).This model uses the latest OECD I-O tables of Canada and 32 trading partner countries(see Table 4). TABLE 4: List of Countries and their I-O Table Base Year Country OECD/Non-OECD Base Year1 Argentina Non-OECD 19972 Australia OECD 2004/20053 Austria OECD 20044 Belgium OECD 20045 Brazil Non-OECD 20056 Canada OECD 20057 China Non-OECD 20058 Denmark OECD 20049 Finland OECD 200510 France OECD 200511 Germany OECD 200512 Greece OECD 200513 India Non-OECD 2003/200414 Indonesia Non-OECD 200515 Ireland OECD 200516 Israel OECD 200417 Italy OECD 200418 Japan OECD 200519 South Korea OECD 200520 Mexico OECD 200321 Netherlands OECD 200522 New Zealand OECD 2002/200323 Norway OECD 200524 Poland OECD 200425 Portugal OECD 200526 Russia Non-OECD 200027 South Africa Non-OECD 200528 Spain OECD 200429 Sweden OECD 200530 Switzerland OECD 200131 Turkey Non-OECD 200232 United Kingdom OECD 200533 United States of America OECD 200534 RoW (U.S.A as proxy) 2005 To close the model on a global scale, an aggregated “Rest of the World (RoW)” categoryis also required. Since world-average I-O tables are not available, most studies approximatethe RoW by using proxy I-O tables of different countries. For example, Giljum et al. (2008)uses the I-O table of Argentina, Ahmad and Wyckoff (2003) uses United States and Nakano et al. (2009) uses Indonesia. This model uses the United States (a country with relatively 42 low  footprint  intensity)  as  a  proxy  country  for  the  RoW  category,  which  producesconservative results16. 3.2.3 Bilateral Trade Database (BTD)Bilateral trade database (BTD) for commodities are also available from the OECD STANdatabase for all OECD member countries with over 70 partner countries or regions17. BTDonly captures OECD trade with the rest  of  the world,  and therefore does not  record tradebetween two non-OECD countries (Giljum, Lutz, & Jungnitz, 2008). BTD is disaggregatedinto 28 sectors using the ISIC classification code, which allows consistent sectorharmonization between the OECD I-O tables. It is noted that the BTD does not cover alltrade transactions happening around the world. However, when considering data for OECDcountries, it covers roughly 90% of world trade (STAN Bilateral Trade Database, 2010).Trade data for services, on the other hand, are published separately as “trade in services”under the OECD International Trade and Balance of Payments database. It supplements theservice sector trade data which is missing from the BTD. All services are aggregated intoone service sector, which is one of the major limiting factors of this database.Table 5 shows the sector classification of I-O tables and BTD and their concordance withthe ISIC Revision 3 code. The shaded areas indicate the sectors which were aggregatedduring the harmonization process. As a result, the number of sectors is reduced to 26 18.Also, figures in the I-O tables are published in local currencies while BTD is published in USdollar terms. Harmonization between the two datasets requires currency conversion,which is done using OECD yearly average exchange rate statistics19. 16 See Appendix C for RoW sensitivity analysis. 17 See OECD STI Division of Economic Analysis and Statistics (2009)} for more details on BTD.18 Several sectors in the I-O table needed aggregation and adjustment in order to be consistent with theBTD sector classification. Namely, “Mining and Quarrying (energy)” and “Mining and Quarrying(non-energy)”  was  aggregated  to  “Mining  and  Quarrying”;  “Production,  collection  and  distribution  ofelectricity”, “Manufacture of gas; distribution of gaseous fuels through mains” and “Steam and hot watersupply” was aggregated to “Electricity, water and gas supply”; and all service sectors (sector 29-48)were aggregated to “Services”.19 See Appendix D for a list of yearly average exchange rates (local currency/USD). 43 TABLE 5: Sector Classification of I-O Tables and BTD and Concordance with ISIC Rev.3 ISIC Rev.3 Code I-O Sector BTD Sector Sector Description1+2+5 1 1 Agriculture, hunting, forestry and fishing10+11+12 2 2 Mining and quarrying (energy)13+14 3 2 Mining and quarrying (non-energy)15+16 4 3 Food products, beverages and tobacco17+18+19 5 4 Textiles, textile products, leather and footwear20 6 5 Wood and products of wood and cork21+22 7 6 Pulp, paper, paper products, printing and publishing23 8 7 Coke, refined petroleum products and nuclear fuel24ex2423 9 8 Chemicals excluding pharmaceuticals2423 10 9 Pharmaceuticals25 11 10 Rubber and plastics products26 12 11 Other non-metallic mineral products271+2731 13 12 Iron & steel272+2732 14 13 Non-ferrous metals28 15 14 Fabricated metal products, except machinery and equipment29 16 15 Machinery and equipment, not elsewhere classified (n.e.c)30 17 16 Office, accounting and computing machinery31 18 17 Electrical machinery and apparatus, n.e.c32 19 18 Radio, television and communication equipment33 20 19 Medical, precision and optical instruments34 21 20 Motor vehicles, trailers and semi-trailers351 22 21 Building & repairing of ships and boats353 23 22 Aircraft and spacecraft352+359 24 23 Railroad equipment and transport equipment n.e.c.36+37 25 24 Manufacturing n.e.c; recycling (include Furniture)401 26 25 Production, collection and distribution of electricity402 27 25 Manufacture of gas; distribution of gaseous fuels through mains403 28 25 Steam and hot water supply41 29 26TradeinServicesData Collection, purification and distribution of water45 30 Construction50+51+52 31 Wholesale and retail trade; repairs55 32 Hotels and restaurants60 33 Land transport; transport via pipelines61 34 Water transport62 35 Air transport63 36 Supporting & auxiliary transport activities; activities of travel agencies64 37 Post and telecommunications65+66+67 38 Finance and insurance70 39 Real estate activities71 40 Renting of machinery and equipment72 41 Computer and related activities73 42 Research and development74 43 Other Business Activities75 44 Public administration and defense; compulsory social security80 45 Education85 46 Health and social work90-93 47 Other community, social and personal services95+99 48 Private households with employed persons & extra-territorialorganizations & bodies 44 3.2.4 National Footprint Account (NFA)National Footprint Account (NFA) data are available from the Global Footprint Network(GFN) for 241 countries, territories and regions from 1961 to 2007 (GFN, 2010)20. There isa three year lag between the data year and the version year due to time-lag of the sourcedata (i.e.  NFA2008 is  based on 2005 data).  NFA are produced using yield and productiondata provided by major international institutions such as the UN Comtrade database, Foodand Agriculture Organization (FAO) and the International Energy Agency (IEA). This modeluses the NFA2008 (2005 data) EFP data for all countries in the model because the latest I-Otables are only available for years around 2005. 3.2.5 Other dataCO2 emission data from the International Energy Agency (IEA) are used when allocatingcarbon footprints to the respective sectors in the I-O table. National Accounting Matrix withEnvironmental Accounts (NAMEA) data are used for determining the ratio of directhousehold CO2 emissions and industry CO2 emissions.  For  countries  where  NAMEA  datawere not available, the ratio of the United Kingdom was used as a proxy21. 3.3 Assumptions and LimitationsThe sections that follow mainly focus on the assumptions used to develop the dataneeded in the analysis. The more general assumptions of the I-O analysis are listed inAppendix B. 3.3.1 Base Year Difference Between I-O Tables and BTDIdeally, all data points in the model need to be chronologically aligned to produce themost accurate results. However, due to different national statistical cycles and frequencies,it is often difficult to obtain data that covers the same year.The  core  calculation  in  this  model  involves  multiplying  BTD  and  the  Leontief  Inversematrices obtained from I-O tables. Although BTD is available for the same base year (2005)for all countries in the model, only roughly 80% (27 out of 34 countries) of I-O tables were 20 EFC data is freely available from GFN’s website, while EFP and EFI data are licensed. Free licenses areavailable for academic purposes. 21 U.K. ratio was closest to the average ratio. 45 available for years around 2005 (2004 and 2005). For the rest of 20% (7 countries), latesttables were only available from years between 2000 and 2003, except for Argentina whichonly  has  I-O  tables  from  1997.  I-O  tables  reflect  the  structure  of  the  economy,  and  thusdiffer more or less every year depending on various socio-economic and political factors.The farther apart the years are between I-O tables and BTD, the more likely they are toproduce inaccurate results.This model assumes that 2 to 3 year differences do not significantly distort the results.Even in cases where significant year lag is seen (Argentina in particular) influences areconsidered relatively minor given their relative importance to the total imports. 3.3.2 Approximations Using ProxiesIn cases where specific data are not available for a certain country, proxy measures areused. Namely, the economic structures of the rest of the world (RoW) category is proxied bythe United States; the ratio of household and industry CO2 emissions for Argentina,Australia, Brazil, China, Finland, Greece, India, Indonesia, Israel, South Korea, Mexico,Russia, South Africa and Turkey are proxied by the ratio of the United Kingdom (75%industry emissions and 25% household emissions). 3.3.3 Sector AggregationService sectors needed to be aggregated to a single sector during the sectorharmonization process between I-O tables and BTD. This is one of the major limitations ofthis model in terms of data. In general, “the more sectors the model can identify thestronger the analysis will be as more interdependencies between sectors that are distinct intheir production technologies can be quantified.” (Wiedmann, Lenzen, Turner, Minx, & JohnBarrett, 2007). However, it is better to include an aggregated service sector than to omit itall together, as is the case with the existing NFA calculation method. This point is explainedin further detail in the “homogeneous sectors” section in Appendix B. 46 Chapter 4: Results 4.1 SummaryFigure 12 below summarizes the results of each type of footprint (EFP, EFI, EFC and EFE)of Canada on a diagram. FIGURE 11: Summary of Model Results (Year: 2005, Focal Country: Canada) EFP was 12.51 global hectares (gha) per capita in 2005 (NFA 2005). 1.68gha per capitaof  carbon  footprint  was  first  separated  from  the  EFP and directly allocated to domesticconsumption as non-tradable direct household emissions (mostly residential energy useand personal automobile use). Of the rest, 10.83gha per capita, 5.73gha per capita wasconsumed domestically through consumption of goods and services and 5.08gha per capitawas exported to other countries. The ecological footprint embodied in imports (EFI) fromall trading partner countries totaled 2.54gha per capita, of which 2.36gha per capita wasconsumed domestically and 0.18gha per capita was re-exported. As a result, the ecologicalfootprint of consumption (EFC) totaled 9.77gha per capita and the ecological footprint ofexports (EFE) totaled 5.26gha per capita. Country A Country B Country X Trading Input EF (1) 12.51gha/capita EFP (2) 2.54gha/capita EFI Output 5.73+2.36+1.68 =9.77gha/capita EFC EFE 5.08+0.18 =5.26gha/capita Household Carbon Footprint 1.68gha/capita Canada Economic System (Input-Output Table) (1.a) 5.73gha/capita (2.b) 0.18gha/capita 47 4.2 Ecological Footprint of Imports (EFI) 4.2.1 By Trading Partner CountriesEFI was estimated by summing the embodied ecological footprint in the imports fromeach trading partner to Canada. Table 6 and Figure 13 show the results of the EFI estimatesby trading partner countries and their contribution to the total imported embodiedfootprint. TABLE 6: EFI of Canada by Trading Partner Country (Unit: gha) *Population of Canada in 2005: 32,359,000 people Country Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Total Share 1 Argentina 444,055 112,123 34,448 48,034 375,356 1,014,017 1.24% 2 Australia 295,665 327,500 98,771 8,202 257,569 987,707 1.20% 3 Austria 19,449 3,369 27,234 1 44,859 94,912 0.12% 4 Belgium 33,207 7,734 21,618 1,512 385,131 449,201 0.55% 5 Brazil 805,344 861,865 588,027 30,036 566,825 2,852,098 3.47% 6 China 4,037,731 952,144 924,654 721,934 8,580,038 15,216,501 18.53% 7 Denmark 68,584 166 8,236 16,362 159,555 252,903 0.31% 8 Finland 32,081 113 201,363 5,670 293,063 532,291 0.65% 9 France 258,674 21,702 87,361 23,388 291,347 682,472 0.83% 10 Germany 155,619 1,040 72,412 21,513 251,502 502,085 0.61% 11 Greece 25,168 1,849 1,344 1,195 106,972 136,529 0.17% 12 India 614,541 6,207 190,768 32,902 663,592 1,508,009 1.84% 13 Indonesia 523,016 13,294 257,407 241,623 288,977 1,324,317 1.61% 14 Ireland 13,381 17,873 9,534 10,720 61,640 113,148 0.14% 15 Israel 5,330 172 68 1,057 17,225 23,852 0.03% 16 Italy 130,244 5,511 13,376 7,944 272,347 429,422 0.52% 17 Japan 17,203 6 12,375 56,986 711,800 798,370 0.97% 18 South Korea 18,682 93 5,380 40,187 664,129 728,471 0.89% 19 Mexico 702,375 402,530 262,995 124,162 1,694,218 3,186,279 3.88% 20 Netherlands 38,015 8,282 4,985 23,747 413,948 488,977 0.60% 21 New Zealand 29,265 216,456 215,975 148,011 61,439 671,145 0.82% 22 Norway 25,988 1,644 94,530 430,220 633,470 1,185,851 1.44% 23 Poland 42,999 98 23,580 2,780 394,453 463,911 0.57% 24 Portugal 10,190 5,853 27,063 7,722 70,151 120,978 0.15% 25 Russia 211,433 3,948 196,974 44,221 2,114,288 2,570,865 3.13% 26 South Africa 102,859 50,026 105,011 30,063 44,887 332,847 0.41% 27 Spain 53,157 8,242 15,881 23,279 171,330 271,888 0.33% 28 Sweden 24,711 1,382 241,928 3,461 59,338 330,819 0.40% 29 Switzerland 12,877 8,475 24,332 76 7,289 53,048 0.06% 30 Turkey 108,625 6,793 16,104 7,489 418,813 557,823 0.68% 31 UK 90,784 16,054 16,255 16,323 166,458 305,874 0.37% 32 USA 15,109,983 864,900 9,653,777 985,654 9,272,997 35,887,311 43.71% 33 RoW 3,801,769 217,614 2,428,952 247,997 1,329,604 8,025,936 9.78% Total EF 27,863,003 4,145,058 15,882,720 3,364,469 30,844,609 82,099,860 *Per Capita EF 0.86 0.13 0.49 0.10 0.95 2.54 48 FIGURE 12: EFI of Canada by Trading Partner Country Share (From largest to smallest) The  results  show  that,  in  2005,  about  44%  of  Canada’s  imported  EF  came  from  theUnited States, followed by China with about 19%. These two countries alone constitutemore than 60% of Canada’s imported footprint. This seems reasonable given Canada’sstrong economic ties with these countries. In actual monetary value, United States andChina’s share of Canada’s total imports in 2005 were 56% and 8%, respectively.The rest  of  the world (RoW) combined (for  which I-O tables  are  not  available  and areproxied  by  technological  levels  of  the  U.S.)  contribute  about  10%  of  Canada’s  importedfootprint.  However,  applying  U.S.  technology  levels  to  the  RoW  category  most  likely 49 underestimates  the  actual  size  of  the  footprint;  implying  that  both  the  total  EFI and  theshare of RoW are conservative figures. Availability of I-O tables for important tradingpartner countries such as the OPEC countries and Southeast Asian countries would providea more accurate picture of Canada’s ecological dependence.Other large sources of imported footprint include countries with which Canada has freetrade agreements like Mexico (NAFTA) or large natural resource exporting (agriculturalproducts, fuel and minerals, etc.) countries like Brazil and Russia. Table 7 presents the data in finer resolution by showing country share within eachindustrial sector. Some countries that are less represented in the total share standout inspecific sector share. For example, 26% of the “Mining and Quarrying” sector footprintcomes from Norway, about 20% of “Non-Ferrous Metals” from Australia,  and over 50% of“Radio, Television and Communication Equipment” from South Korea and Japan. TABLE 7: Country Share in Each Industrial Sector 1 2 3 4 5 6 7 Agriculture, Hunting, Forestry and Fishing Mining and Quarrying Foodproducts, Beverages and Tobacco Textiles, Textile Products, Leather and Footwear Wood and Products of Wood and Cork Pulp, Paper, Paper Products, Printing and Publishing Coke, Refined Petroleum Products and Nuclear Fuel 1 Argentina 1.73% 0.18% 1.74% 0.28% 1.48% 0.05% 0.79% 2 Australia 0.43% 0.53% 4.47% 0.13% 0.04% 0.15% 2.74% 3 Austria 0.00% 0.01% 0.09% 0.03% 0.83% 0.42% 0.00% 4 Belgium 0.04% 0.46% 0.43% 0.10% 0.46% 0.30% 16.21% 5 Brazil 3.33% 0.44% 8.83% 1.69% 7.26% 1.77% 2.88% 6 China 1.84% 0.40% 4.77% 62.52% 19.49% 17.97% 3.53% 7 Denmark 0.03% 0.84% 0.60% 0.06% 0.02% 0.11% 0.11% 8 Finland 0.03% 0.00% 0.16% 0.05% 0.96% 7.94% 15.34% 9 France 0.11% 0.03% 2.26% 0.20% 0.25% 1.19% 1.31% 10 Germany 0.13% 0.02% 0.75% 0.18% 1.10% 0.64% 0.59% 11 Greece 0.02% 0.09% 0.17% 0.04% 0.00% 0.03% 0.01% 12 India 0.83% 0.09% 1.56% 7.85% 1.52% 0.58% 2.33% 13 Indonesia 3.35% 0.04% 0.44% 1.29% 2.09% 1.49% 0.00% 14 Ireland 0.01% 0.06% 0.35% 0.03% 0.01% 0.05% 0.00% 15 Israel 0.02% 0.02% 0.01% 0.02% 0.00% 0.01% 0.08% 16 Italy 0.10% 0.19% 0.88% 0.46% 0.14% 0.42% 3.40% 17 Japan 0.04% 0.02% 0.06% 0.06% 0.01% 0.26% 0.38% 18 South Korea 0.03% 0.00% 0.09% 0.37% 0.02% 0.39% 0.42% 19 Mexico 4.72% 0.93% 1.02% 2.51% 0.35% 1.41% 0.24% 20 Netherlands 0.28% 0.07% 0.33% 0.35% 0.04% 0.60% 2.94% 21 New Zealand 0.45% 0.00% 3.74% 0.11% 0.04% 0.08% 0.00% 22 Norway 0.00% 26.44% 0.07% 0.01% 0.00% 0.00% 3.18% 23 Poland 0.01% 0.00% 0.35% 0.27% 2.41% 0.25% 0.00% 24 Portugal 0.02% 0.00% 0.25% 0.13% 0.47% 0.05% 0.67% 25 Russia 0.04% 50.94% 2.18% 0.19% 0.56% 0.00% 2.60% 26 South Africa 0.98% 0.21% 0.43% 0.03% 0.02% 0.00% 0.21% 27 Spain 0.18% 0.23% 0.38% 0.14% 0.22% 0.29% 3.33% 28 Sweden 0.02% 0.16% 0.67% 0.07% 0.68% 2.01% 2.22% 29 Switzerland 0.00% 0.00% 0.19% 0.01% 0.00% 0.02% 0.00% 30 Turkey 0.21% 0.01% 0.41% 1.29% 0.00% 0.03% 6.73% 31 UK 0.05% 0.89% 0.53% 0.04% 0.02% 0.21% 0.48% 32 USA 61.35% 6.93% 52.37% 11.14% 52.80% 59.90% 18.25% 33 RoW 19.64% 9.77% 9.42% 8.33% 6.72% 1.38% 9.02% Country Industry 50 8 9 10 11 12 13 14 Chemicals excluding Pharmaceuticals Pharmaceuticals Rubber and Plastics Products Other Non- Metallic Mineral Products Iron and Steel Non-Ferrous Metals Fabricated Metal Products 1 Argentina 0.34% 2.06% 0.20% 0.30% 9.18% 29.01% 0.12% 2 Australia 0.55% 19.60% 0.18% 0.09% 0.26% 19.46% 0.17% 3 Austria 0.13% 0.00% 0.06% 0.55% 0.20% 0.00% 0.16% 4 Belgium 1.15% 0.00% 0.38% 0.67% 1.54% 0.00% 0.18% 5 Brazil 1.37% 3.43% 0.91% 6.35% 7.47% 31.56% 0.92% 6 China 10.44% 0.00% 26.99% 0.00% 14.72% 0.00% 47.78% 7 Denmark 0.28% 0.00% 0.13% 0.30% 0.18% 0.00% 0.22% 8 Finland 0.35% 0.00% 0.85% 0.30% 0.94% 0.00% 0.34% 9 France 2.00% 0.00% 0.60% 1.43% 1.51% 0.00% 0.83% 10 Germany 2.28% 0.00% 0.63% 0.87% 1.42% 0.00% 0.75% 11 Greece 0.34% 0.00% 0.14% 0.18% 0.08% 0.00% 0.07% 12 India 5.06% 41.53% 1.85% 3.85% 1.69% 1.51% 4.01% 13 Indonesia 0.77% 2.80% 0.91% 2.67% 0.37% 0.77% 0.73% 14 Ireland 0.36% 0.00% 0.05% 0.15% 0.05% 0.00% 0.30% 15 Israel 0.07% 1.45% 0.06% 0.03% 0.01% 0.00% 0.07% 16 Italy 0.79% 0.00% 0.55% 3.75% 0.86% 0.00% 0.48% 17 Japan 1.54% 6.63% 1.47% 1.21% 1.16% 1.92% 0.84% 18 South Korea 2.55% 1.26% 1.99% 0.45% 2.25% 4.28% 1.21% 19 Mexico 1.46% 0.00% 2.87% 3.42% 1.43% 0.00% 4.71% 20 Netherlands 1.34% 0.00% 0.57% 0.95% 0.52% 0.00% 0.51% 21 New Zealand 0.74% 0.00% 0.09% 0.06% 0.44% 0.00% 0.05% 22 Norway 0.00% 1.28% 0.01% 0.03% 0.00% 1.45% 0.46% 23 Poland 0.40% 0.00% 0.33% 1.49% 2.00% 0.00% 0.89% 24 Portugal 0.01% 0.00% 0.43% 0.32% 0.04% 0.00% 0.12% 25 Russia 2.66% 0.00% 0.00% 0.06% 4.48% 10.05% 0.00% 26 South Africa 0.00% 0.00% 0.05% 0.00% 0.60% 0.00% 0.00% 27 Spain 0.57% 0.00% 0.73% 1.12% 1.25% 0.00% 0.18% 28 Sweden 0.14% 0.00% 0.26% 0.14% 1.11% 0.00% 0.54% 29 Switzerland 0.42% 0.00% 0.02% 0.04% 0.01% 0.00% 0.03% 30 Turkey 0.37% 0.00% 0.41% 11.15% 6.31% 0.00% 0.53% 31 UK 0.55% 19.95% 0.22% 0.29% 0.21% 0.00% 0.18% 32 USA 58.78% 0.00% 53.01% 54.42% 33.25% 0.00% 29.28% 33 RoW 2.18% 0.00% 3.02% 3.37% 4.46% 0.00% 3.32% 15 16 17 18 19 20 21 Machinery and Equipment Office, Accounting and Computing Machinery Electrical Machinery and Apparatus, nec Radio, Television and Communication Equipment Medical, Precision and Optical Instruments Motor Vehicles, Trailers and Semi- Trailers Building and Repairing of Ships and Boats 1 Argentina 0.17% 0.06% 0.07% 0.14% 0.15% 0.16% 0.00% 2 Australia 0.40% 0.00% 0.16% 0.00% 2.74% 0.11% 2.42% 3 Austria 0.25% 0.13% 0.09% 0.83% 0.39% 0.21% 0.00% 4 Belgium 0.63% 1.60% 0.15% 1.26% 0.54% 0.56% 0.00% 5 Brazil 1.73% 0.07% 1.24% 3.00% 0.56% 1.61% 0.05% 6 China 24.58% 61.51% 32.13% 0.00% 60.66% 5.93% 0.00% 7 Denmark 0.51% 0.24% 1.80% 2.08% 1.53% 0.15% 5.11% 8 Finland 1.49% 1.13% 0.32% 0.95% 1.95% 0.82% 0.04% 9 France 1.38% 1.42% 0.72% 4.36% 2.25% 0.11% 0.99% 10 Germany 1.72% 0.53% 0.57% 1.53% 2.77% 1.20% 3.84% 11 Greece 0.02% 0.05% 0.07% 0.02% 0.10% 0.01% 0.00% 12 India 1.62% 0.31% 1.07% 1.29% 1.37% 0.32% 0.00% 13 Indonesia 0.11% 0.00% 0.51% 12.90% 5.96% 0.05% 2.09% 14 Ireland 0.35% 0.11% 0.05% 0.60% 1.19% 0.01% 0.03% 15 Israel 0.03% 0.00% 0.06% 0.37% 0.06% 0.01% 0.00% 16 Italy 1.54% 0.30% 0.28% 1.16% 1.85% 0.23% 0.38% 17 Japan 1.85% 2.27% 0.84% 21.66% 4.92% 3.33% 0.19% 18 South Korea 1.98% 1.77% 1.00% 32.39% 1.94% 3.72% 6.48% 19 Mexico 6.70% 6.12% 17.11% 0.00% 0.00% 8.89% 0.00% 20 Netherlands 1.33% 3.12% 0.90% 0.95% 3.66% 0.04% 0.44% 21 New Zealand 0.34% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 22 Norway 2.92% 0.02% 0.04% 1.10% 0.37% 0.01% 0.00% 23 Poland 1.45% 2.79% 0.69% 6.17% 0.89% 0.08% 0.53% 24 Portugal 0.09% 0.68% 0.03% 3.52% 0.05% 0.02% 0.06% 25 Russia 0.04% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 26 South Africa 0.00% 0.00% 0.08% 0.24% 0.00% 0.06% 0.00% 27 Spain 0.38% 0.27% 0.24% 0.71% 1.29% 0.18% 0.05% 28 Sweden 1.53% 0.25% 0.28% 0.00% 0.91% 0.58% 0.04% 29 Switzerland 0.07% 0.01% 0.00% 0.12% 0.22% 0.01% 0.00% 30 Turkey 0.38% 2.02% 0.10% 0.32% 0.83% 0.12% 0.49% 31 UK 0.48% 0.22% 0.23% 2.31% 0.86% 0.16% 10.43% 32 USA 41.89% 8.69% 35.74% 0.00% 0.00% 70.56% 56.55% 33 RoW 2.05% 4.32% 3.46% 0.00% 0.00% 0.76% 9.79% Country Industry Country Industry 51 However, since unidirectional trade MRIO models do not account for trade between twonon-focal countries, some figures in Table 7 may not attribute the footprint to its propersource.  For  example,  some  products  may  be  exported  from  China  to  Canada  through  theUnited States. In that case, the unidirectional trade MRIO model attributes the source of thefootprint to the United States instead of China, where the production actually took place. Toaccurately identify the origin of the footprint, a more sophisticated multidirectional tradeMRIO model is required. 22 23 24 25 26 Aircraft and Spacecraft Railroad and Transport Equipment, nec Manufacturing nec; Recycling Electricity and Gas Services 1 Argentina 0.00% 8.46% 0.15% 0.00% 0.34% 2 Australia 3.45% 2.83% 0.32% 0.00% 1.77% 3 Austria 0.00% 0.00% 0.11% 0.00% 0.21% 4 Belgium 0.00% 0.00% 0.39% 0.00% 0.45% 5 Brazil 0.00% 0.00% 0.48% 0.00% 0.68% 6 China 0.00% 0.00% 73.94% 0.00% 9.08% 7 Denmark 0.00% 0.00% 0.13% 0.00% 0.14% 8 Finland 0.00% 0.00% 0.16% 0.00% 1.10% 9 France 0.00% 0.00% 0.28% 0.00% 1.89% 10 Germany 0.00% 0.00% 0.23% 0.00% 0.92% 11 Greece 0.00% 0.00% 0.02% 0.00% 3.03% 12 India 30.06% 4.90% 0.00% 0.00% 2.43% 13 Indonesia 0.03% 1.13% 2.32% 0.00% 0.97% 14 Ireland 0.00% 0.00% 0.03% 0.00% 0.46% 15 Israel 0.00% 0.00% 0.08% 0.00% 0.05% 16 Italy 0.00% 0.00% 0.52% 0.00% 0.74% 17 Japan 45.68% 80.62% 0.58% 0.00% 1.37% 18 South Korea 4.17% 0.00% 0.32% 0.00% 0.70% 19 Mexico 0.00% 0.00% 4.63% 0.00% 1.34% 20 Netherlands 0.00% 0.00% 0.14% 0.00% 1.88% 21 New Zealand 0.00% 0.00% 0.02% 0.00% 0.87% 22 Norway 0.00% 0.00% 0.01% 0.00% 0.71% 23 Poland 0.00% 0.00% 1.55% 0.00% 0.70% 24 Portugal 0.00% 0.00% 0.02% 0.00% 0.30% 25 Russia 0.00% 0.00% 0.56% 0.00% 4.17% 26 South Africa 0.00% 0.00% 0.19% 0.00% 0.34% 27 Spain 0.00% 0.00% 0.09% 0.00% 0.56% 28 Sweden 0.00% 0.00% 0.21% 0.00% 0.54% 29 Switzerland 0.00% 0.00% 0.01% 0.00% 0.38% 30 Turkey 0.00% 0.00% 0.39% 0.00% 0.92% 31 UK 16.61% 2.06% 0.04% 0.00% 1.34% 32 USA 0.00% 0.00% 10.21% 100.00% 46.62% 33 RoW 0.00% 0.00% 1.90% 0.00% 12.97% Country Industry 52 4.2.2 By Industrial SectorsTable 8 and Figure 13 summarize the EFI estimates by industrial sectors and their share.The two sectors “Agriculture, Hunting, Forestry and Fishing” and “Food products, Beveragesand Tobacco” combined total to about 45% of the imported eco-footprint. The “Agriculture,Hunting, Forestry and Fishing” sector is only 1.7% of the total imports in monetary value,but make up more than a quarter of the total imported footprint.Significant portion of the imported footprint also comes from the textile products(7.8%), motor vehicles (6.4%), manufactured goods (5.4%) and mining and quarrying(4.6%) sectors. Manufactured and processed products embody high levels of energy andmaterials input and have consequently large footprints (carbon footprint in particular). TABLE 8: EFI of Canada by Industrial Sectors (Unit: gha) *Population of Canada in 2005: 32,359,000 people Sectors Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Total Share 1 Agriculture, Hunting, Forestry and Fishing 12,808,528 1,471,855 7,712,116 1,157,441 398,255 23,548,195 28.7% 2 Mining and Quarrying 135,563 12,641 169,743 429,462 3,028,525 3,775,935 4.6% 3 Food products, Beverages and Tobacco 6,506,424 1,318,807 3,958,663 657,626 782,992 13,224,512 16.1% 4 Textiles, Textile Products, Leather and Footwear 2,426,903 478,100 760,622 364,277 2,342,843 6,372,746 7.8% 5 Wood and Products of Wood and Cork 908,512 117,029 555,695 82,176 684,976 2,348,389 2.9% 6 Pulp, Paper, Paper Products, Printing and Publishing 390,413 37,792 312,717 35,140 503,800 1,279,862 1.6% 7 Coke, Refined Petroleum Products and Nuclear Fuel 51,745 11,287 66,505 13,689 744,076 887,301 1.1% 8 Chemicals excluding Pharmaceuticals 470,051 48,442 264,999 45,826 1,105,492 1,934,809 2.4% 9 Pharmaceuticals 15,810 4,310 4,655 2,457 92,948 120,180 0.1% 10 Rubber and Plastics Products 313,538 32,856 175,572 36,689 1,059,396 1,618,050 2.0% 11 Other Non-Metallic Mineral Products 33,767 5,545 19,554 3,517 445,248 507,631 0.6% 12 Iron and Steel 124,135 26,157 75,522 15,431 1,369,637 1,610,881 2.0% 13 Non-Ferrous Metals 13,002 11,055 7,802 1,400 319,839 353,098 0.4% 14 Fabricated Metal Products 126,508 21,981 54,256 18,210 1,134,888 1,355,843 1.7% 15 Machinery and Equipment 361,903 54,395 218,210 49,650 2,917,068 3,601,226 4.4% 16 Office, Accounting and Computing Machinery 332,439 71,159 102,574 56,149 2,305,327 2,867,647 3.5% 17 Electrical Machinery and Apparatus, nec 208,058 41,132 84,616 31,411 2,049,145 2,414,361 2.9% 18 Radio, Television and Communication Equipment 13,821 1,369 7,899 11,617 312,462 347,168 0.4% 19 Medical, Precision and Optical Instruments 61,760 13,940 24,166 12,772 970,659 1,083,297 1.3% 20 Motor Vehicles, Trailers and Semi-Trailers 667,800 70,044 423,608 72,496 4,018,269 5,252,218 6.4% 21 Building and Repairing of Ships and Boats 5,612 396 3,294 585 72,595 82,482 0.1% 22 Aircraft and Spacecraft 2,604 485 568 984 178,260 182,901 0.2% 23 Railroad and Transport Equipment, nec 481 33 257 862 75,048 76,681 0.1% 24 Manufacturing nec; Recycling 920,646 194,271 294,383 158,209 2,845,760 4,413,269 5.4% 25 Electricity and Gas 3,829 219 2,446 250 36,325 43,070 0.1% 26 Services 959,149 99,757 582,279 106,147 1,050,776 2,798,108 3.4% Total EF 27,863,003 4,145,058 15,882,720 3,364,469 30,844,609 82,099,860 *Per Capita EF 0.86 0.13 0.49 0.10 0.95 2.54 53 FIGURE 13: EFI of Canada by Industrial Sector Share 54 4.3 Ecological Footprint of Consumption (EFC)Table  10  and  Figure  15  summarize  the  EFC results which sum the domesticconsumption  of  domestically  produced  goods  and  services  (1.a  of  Figure  11)  and  thedomestic consumption of imported goods and services (2.a of Figure 11).Consumption of various services (35%), food and beverage products (20%),agricultural and marine products (15%) and direct household energy use (17%) are thefour dominant factors of Canada’s EFC. The “services” sector includes, for example, sectorslike “hotels and restaurants” which requires a lot of cropland and grazing land, and “airtransport” which contributes to the carbon footprint22.  Overall,  the carbon footprint is thelargest component, contributing to about 46% of the total eco-footprint, while croplandand forest land together contribute another 48%. TABLE 9: EFC of Canada by Industrial Sectors (Unit: gha) *Population of Canada in 2005: 32,359,000 people 22 See page 43 Table 5 for more detailed breakdown of the “services” sector. Sectors Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Built-up Land Total Share 1 Agriculture, Hunting, Forestry and Fishing 21,491,879 2,638,123 20,845,631 2,310,710 958,285 6,840 48,251,468 15.3% 2 Mining and Quarrying 55,268 6,673 55,307 5,886 1,328,069 10,152 1,461,355 0.5% 3 Food products, Beverages and Tobacco 24,221,885 2,420,425 31,990,952 2,323,002 2,610,980 56,957 63,624,201 20.1% 4 Textiles, Textile Products, Leather and Footwear 341,632 49,621 213,224 40,641 219,397 4,302 868,818 0.3% 5 Wood and Products of Wood and Cork 780,773 78,231 1,027,961 74,987 122,874 1,608 2,086,435 0.7% 6 Pulp, Paper, Paper Products, Printing and Publishing 798,023 81,609 1,025,324 77,483 1,651,095 12,362 3,645,895 1.2% 7 Coke, Refined Petroleum Products and Nuclear Fuel 174,561 22,027 160,098 19,073 3,388,881 23,247 3,787,887 1.2% 8 Chemicals excluding Pharmaceuticals 108,997 14,615 86,722 12,348 490,304 3,082 716,069 0.2% 9 Pharmaceuticals 107,372 13,784 94,859 11,851 233,460 3,247 464,573 0.1% 10 Rubber and Plastics Products 292,663 42,484 183,034 34,803 487,870 2,429 1,043,282 0.3% 11 Other Non-Metallic Mineral Products 19,775 2,753 14,167 2,292 149,889 468 189,344 0.1% 12 Iron and Steel 28,638 4,215 17,021 3,435 273,155 (1,187) 325,276 0.1% 13 Non-Ferrous Metals - - - - - - - 0.0% 14 Fabricated Metal Products 37,465 5,115 28,398 4,291 216,311 1,632 293,212 0.1% 15 Machinery and Equipment 175,983 23,638 139,396 19,957 615,626 10,247 984,846 0.3% 16 Office, Accounting and Computing Machinery 22,845 3,398 13,034 2,758 72,115 4 114,154 0.0% 17 Electrical Machinery and Apparatus, n.e.c 18,181 2,436 14,489 2,059 72,501 822 110,488 0.0% 18 Radio, Television and Communication Equipment 64,771 9,317 41,816 7,659 196,967 1,214 321,744 0.1% 19 Medical, Precision and Optical Instruments - - - - - - - 0.0% 20 Motor Vehicles, Trailers and Semi-Trailers 887,861 127,774 572,340 105,018 2,613,735 17,366 4,324,094 1.4% 21 Building and Repairing of Ships and Boats 6,672 846 6,055 731 25,494 537 40,336 0.0% 22 Aircraft and Spacecraft 23,475 3,481 13,560 2,829 88,802 71 132,218 0.0% 23 Railroad and Transport Equipment, n.e.c 42,507 5,805 32,207 4,869 167,492 1,525 254,405 0.1% 24 Manufacturing n.e.c; Recycling 495,630 58,454 517,384 52,075 1,845,425 9,973 2,978,941 0.9% 25 Electricity and Gas 106,604 11,624 125,858 10,718 14,043,975 23,490 14,322,269 4.5% 26 Services 19,707,864 1,987,959 25,743,054 1,899,550 60,564,716 1,466,261 111,369,406 35.2% Direct Household Consumption - - - - 54,465,165 -                    54,465,165 17.2% TOTAL 70,011,324 7,614,408 82,961,891 7,029,023 146,902,585 1,656,650 316,175,881 *Per Capita EFc 2.16 0.24 2.56 0.22 4.54 0.05 9.77 55 FIGURE 14: EFC of Canada by Sector Share 56 4.4 Ecological Footprint of Exports (EFE)Table 10 and Figure 15 summarize the EFE results representing the sum of the foreignconsumption of domestically produced goods and services (1.b of Figure 12) and theforeign consumption of imported goods and services (2.b of Figure 12).“Agriculture,  Hunting,  Forestry  and  Fishing”  (35%),  “Wood  and  products  of  Wood  andCork” (18%) and “Food products, Beverages and Tobacco” (16%) are the three biggestsectors that export Canadian bio-capacity. “Mining and Quarrying” (6%) (includes oil, gasand coal) is also a substantial part of foreign footprints on Canada, ccounting for 25% of thetotal  exported  carbon  footprint.  These  facts,  along  with  the  fact  that  Canada  earnssubstantial income from natural resources (about 30% including energy), suggest thatforeign demand on Canada’s bio-capacity can not only jeopardize Canada’s environmentbut also its long-term economic-base unless it is monitored and managed sustainably. TABLE 10: EFE of Canada by Industrial Sectors (Unit: gha) *Population of Canada in 2005: 32,359,000 people Sectors Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Total Share 1 Agriculture, Hunting, Forestry and Fishing 21,694,113 1,905,992 32,677,254 1,947,371 859,151 59,083,881 34.74% 2 Mining and Quarrying 302,776 26,029 464,866 26,887 9,852,977 10,673,534 6.28% 3 Food products, Beverages and Tobacco 9,852,010 851,936 15,049,441 877,428 868,282 27,499,098 16.17% 4 Textiles, Textile Products, Leather and Footwear 57,595 5,878 74,183 5,586 146,946 290,187 0.17% 5 Wood and Products of Wood and Cork 11,113,192 951,947 17,115,038 985,147 1,157,592 31,322,917 18.42% 6 Pulp, Paper, Paper Products, Printing and Publishing 1,797,253 154,726 2,755,979 159,714 2,746,962 7,614,634 4.48% 7 Coke, Refined Petroleum Products and Nuclear Fuel 83,689 8,926 101,871 8,313 1,962,019 2,164,818 1.27% 8 Chemicals excluding Pharmaceuticals 263,860 23,730 389,026 23,964 1,935,536 2,636,115 1.55% 9 Pharmaceuticals 82,983 7,792 117,289 7,704 233,593 449,360 0.26% 10 Rubber and Plastics Products 230,300 25,632 263,913 23,419 1,822,246 2,365,509 1.39% 11 Other Non-Metallic Mineral Products 29,273 2,726 41,724 2,706 325,326 401,755 0.24% 12 Iron and Steel 163,290 15,121 234,051 15,052 3,619,911 4,047,426 2.38% 13 Non-Ferrous Metals - - - - - - 0.00% 14 Fabricated Metal Products 87,962 8,436 121,617 8,256 495,191 721,462 0.42% 15 Machinery and Equipment 140,180 13,688 190,057 13,281 573,795 931,002 0.55% 16 Office, Accounting and Computing Machinery 18,147 1,841 23,547 1,754 46,767 92,056 0.05% 17 Electrical Machinery and Apparatus, n.e.c 49,004 4,686 67,961 4,593 189,543 315,787 0.19% 18 Radio, Television and Communication Equipment 108,014 10,605 145,555 10,263 305,914 580,351 0.34% 19 Medical, Precision and Optical Instruments - - - - - - 0.00% 20 Motor Vehicles, Trailers and Semi-Trailers 548,294 51,362 776,846 50,840 2,099,318 3,526,660 2.07% 21 Building and Repairing of Ships and Boats 4,035 355 6,070 363 16,908 27,732 0.02% 22 Aircraft and Spacecraft 64,674 6,744 81,099 6,346 224,983 383,845 0.23% 23 Railroad and Transport Equipment, n.e.c 35,261 3,277 50,363 3,256 137,423 229,579 0.13% 24 Manufacturing n.e.c; Recycling 468,263 42,207 688,938 42,576 1,597,178 2,839,162 1.67% 25 Electricity and Gas 14,498 1,234 22,456 1,281 2,387,391 2,426,859 1.43% 26 Services 1,650,908 149,293 2,421,411 150,355 5,088,868 9,460,835 5.56% TOTAL 48,859,574 4,274,162 73,880,555 4,376,455 38,693,819 170,084,566 *Per Capita EFE 1.51 0.13 2.28 0.14 1.20 5.26 57 FIGURE 15: EFE of Canada by Industrial Sector Share 58 Chapter 5: Discussion and Conclusion 5.1 Discussion 5.1.1 Comparison with Existing NFA ResultsIn  this  section,  I  will  compare  the  results  from  the  MRIO  model  to  the  existing  NFAestimates. Table 11 compares the EFC, EFI and EFE estimates from both the existing NFAapproach and the MRIO approach. TABLE 11: Comparison of NFA approach and MRIO approach (Year: 2005) Results from the MRIO approach were about 30% higher than NFA for EFC, about 20%lower  for  EFI,  and  about  40%  lower  for  EFE. The higher EFC of  the  MRIO  result  is  mostlyattributable to the significantly lower EFE (i.e., larger portion of the production footprint ofCanada  attributed  to  domestic  demand  than  were  assumed  in  the  NFA).  This  result  maysuggest that trade-related footprints in the NFA method were overestimated for both theimports and exports due to the use of world-average technological levels. This hypothesis isalso supported by the fact  that  EFE results yielded larger deviation (40% difference) thanthe EFI results (20%) because domestic technology level of Canada is likely much higher EF Consumption TOTAL 7.33gha 9.77gha Cropland 1.53 21% 2.16 22% Grazing Land 0.26 4% 0.24 2% Forest Land 1.32 18% 2.56 26% Fishing Grounds 0.17 2% 0.22 2% Carbon Footprint 4 55% 4.54 46% Built-up Land 0.05 1% 0.05 1% EF Imports TOTAL 3.25gha 2.54gha Cropland 0.48 15% 0.86 34% Grazing Land 0.08 2% 0.13 5% Forest Land 0.58 18% 0.49 19% Fishing Grounds 0.16 5% 0.11 4% Carbon Footprint 1.95 60% 0.95 37% EF Exports TOTAL 8.44gha 5.26gha Cropland 1.77 21% 1.51 29% Grazing Land 0.06 1% 0.13 2% Forest Land 3.62 43% 2.28 43% Fishing Grounds 0.24 3% 0.14 3% Carbon Footprint 2.75 33% 1.2 23% NFA MRIO 59 than the world-average. This deviation is consistent with GFN’s own assessment of theweakness of their method, which says: “using world-average efficiencies for all tradedgoods is an overestimate of the footprint of exports for countries with higher-than-averageproduction efficiency. In turn, it underestimates that country’s footprint of consumption”(Global  Footprint  Network,  2010a).  This  is  exactly  what  has  been  proven  with  the  MRIOmodel.The breakdown footprints of each land use type for all dimensions of the footprint arevery similar for the most part, except for some differences in the composition of EFI and EFE.The relatively small imported carbon footprint with the MRIO model may be attributed tothe allocation process in the model where direct household emissions were excluded asnon-tradable emission. During the allocation process, the ratio of industry and householdemission was proxied by that of the U.K. for many countries because of data deficiencies.This  may  have  resulted  in  the  underestimation  of  embodied  carbon  footprint.  For  othermajor differences such as the higher percentage of cropland in the MRIO EFI (34% withMRIO compared to 15% with NFA) and EFE (29% with MRIO compared to 21% with NFA)could  be  explained  by  the  high  sector  aggregation  employed  in  this  MRIO  model  whichdistorts the proportion of each land use type requirement. For example, even if a sectordemands a unit of input from only the forestry sector (and therefore requires only “forestland”), the model assigns the sector cropland, grazing land and forest land because all threeland use types are allocated to the “agriculture, hunting, forestry and fishing” sector. This isan unavoidable weakness of MRIO models that have low sector disaggregation levels.Being faced with such discrepancies, it is difficult to conclude which model lies closer tothe truth. Since embodied footprints cannot be measured directly, “one will always have torely on an indirect allocation through modeling approaches.”(Wiedmann, 2009b). Bothmodels have their deficiencies and are based on assumptions. However, in the NFAcalculations, the calculations of derived product footprints are estimated only through theirobvious physical relationship to raw biological materials (e.g. the amount of wheat used toproduce bread). This procedure is less applicable and becomes highly inaccurate when itcomes to more highly manufactured products like electrical machinery (Wiedmann, 2009b).The same is true for the footprints of other activities that are much higher up theproduction  system  (e.g.  timber  for  cardboard  packaging  for  a  computer  or  food  for  abusiness lunch at a bank which is selling financial service) (Wiedmann, 2009b). In his study 60 of comparing trade-embodied energy footprints from the NFA and MRIO models,Wiedmann (2009) concludes that for such complex problems, “multi-region input-output(MRIO) models – once fully developed – will be particularly appropriate to estimate theEcological Footprints embodied in trade flows with the possibility to track their origin viainter-industry linkages, international supply chains and multi-national trade flows”. Thus, itis likely that the MRIO model potentially reflects the reality more accurately than the NFAmethod.However, one can also argue that despite the structural differences of the twoapproaches, the results are fairly similar. The purpose of the ecological footprint, fromwhen  it  was  first  developed,  were  not  meant  to  produce  precise  figures  but  rather  toprovide a rough (under)estimate of the environmental impact of a specified population(Wackernagel & Rees, 1996). In this respect, the results from each approaches mutuallyconfirms the other and together confirm the fact that Canadians, on average, requiredabout 7 to 10 global hectares of bio-capacity per capita in 2005 to support their lifestyles.This is approximately four to five times the equitable share of global bio-capacity. 5.1.2 Summary on the Strengths and Weaknesses of I-O Based EFAThis section provides some reflection on the merits and demerits of using the I-O basedEFA. Table 12 gives a summary of the main points. TABLE 12: Strengths and Weaknesses of I-O Based EFA Strengths Weaknesses l Able to attribute environmental impacts tovarious consumption activities. l Suitable for calculating embodied resourceuse especially for activities higher up theproduction system. l Consistent with UN accounting frameworkand other economic database. l MIOT-based I-O analysis may not reflectphysical realities accurately. l Important  information  may  be  lost  duringsector aggregation process StrengthsOne of the strengths of the I-O based EFA is that it allows attributing ecologicalfootprints to almost any consumption activity, of regions, nations, governments, cities, etc. 61 (Wiedmann, Lenzen, Turner, Minx, & John Barrett, 2007). With the existing NFA method,assigning ecological footprints to sub-national entities required scaling techniques orbottom-up accounting methods (i.e. using life-cycle analysis (LCA) to calculate ecologicalfootprints for each products and activities.) With the I-O based EFA, as long as there is afinal demand vector for the respective consumption activity, it is fairly easy to calculate theecological footprint of that activity.As already mentioned, I-O based EFA is especially suitable for assigning ecologicalfootprints to products and services that are higher up the production system23  ortrade-embodied ecological footprints. I-O tables capture the complex inter-industriallinkages through money flows that are otherwise very difficult to trace using the existingmethod.Much of the data required to construct the I-O based EFA is consistent with theestablished  UN  accounting  standards  and  is  desirable  to  push  for  further  integrating  thetwo. This “will underpin and lend credibility to a Footprint accounting standard. Attentionshould also be paid to retaining commodity classification as disaggregated and relevant forthe Footprint as possible” (Wiedmann, 2009b). WeaknessesMost of the weaknesses of the I-O based EFA are either caused by basing the analysison the monetary I-O tables (MIOTs) or lack of detail in the current data.Environmentally extended use of the MIOTs assumes a constant ratio between themonetary transaction and physical transaction. In other words, I-O based EFA is totallydependent on the premise that resource use per dollar ratio is uniform and coherent acrossand within sectors24. This is a rather naïve assumption and compromises the results tosome degree.As mentioned in page 60 using the example of the “agriculture, hunting, forestry andfishing” sector, results may be distorted when sector aggregation levels are high. In general,the fewer sectors in the model, the more information is lost about the inter-industriallinkages.  For  example,  one  cannot  assess  whether  a  sector  required  input  from  theagricultural sector or the forestry sector if they are both aggregated as one. This will 23 e.g. example of the banking service on page 60.24 See appendix A (page 83), “linearity” for more explanation. 62 inevitably cause over- or underestimation of certain land type requirement.At least some of these weaknesses, however, can be overcome by improved dataavailability, careful assumption setting and further disaggregation research. 5.1.3 Policy ImplicationsIn this section, I step back from the methodological discussion, and briefly offer somebroader public policy implications of this study, especially respecting the trade aspect ofthe ecological footprint. Before that, however, I once again stress the general importance offraming the overall sustainability discussion within a larger interregional analyticframework. The essence of this argument is already briefly mentioned in the introductorychapter. Trade introduces a psychological disconnect between people’s actions and theirimpact on the environment. The interregional analytic framework, as presented byKissinger and Rees (2009) is “one based on a recognition that sustainability anywhere islinked, directly and indirectly, to sustainability elsewhere” (Kissinger & Rees, 2009a). Thisanalytic framework is becoming increasingly important in a globalizing world whereproduction activities and consumption activities are connected via complex trade network.In recognition of such emerging realities, the European Union has indicated SustainableConsumption  and  Production  (SCP)  as  one  of  the  key  objectives  in  their  EU  SustainableDevelopment Strategy (SDS) (Nash, 2009). The European Commission’s “Communicationon the Sustainable Consumption and Production and Sustainable Industrial Policy ActionPlan” explicitly states: “The challenges are directly linked to our way of life. The way we produce and consume contributes to global warming, pollution, material use, and natural resource depletion. The impacts of consumption in the EU are felt globally, as the EU is dependent on the imports of energy and natural resources. Furthermore, an increasing proportion of products consumed in Europe are produced in other parts of the world. The need to move towards more sustainable patterns of consumption and production is more pressing than ever.”(European Commission, 2008) However, interregional dependencies have far-reaching implications beyond just theenvironmental, but also for economic and political sustainability, as outlined below. 63 Environmental ImplicationsEnvironmental problems can be categorized into different types according to their scaleof impact and their source. Climate Change, as caused by GHG emissions into theatmosphere, has numerous sources and the impacts are global. On the other hand,problems like SO2 pollution and water contamination are relatively more localized andoften have an obvious source. In both cases, however, both the producers and theconsumers of the related activities are responsible for the cause (i.e. demand inducessupply and supply also induces demand). As already mentioned, environmental policieshave traditionally taken a producer-centric view of environmental impacts, perhaps due tothe market-driven economies’ tendency to avoid interfering with consumer’s preferences(Lenzen, Murray, Sack, & Wiedmann, 2007). However, facilitated by indicators like theecological footprint, awareness of consumer’s responsibility is increasing. TABLE 13: Research and Policy Questions that can be Answered Using Environmentally-Extended Input-Output AnalysisAdapted from Moll & Watson (2009)and Wiedmann et al.,( 2009). 64 In this context, sector-level trade embodied footprint analysis can inform policy makersabout Canada’s impact on other countries and how it can be corrected. Table 13 highlightsseveral policy questions that can be at least partially answered using the results of I-Obased studies like this one (but preferably with more sector detail). Unlike the existing NFAmethod which presents only imported footprints in aggregate, the I-O based method is ableto identify sectors that are “hot spots” where policies can most effectively reduce resourceuse (Wiedmann et al., 2009). Actual policy tools can take the form of taxes collected by theexporting country for certain goods and services that would be used to maintain the relatedproductive bio-capacity in the exporting region (e.g. reforestation fund for wood products).Labeling and certification methods, although more voluntary, are also common tools thatguide consumers to make better choices (e.g. seen in products like seafood, coffee beansand electronic devices). The word “consumer responsibility” has an obligatory connotation,but in fact it should be in the consumer’s greatest self-interest to support the preservationor improvement of the distant supportive bio-capacity.Another environmental implication of detailed studies that take into account distantenvironmental impacts is the potential empirical evidence against the concept ofenvironmental  Kuznets  curves  (EKC).  Although  it  is  already  highly  contested,  EKChypothesizes that a country’s relationship between per capita income and variousindicators of environmental degradation has an inverted U-shape (IBRD, 1992). EKC is atthe  center  of  the  “growth  decoupling”  argument25. The theory attributes decreasingdomestic environmental degradation to higher income because of increased efficiency andtechnological advancement. However, some studies suggest the exact opposite of the EKC –that environmental impact is positively correlated with affluence (Dietz, Rosa, & York,2007; Ehrlich & Holdren, 1971). Analyzed from an interregional framework, the apparentimprovement in domestic environmental quality observed in some studies that supportEKC  may  merely  be  an  effect  of  moving  footprint-intensive  sectors  to  other  countries.  Atime series of sector-level footprint studies may reveal that footprint-intensive sectors have(or have not) shifted their production to overseas. 25 Decoupling refers to a state in which economic growth is no longer correlated to environmentaldegradation. 65 Economic and Political ImplicationsAccording to the NFA, Canada had 15.4gha of bio-capacity per capita in 2005 – one ofthe  highest  figures  in  the  world.  This  is  a  result  of  Canada’s  vast  territory  and  relativelysmall population. Although Canada is one of the more fortunate countries in terms ofnatural endowments, about one fourth of the natural resources required to support thecountry’s consumption is met with imports. Also, Canadians export about 42% of theirproduction-related bio-capacity use to other countries. Countries with higher populationdensities than Canada depend more for survival on imported “natural income” fromexternal sources like Canada which their natural capital are already being rapidly depleted.According to Kissinger and Rees; “In just over a century, high volume productionagriculture on the Canadian prairies has all but eliminated the natural grassland habitatand the rich flora and fauna associated with it. In just over a century, productionagriculture has permanently dissipated almost half of the rich grassland soils that requiredmillennia to accumulate on the post-glacial plains.”(Kissinger & Rees, 2009b)In a “full-world”26 when natural capital is the new limiting factor of productivity,bio-capacity and productive ecosystems are directly linked to a country’s economicwellbeing and hence political interest. This relationship will only intensify, as we undergo“peak everything” of important parameters that underpin our current economies includingoil, minerals and fresh water (Heinberg, 2007). The increasing strategic importance ofnatural resources means that countries like Canada - which depends about 30% of itsincome  on  natural  resources  and  exports  about  42%  of  its  bio-capacity  use  to  othercountries - should wisely manage its ecological base to secure sustainable use.Finally, specialization in certain products or crops is more efficient in theory, but it isbuilt upon stable biophysical and geopolitical context, and other assumptions 27(Kissinger& Rees, 2010). At a time of uncertainty and change, investment in diverse ecosystems andsocietal structures would make them more resilient. 26 A term that refers to a planet which is ecologically at carrying capacity or “full of people and our stuff”(H. E. Daly, 1994)27 Capital and labor immobility, for example (H. Daly & Goodland, 1994). 66 5.2 Summary and ConclusionThis thesis analyzed the ecological footprint of Canada by developing a multi-regionalinput-output (MRIO) model as an alternative to the existing NFA method. The MRIO modelyielded  a  higher  ecological  footprint  for  Canada  than  the  NFA  method.  This  result  isconsistent with theory, which suggests that Canada’s eco-footprint was underestimated inthe NFA because of the use of world-average efficiencies for calculating EFI and EFE. Thus,this implies that the MRIO model is potentially a more accurate method for estimatingeco-footprints. However, since both models have their weaknesses and assumptions,neither is perfect, but each provides a rough estimate of the true ecological footprint.Therefore, I conclude that findings from this research confirm the results from the existingNFA method but suggest that actual ecological footprints may be larger. Once fullydeveloped with more accurate data, the MRIO models are potentially very useful andappropriate for ecological footprint calculations. The significance of this research is that itcontributed to the establishment of the model framework. The basic structure of the modelis easily transferable and flexible to enable future research when more detailed databecome available.In the NFA results, Canada is ranked 7th in the world for the size of its footprint (GlobalFootprint Network, 2010a). Its absolute footprint size and the distributions revealed in theresearch findings have important implications for environmental, economic and politicalpolicies that extend well beyond its territorial boundaries. There is a large gap between thescales of the current sustainability governance and our economic activities. In anincreasingly interconnected world, we must recognize and respond to the fact that“sustainability anywhere is linked, directly and indirectly, to sustainability elsewhere”(Kissinger & Rees, 2009a). Studies like this one can inform the development of new toolsand policies that reflect such inter-dependencies. 67 5.3 Future Research AgendasLastly, I provide several future research agendas and improvements that are relevant tothe I-O based ecological footprint estimations. Improvement of DataImprovement of data coverage and consistency would significantly increase therobustness of the model. In an increasingly globalizing world, availability of internationallycomparable coherent datasets is critical for policy research and analysis. Ideally, thereshould be an international framework among national statistical agencies for standardizingsector classification, reporting year, and other formats of key economic and environmentalstatistics. Such an initiative has been already started jointly by the United Nations, WorldBank, IMF, Eurostat and the OECD called the Integrated Environmental and EconomicAccounting (SEEA). It will establish an international statistical framework from whichimportant economic-environment indicators and climate change related policies will beformed28. Multidirectional trade MRIO modeling would also become easier when theseconsistent datasets become available in the future. Time Series AnalysisThis model estimated the ecological footprint of Canada only for a single year. Withincreased data consistency and modeling software capacity, a time series analysis wouldprovide a more dynamic picture of the economy-environment relationship as well as traderelationship trends. For example, one could observe whether a country is increasinglybecoming dependent or independent of imported bio-capacity from a particular othercountries. In other words, dynamic analysis elicits trends and a monitoring mechanism forassessing the degree to which countries are making progress towards their goal. Scenario AnalysisOne of the strength of the I-O based method is the potential for scenario analysis. Theoriginal monetary I-O analysis is commonly used in policy research to estimate economicimpacts of a planned government spending or a mega event like the Olympics. Theenvironmentally-extended I-O analysis can be potentially used in a similar fashion to 28 UN Statistics Division - UNCEEA (http://unstats.un.org/unsd/envaccounting/ceea/default.asp) 68 roughly estimate the marginal change in environmental impacts (in terms of increase ordecrease in footprint size) by plugging in the amount of money to be spent to its respectivesector. However, since the ecological footprint is calculated at a fixed point in time, it doesnot take into account the potential efficiency improvement in the future. Thus, the potentialfor scenario analysis is limited to a short- to mid-term timeframe. Despite this, I believe theI-O based method provides a good policy and planning research tool. 69 BibliographyAckerman, F., Ishikawa, M., & Suga, M. (2007). The carbon content of Japan–US trade.Energy Policy, 35(9), 4455-4462.Ahmad, N., & Wyckoff, A. (2003). Carbon Dioxide Emissions Embodied in InternationalTrade of Goods. OECD, Paris.Andrew, R., Peters, G., & Lennox, J. (2009a). Approximation and Regional Aggregation inMulti-Regional Input-Output Analysis for National Carbon Footprint Accounting.Economic Systems Research, 21(3), 311-335.Andrew, R., Peters, G., & Lennox, J. (2009b). Approximation and Regional Aggregation inMulti-Regional Input-Output Analysis for National Carbon Footprint Accounting.Economic Systems Research, 21(3), 311-335.Bagliani, M., Galli, A., Niccolucci, V., & Marchettini, N. (2008). Ecological footprint analysisapplied to a sub-national area: the case of the Province of Siena (Italy). Journal ofenvironmental management, 86(2), 354-64.Bicknell, K. (1998). New methodology for the ecological footprint with an application to theNew Zealand economy. Ecological Economics, 27(2), 149-160.Chapagain, A. K., & Hoekstra, A. Y. (2007). The water footprint of coffee and teaconsumption in the Netherlands. Ecological Economics, 64(1), 109-118.Daly, H. E. (1994). From Empty World Economics to Full World Economics: Recognizing anHistoric Turning Point in Economic Development. In R. Goodland, H. E. Daly, & S. E.Serafy (Eds.), Environmentally Sustainable Ecological Development: Building onBrundtland. Paris: UNESCO: Island Press.Daly, H. E. (1999). Uneconomic growth in theory and in fact (pp. 1-15). Trinity College,Dublin: The First Annual Faesta Lecture.Daly, H. E., & Farley, J. (2004). Ecological Economics: Principles and Applications (1st ed.).Washington (DC): Island Press.Daly, H., & Goodland, R. (1994). An ecological-economic assessment of deregulation ofinternational commerce under GATT. Ecological Economics. 70 Davis, C. (1990). Regional Economic Impact Analysis and project Evaluation. Vancouver:UBC Press.Davis, S. J., & Caldeira, K. (2010). Consumption-based accounting of CO2 emissions.Proceedings of the National Academy of Sciences of the United States of America,107(12), 5687-92.Davis, S. J., & Caldeira, K. (2010). Consumption-based accounting of CO2 emissions.Proceedings of the National Academy of Sciences of the United States of America,107(12), 5687-92.Diamond, J. (2005). Collapse: How Societies Choose to Fail or Succeed. New York: VikingPress.Diaz, R. J., & Rosenberg, R. (2008). Spreading dead zones and consequences for marineecosystems. Science (New York, N.Y.), 321(5891), 926-9.Dietz, T., Rosa, E. a, & York, R. (2007). Driving the human ecological footprint. Frontiers inEcology and the Environment, 5(1), 13-18.Ehrlich, P., & Holdren, J. (1971). Impact of Population Growth. Science, 171, 1212-17.European Commission. (2008). European Commission communication on the sustainableconsumption and production and sustainable industrial policy action plan.Ferng, J.-jiun. (2001). Using composition of land multiplier to estimate ecological footprintsassociated with production activity. Ecological Economics, 37, 159- 172.Fletcher, J. (1989). Input-output analysis and tourism impact studies. Annals of TourismResearch, 16(4), 514-529.Ghertner, D. A., & Fripp, M. (2007). Trading away damage: Quantifying environmentalleakage through consumption-based, life-cycle analysis. Ecological Economics, 63(2-3),563-577.Giljum, S., & Eisenmenger, N. (2004). North-South Trade and the Distribution ofEnvironmental Goods and Burdens: a Biophysical Perspective. The Journal ofEnvironment & Development, 13(1), 73-100. 71 Giljum, S., Lutz, C., & Jungnitz, A. (2008). SERI STUDIES The Global Resource AccountingModel ( GRAM ) A methodological concept paper (pp. 1-40).Global Footprint Network. (2008). Guidebook to the National Footprint Accounts 2008.Global Footprint Network. (2009). Ecological Footprint Atlas 2009.Global Footprint Network. (2010a). Ecological Footprint Atlas 2010.Global Footprint Network. (2010b). Calculation Methodology for the National FootprintAccounts , 2010 Edition.Heinberg, R. (2007). Peak Everything: Waking Up to the Century of Declines. New SocietyPublishers.Hertwich, E. G., & Peters, G. (2010). Multiregional Input-Output Database OPEN:EUTechnical Document. Framework (pp. 1-22).Hubacek, K., & Giljum, S. (2003). Applying physical input–output analysis to estimate landappropriation (ecological footprints) of international trade activities. EcologicalEconomics, 44(1), 137-151.IBRD. (1992). World Development Report 1992. Development and the environment. NewYork: Oxford University Press.IEA. (2010). Key World Energy Statistics 2010. Statistics. OECD Publishing.Kissinger, M., & Rees, W. E. (2009a). Assessing Sustainability in a Globalizing World. Journalof Industrial Ecology, 13(3), 357-360.Kissinger, M., & Rees, W. E. (2009b). Footprints on the prairies: Degradation andsustainability of Canadian agricultural land in a globalizing world. EcologicalEconomics, 68(8-9), 2309-2315.Kissinger, M., & Rees, W. E. (2010). Importing terrestrial biocapacity: The U.S. case andglobal implications. Land Use Policy, 27(2), 589-599. 72 Kitzes, J., Galli, A., Bagliani, M., Barrett, John, Dige, G., Ede, S., et al. (2009). A researchagenda for improving national Ecological Footprint accounts. Ecological Economics,68(7), 1991-2007.Lenzen, M., Murray, J., Sack, F., & Wiedmann, T. (2007). Shared producer and consumerresponsibility — Theory and practice. Ecological Economics, 61(1), 27-42.Lenzen, M., Pade, L.-L., & Munksgaard, Jesper. (2004). CO2 Multipliers in Multi-regionInput-Output Models. Economic Systems Research, 16(4), 391-412.Leontief, W. (1966). Input-Output Economics. New York: Oxford University Press.Mcdonald, G., & Patterson, M. (2004). Ecological Footprints and interdependencies of NewZealand regions. Ecological Economics, 50(1-2), 49-67.Millennium Ecosystem Assessment. (2005). Ecosystems and Human-wellbeing: Synthesis.Moll, S., & Watson, D. (2009). Environmental Pressures from European Consumption andProduction A study in integrated environmental and economic analysis. ETC/SCPWorking Paper, (February), 1-67.Munksgaard, J. (2001). CO2 accounts for open economies: producer or consumerresponsibility? Energy Policy, 29(4), 327-334.Muñoz, P., & Steininger, K. W. (2010). Austria?s CO2 responsibility and the carbon contentof its international trade. Ecological Economics, 69(10), 2003-2019.Nakano, S., Okamura, A., Sakurai, N., Suzuki, M., Tojo, Y., & Yamano, N. (2009). TheMeasurement of CO2 Embodiments in International Trade: Evidence from theHarmonised Input-Output and Bilateral Trade Database.Nash, H. (2009). The European Commission?s sustainable consumption and productionand sustainable industrial policy action plan. Journal of Cleaner Production, 17(4),496-498.Ng, Y. (1993). Relative income, aspiration, environmental quality, individual and politicalmyopia Why may the rat-race for material growth be welfare-reducing? MathematicalSocial Sciences, 26(1), 3-23. 73 OECD STI Division of Economic Analysis and Statistics. (2009). STAN Bilateral TradeDatabase Documentation.Pearce, D., Markandya, A., & Barbier, E. (1989). Blueprint for a Green Economy. London,UK: Earthscan Publications.Peters, G., & Hertwich, E. (2007). Opportunities and Challenges for Environmental MRIOModeling: Illustrations with the GTAP Database. 16th International Input-OutputConference (pp. 1-26). Retrieved January 14, 2011, fromhttp://www.iioa.org/pdf/16th Conf/Papers/Peters_LP.pdf.Rees, W. E. (1992). Ecological footprints and appropriated carrying capacity: what urbaneconomics leaves out. Environment and Urbanization, 4(2), 121-130.Rees, W. E. (1995). Achieving Sustainability: Reform or Transformation? Journal ofPlanning Literature, 9(4), 343-361.Rees, W. E. (2002). Globalization and Sustainability: Conflict or Convergence? Bulletin ofScience, Technology & Society, 22(4), 249-268.Rees, W. E. (2006). Ecological Footprints and Bio-Capacity: Essential Elements inSustainability Assessment. In D. Jo & V. L. Herman (Eds.), Renewables-BasedTechnology: Sustainability Assessment. Chichester, UK: John Wiley and Sons.Rees, W. E. (2010a). Cities as Dissipative Structures: Global Change and the Vulnerability ofUrban Civilization. Presented at the International Symposium on SustainabilityScience: The Emerging Paradigm and the Urban Environment (Montclair StateUniversity, 25-27 October 2010 [in press]).Rees, W. E. (2010b). Globalization and Extended Eco-footprints: Neo-colonialism and(Un)sustainability. In J. R. Engel, L. Westra, & K. Bosselmann (Eds.), Democracy,Ecological Integrity and International Law. Newcastle: Cambridge Scholars Publishing.Rehdanz, K., & Maddison, D. (2005). Climate and happiness. Ecological Economics, 52(1),111-125.Richardson, H. (1972). Input-Output and Regional Economics. London: Weidenfeld andNicolson. 74 Simon, J. (1995). The State of Humanity: Steadily Improving. Washington (DC).Stiglitz, J., & Sen, A. (2009). Report by the Commission on the Measurement of EconomicPerformance and Social Progress. Sustainable Development.Suh, S. (Ed.). (2009). Handbook of Input-Output Economics in Industrial Ecology. Media.New York: Springer.Suh, S., & Kagawa, S. (2005). Industrial ecology and input-output economics: anintroduction. Economic Systems Research, 17(4), 349-364.Turner, K., Lenzen, M., & Wiedmann, T. (2007). Examining the global environmental impactof regional consumption activities--Part 1: A technical note on combining input-outputand ecological footprint analysis. Ecological Economics, 1-8.UN Department of Economic and Social Affairs. (2010). Trends in Resource Use. World (pp.1-2).UNEP. (2011). Disasters and Conflict. Retrieved from http://www.unep.org/.UNFPA. (2011). Population Trends. Retrieved November 12, 2011, fromhttps://www.unfpa.org/public/.US Census Bureau. (2011). World Population Growth Rates. Retrieved November 12, 2011,from http://www.census.gov/.Wackernagel, M., & Rees, W. E. (1996). Our Ecological Footprint: Reducing Human Impacton the Earth. New Society Publishers.Wackernagel, M., Schulz, N. B., Deumling, D., Linares, A. C., Jenkins, M., Kapos, V., et al.(2002). Tracking the ecological overshoot of the human economy. Proceedings of theNational Academy of Sciences of the United States of America, 99(14), 9266-71. WCED. (1987). Our Common Future. Oxford, UK: Oxford University Press.Wiedmann, T. (2009a). A review of recent multi-region input – output models used forconsumption-based emission and resource accounting. Ecological Economics, 69(2),211-222. 75 Wiedmann, T. (2009b). A first empirical comparison of energy Footprints embodied intrade — MRIO versus PLUM. Ecological Economics, 68(7), 1975-1990.Wiedmann, T. (2009c). A review of recent multi-region input–output models used forconsumption-based emission and resource accounting. Ecological Economics, 69(2),211-222.Wiedmann, T., Lenzen, M., Turner, K., & Barrett, John. (2006). Examining the globalenvironmental impact of regional consumption activities — Part 2 : Review of input –output models for the assessment of environmental impacts embodied in trade.Ecological Economics, 62(1), 37-44.Wiedmann, T., Lenzen, M., Turner, K., Minx, J., & Barrett, John. (2007). MultiregionalInput-Output Modelling Opens New Opportunities for the Estimation of EcologicalFootprints Embedded in International Trade. Paper prepared for the InternationalEcological Footprint Conference, Cardiff, 8-10 May.Wiedmann, T., Minx, J., Barrett, J, & Wackernagel, M. (2006). Allocating ecological footprintsto final consumption categories with input–output analysis. Ecological Economics,56(1), 28-48.Wiedmann, T., Wilting, H., Lutter, S., Palm, V., Giljum, S., Wadeskog, A., et al. (2009).Development of a methodology for the assessment of global environmental impacts oftraded goods and services. Retrieved July 31, 2010, fromhttp://www.old.seri.at/documentupload/SERI PR/eipot_final_report.pdf.Wixted, B., Yamano, N., & Webb, C. (2006). Input-Output Analysis in an increasinglyglobalized world: applications of OECD?s harmonized international tables.STI/Working Paper.WWF. (2010). Living Planet Report 2010.Yamano, N., & Ahmad, N. (2006). The OECD Input-Output Database: 2006 Edition.STI/Working Paper. OECD, Paris. 76 Appendices 77 Appendix A: Major Assumptions and Limitations of the I-O AnalysisThere are several major assumptions inherent in the I-O analysis that one must beaware of when applying the analysis. These are: (1) fixed input coefficient; (2) linearity; (3)omission of unpaid work and (4) homogeneous sectors. A.1 Fixed Input CoefficientThe most commonly cited assumption of the I-O model is the fixed input coefficients.The proportions in which each sector purchases its inputs from other sectors are assumedto be invariant over the period of the analysis (C. Davis, 1990). Davis notes three reasonswhy this assumption may not hold in reality. The first is the technological change.Technological advance or increase in efficiency will likely alter both purchase patterns andquantity. If an industry is able to produce the same amount of output with less input usingdifferent material, this will significantly change the input coefficients.  The second point isthe relative price changes. When one commodity becomes expensive relative to others,there is a tendency for the economy to shift to substitutes. This will also likely alter thepurchase patterns and hence input coefficients. Finally, the input coefficients of the modelmay be changed by the location of a new firm. Entry or exit of a firm of significant size mayalter the weighted average input pattern of the relevant sector. A.2 LinearityLinearity is another major assumption of the I-O model. There are linearityassumptions manifested in two different levels: between input and output and betweenoutput and the ecological footprints.Firstly, all inputs into a particular sector are assumed to be proportional to the output of 78 that sector. This assumption rules out the possibility of internal economies of scale, wherean x% increase in output may require less than x% of inputs from other sectors (Bicknell,1998).Secondly, the ecological footprint/output ratio used to convert monetary units toecological footprints assumes proportional relationship between output and the landrequirement. Such an assumption can be particularly inappropriate when inter-sectoralprices differ greatly. For example, sector C provides sector A with $5 million worth of goodsand sector B with $10 million worth of goods. If inter-sectoral prices are constant, we couldinfer from this that sector C provides sector B with 2 times more physical output than theyprovide  for  Sector  A.  However,  for  some  reason  if  sector  C  was  able  to  charge  Sector  B  asubstantially higher price per unit than sector A, it would distort the physical linkagesbetween sectors (Bicknell, 1998). A.3 Omission of Unpaid WorkSince MIOTs only trace monetary transactions between sectors, unpaid work isexcluded from the transaction table (Bicknell, 1998). For countries where a substantialamount  of  activity  takes  place  outside  of  the  monetary  economy,  the  MIOT-based  I-Oanalysis may underestimate the resource use of the country. A.4 Homogenous SectorsThe  last  major  assumption  is  of  homogenous  sectors.  Different  MIOTs  have  differentclassifications and number of sectors. For example the North American IndustryClassification System (NAICS) is the standardized system for Canada, United States andMexico. Although sectors are grouped in activities of similar input patterns as possible, 79 there will be heterogeneous activities within sectors. This is truer when sectors becomemore aggregated. For example, the marine construction sector includes the construction ofboth small pleasure crafts and ocean-going ships. An $x increase in demand for a pleasurecraft will not have the same impact as an $x increase in demand for shipbuilding (C. Davis,1990). 80 Appendix B: Example of Calculating Ecological Footprint Using I-O AnalysisBelow is an example illustration of the I-O based calculation of ecological footprint, usinga hypothetical simple 3 sector economy. In the actual model, all the calculations are doneusing Excel spreadsheets. Table B.1 is a hypothetical I-O table of a country named A with only 3 sectors: food sector,manufacturing sector and the services sector. TABLE 14: Hypothetical 3 sector I-O Table of Country A (Unit: Million $) Year X Food Manufac. Services Final Demand Net Exports *1 Total Output Food 10 5 5 25 20 65 Manufac. 20 30 25 15 -5 85 Services 5 10 10 50 0 75 Value Added 30 40 35 Total Input 65 85 75 *1 Net Exports (NX) = Exports - Imports Step 1: Allocate EFP to industrial sectors in the I-O tableFor example, presume that the EFP data of country A is the follows: TABLE 15: Ecological Footprint of Production (EFP) Data of Country A Land Use Type EFP (Unit: global hectares) Cropland 90,000 Grazing Land 7,000 Forest Land 140,000 Fishing Grounds 8,000 Carbon Footprint 155,000 Built-up Land 1,000 TOTAL 410,000 There are several assumptions and data required when allocating these footprints to itsrespective related industrial sector in the I-O table: 81 (1) Cropland, Grazing Land, Forest Land and Fishing Grounds are all allocated to the FoodSector (in the actual model, it is allocated to the “Agriculture, Hunting, Forestry andFishing” sector).(2)  The carbon footprint is proportionately allocated to each sector of the economyweighted based on the CO2 emissions data from each sector (see Table B.3 for example.In the actual model, this data is provided by the IEA CO2 emissions data). TABLE 16: CO2 Emissions by Industrial Sector and their Share Sector CO2 Emissions (million tonnes) Share (%) Food Sector 100 19% Manufacturing Sector 240 44% Services Sector 200 37% TOTAL 540 100% (3) Built-up Land is proportionately allocated to each sector of the economy weighed basedon the output value. This is rather an inaccurate assumption, because it assumes that thesectors with the highest output also have the highest associated built-up land. However,there is currently no land use data available for each sector. This is relatively a minor issue,however, considering the minor importance of built-up land in the overall footprint. Using the above assumptions, allocation of EFP to each sector is shown in Table B.4: TABLE 17: Allocation of EFP to its Respective Sector Food Sector Manufacturing Sector Services Sector Cropland 90,000 - - Grazing Land 7,000 - - Forest Land 140,000 - - Fishing Grounds 8,000 - - Carbon Footprint 29,450 68,200 57,350 Built-up Land 200 400 400 TOTAL 274,650 68,600 57,750 82 Step 2: Calculate the Leontief Inverse Matrix ( (I-A)-1 )From Table B.1, the input coefficient (A) is extracted: ? = ??? ?? ?? ?? ?? ???? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?                            (1) Since I is the identity matrix (a matrix with 1 on the diagonal), (I-A) is: ? ? ? = ?? ? ?? ? ? ? ? ? ? ? ? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ? = ? ?? ?? ??? ?? ???????? ?? ?? ?? ????? ??? ?? ??? ?? ?? ?? ?      (2) Therefore the Leontief Inverse matrix is derived by inverting equation (2): ???????????????? = (? ? ?)?? = ??? ?? ?? ?? ?? ???? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?            (3) Step 3: Calculate the direct intensity matrix (EFdir)By dividing the total footprint of each sector (from Table B.4) by the total monetaryoutput data (from Table B.1), the direct footprint intensity matrix is calculated as follows: TABLE 18: Direct Footprint Intensity Matrix Calculation (Unit: gha/million $) Food Sector Manufacturing Sector Services Sector Cropland 90,000/65 = 1385 - - Grazing Land 7,000/65 = 108 - - Forest Land 140,000/65 = 2154 - - Fishing Grounds 8,000/65 = 123 - - Carbon Footprint 29,450/65 = 453 68,200/85 = 802 57,350/75 = 765 Built-up Land 200/65 = 3 400/85 = 5 400/75 = 5*Numbers are rounded up. 83 Thus the direct footprint intensity matrix is: ?????? = ? ?? ? ???? ??? ???? ??? ??? ? ? ? ? ? ??? ? ? ? ? ? ??? ? ? ?? ?                              (4) Each element of the matrix shows how much direct footprint is associated with a unit ofoutput of each sector. Step 4: Calculate the total intensity matrix (EFtot)The total footprint intensity matrix is derived by multiplying the direct footprintintensity matrix and the Leontief Inverse matrix. ????? ? ????? ? ???????????????? = ? ?? ? ???? ??? ???? ??? ??? ? ? ? ? ? ??? ? ? ? ? ? ??? ? ? ?? ? × ??? ?? ?? ?? ?? ???? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ? = ? ?? ? ???? ??? ???? ??? ???? ? ??? ??? ??? ?? ???? ?? ??? ?? ??? ?? ???? ?? ? ?? ?           (5) Each element of the matrix shows how much direct and indirect footprint is associated witha unit of output of each sector. 84 Step 5: Multiply final consumption vectors to the total intensity matrixTo calculate the footprint associated with domestic final demand, the total footprintintensity matrix is multiplied by the final demand vector (third column from the right inTable B.1). In order to conduct this calculation for each land use type, the EFtot matrix needsto be disaggregated into a diagonal matrix by each land use type29. ?????????????????? ? ????????? ? ?????? ????? = ????? ? ?? ??? ? ? ? ??? ?  × ????? ?? ? = ?????????? ????? ?               (6) ?????????????????????? ? ????????? ??? ? ?????? ????? = ? ??? ? ? ? ??? ? ? ? ?? ? × ????? ?? ? = ???????? ??? ?              (7) ????????????????????? ? ???????? ??? ? ?????? ????? = ????? ? ?? ???? ? ? ? ??? ?× ????? ?? ? = ?????????? ????? ?            (8) ???????? ???????????????? ? ????????? ??? ? ?????? ????? = ???? ? ?? ?? ? ? ? ?? ? × ????? ?? ? = ???????? ??? ?             (9) 29 This is s basic rule of linear algebra. 85 ???????????????? ? ???????? ??? ? ?????? ????? = ????? ? ?? ???? ? ? ? ???? ?× ????? ?? ? = ??????????? ????? ?              (10) ????????????????????? ? ??????? ??? ? ?????? ????? = ?? ? ?? ?? ? ? ? ?? ? × ????? ?? ? = ??????? ??? ?                    (11) Thus, the footprint associated with domestic final demand of court y A is: TABLE 19: Ecological Footprint of Domestic Consumption of Country A (Unit: gha) Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Built-up Land TOTAL Food 42,935 3,348 66,774 3,813 32,094 207 149171 Manufacturing 2,909 227 4,523 258 24,632 155 32704 Services 10,388 810 16,155 923 80,064 515 181875 TOTAL EF 56,232 4,385 87,452 4,994 136,790 877 234,497 EF/capita* 1.87 0.15 2.91 0.17 4.56 0.03 7.82*Assuming that country A has a population of 30,000 people 86 Appendix C: Sensitivity Analysis for RoW CategoryBelow are three different scenarios of the rest of the world (RoW) category usingdifferent proxy countries. These countries are: China, Indonesia and the U.S (thesecountries appeared in the literature as proxy countries). The years in the parenthesis arethe base years of the I-O table. TABLE 20: Scenario 1- Proxy = China (2005) Sectors Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Total 1 Agriculture, Hunting, Forestry and Fishing 2,098,449 494,839 480,552 375,196 249,954 3,698,990 2 Mining and Quarrying 475,838 112,208 108,968 85,078 3,094,298 3,876,391 3 Food products, Beverages and Tobacco 929,615 219,214 212,885 166,212 393,665 1,921,591 4 Textiles, Textile Products, Leather and Footwear 734,238 173,142 168,143 131,279 742,951 1,949,753 5 Wood and Products of Wood and Cork 63,345 14,938 14,506 11,326 178,786 282,901 6 Pulp, Paper, Paper Products, Printing and Publishing 29,538 6,966 6,764 5,281 113,402 161,952 7 Coke, Refined Petroleum Products and Nuclear Fuel 71,546 16,871 16,384 12,792 703,696 821,290 8 Chemicals excluding Pharmaceuticals 104,368 24,611 23,901 18,661 230,945 402,487 9 Pharmaceuticals - - - - - - 10 Rubber and Plastics Products 27,700 6,532 6,343 4,953 195,174 240,702 11 Other Non-Metallic Mineral Products - - - - - - 12 Iron and Steel 46,022 10,852 10,539 8,229 318,887 394,528 13 Non-Ferrous Metals - - - - - - 14 Fabricated Metal Products 40,075 9,450 9,177 7,165 390,334 456,202 15 Machinery and Equipment 65,782 15,512 15,064 11,762 429,397 537,516 16 Office, Accounting and Computing Machinery 153,829 36,275 35,227 27,504 1,031,088 1,283,922 17 Electrical Machinery and Apparatus, n.e.c 49,690 11,717 11,379 8,884 335,153 416,823 18 Radio, Television and Communication Equipment - - - - - - 19 Medical, Precision and Optical Instruments 31,213 7,360 7,148 5,581 476,053 527,355 20 Motor Vehicles, Trailers and Semi-Trailers 34,477 8,130 7,895 6,164 262,189 318,856 21 Building and Repairing of Ships and Boats - - - - - - 22 Aircraft and Spacecraft - - - - - - 23 Railroad and Transport Equipment, n.e.c - - - - - - 24 Manufacturing n.e.c; Recycling 149,653 35,290 34,271 26,757 581,821 827,792 25 Electricity and Gas - - - - - - 26 Services 1,253,283 295,539 287,006 224,083 2,506,911 4,566,821 Total EF 6,358,660 1,499,446 1,456,154 1,136,909 12,234,705 22,685,874 Per capita EF 0.20 0.05 0.04 0.04 0.38 0.70 87 TABLE 21: Scenario 2 - Proxy = Indonesia (2005) TABLE 22: Scenario 3- Proxy= U.S.A (2005) Sectors Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Total 1 Agriculture, Hunting, Forestry and Fishing 3,047,099 77,449 1,499,659 1,407,697 112,010 6,143,913 2 Mining and Quarrying 73,023 1,856 35,939 33,735 743,743 888,295 3 Food products, Beverages and Tobacco 1,189,251 30,227 585,301 549,409 143,309 2,497,499 4 Textiles, Textile Products, Leather and Footwear 341,770 8,687 168,205 157,890 451,104 1,127,656 5 Wood and Products of Wood and Cork 90,608 2,303 44,594 41,859 93,467 272,831 6 Pulp, Paper, Paper Products, Printing and Publishing 17,490 445 8,608 8,080 76,314 110,936 7 Coke, Refined Petroleum Products and Nuclear Fuel 6,567 167 3,232 3,034 173,405 186,405 8 Chemicals excluding Pharmaceuticals 73,326 1,864 36,088 33,875 202,476 347,628 9 Pharmaceuticals 25,888 658 12,741 11,960 136,001 187,248 10 Rubber and Plastics Products 169,996 4,321 83,665 78,535 125,420 461,937 11 Other Non-Metallic Mineral Products 13,530 344 6,659 6,250 73,404 100,188 12 Iron and Steel 26,654 677 13,118 12,313 709,475 762,237 13 Non-Ferrous Metals 51,176 1,301 25,187 23,642 1,780,610 1,881,915 14 Fabricated Metal Products 32,694 831 16,091 15,104 344,984 409,704 15 Machinery and Equipment 49,319 1,254 24,273 22,785 602,489 700,120 16 Office, Accounting and Computing Machinery - - - - - - 17 Electrical Machinery and Apparatus, n.e.c 50,045 1,272 24,630 23,120 382,188 481,256 18 Radio, Television and Communication Equipment 234,242 5,954 115,284 108,215 1,686,505 2,150,200 19 Medical, Precision and Optical Instruments 33,823 860 16,646 15,625 2,118,245 2,185,199 20 Motor Vehicles, Trailers and Semi-Trailers 22,389 569 11,019 10,343 356,091 400,412 21 Building and Repairing of Ships and Boats 6,247 159 3,074 2,886 425,412 437,777 22 Aircraft and Spacecraft 55,358 1,407 27,245 25,574 6,380,329 6,489,913 23 Railroad and Transport Equipment, n.e.c 6,255 159 3,078 2,889 81,457 93,838 24 Manufacturing n.e.c; Recycling 190,172 4,834 93,595 87,856 564,499 940,956 25 Electricity and Gas - - - - - - 26 Services 1,422,272 36,150 699,985 657,060 1,713,274 4,528,742 Total EF 7,229,193 183,746 3,557,916 3,339,738 19,476,212 33,786,805 Per capita EF 0.22 0.01 0.11 0.10 0.60 1.04 Sectors Cropland Grazing Land Forest Land Fishing Grounds Carbon Footprint Total 1 Agriculture, Hunting, Forestry and Fishing 2,588,790 148,183 1,653,979 168,872 64,885 4,624,710 2 Mining and Quarrying 38,680 2,214 24,712 2,523 300,686 368,815 3 Food products, Beverages and Tobacco 678,363 38,830 433,406 44,251 50,575 1,245,425 4 Textiles, Textile Products, Leather and Footwear 158,563 9,076 101,306 10,343 251,607 530,895 5 Wood and Products of Wood and Cork 75,504 4,322 48,240 4,925 24,908 157,899 6 Pulp, Paper, Paper Products, Printing and Publishing 6,859 393 4,382 447 5,626 17,707 7 Coke, Refined Petroleum Products and Nuclear Fuel 8,146 466 5,205 531 65,681 80,030 8 Chemicals excluding Pharmaceuticals 11,798 675 7,538 770 21,477 42,258 9 Pharmaceuticals - - - - - - 10 Rubber and Plastics Products 11,822 677 7,553 771 28,005 48,828 11 Other Non-Metallic Mineral Products 1,180 68 754 77 15,035 17,114 12 Iron and Steel 6,208 355 3,966 405 60,987 71,921 13 Non-Ferrous Metals - - - - - - 14 Fabricated Metal Products 4,286 245 2,738 280 37,417 44,966 15 Machinery and Equipment 8,028 460 5,129 524 59,527 73,668 16 Office, Accounting and Computing Machinery 18,286 1,047 11,683 1,193 91,603 123,812 17 Electrical Machinery and Apparatus, n.e.c 5,689 326 3,635 371 73,399 83,420 18 Radio, Television and Communication Equipment - - - - - - 19 Medical, Precision and Optical Instruments - - - - - - 20 Motor Vehicles, Trailers and Semi-Trailers 5,890 337 3,763 384 29,337 39,711 21 Building and Repairing of Ships and Boats 687 39 439 45 6,868 8,078 22 Aircraft and Spacecraft - - - - - - 23 Railroad and Transport Equipment, n.e.c - - - - - - 24 Manufacturing n.e.c; Recycling 20,999 1,202 13,416 1,370 46,664 83,651 25 Electricity and Gas - - - - - - 26 Services 151,991 8,700 97,107 9,915 95,317 363,029 Total EF 3,801,769 217,614 2,428,952 247,997 1,329,604 8,025,936 Per capita EF 0.12 0.01 0.08 0.01 0.04 0.25 88 As evident from the sensitivity analysis, the imported footprints from the RoW differsignificantly  depending  on  which  country  is  chosen  as  the  proxy.  Figure  C.1  shows  thepercentage of RoW category by different scenarios in relation to the total importedfootprint. In this model, U.S. was chosen as the proxy as it produced the most conservativeresults. FIGURE 16: Percentage of RoW to Total EFI 89 Appendix D: Exchange Rate Table TABLE 23: US dollar per Local Currency by Year (1997-2005) Country 1997 1998 1999 2000 2001 2002 2003 2004 2005 1 Argentina 0.9997 0.9997 0.9997 0.9996 0.9994 3.240 2.945 2.951 2.926 2 Australia 0.742 0.628 0.645 0.580 0.517 0.543 0.649 0.735 0.764 3 Austria 1.128 1.112 1.065 0.921 0.895 0.941 1.129 1.242 1.244 4 Belgium 1.128 1.111 1.065 0.921 0.895 0.941 1.129 1.242 1.244 5 Brazil 0.928 0.862 0.551 0.547 0.426 0.342 0.325 0.342 0.411 6 Canada 0.772 0.674 0.673 0.673 0.646 0.637 0.714 0.769 0.825 7 China 0.121 0.121 0.121 0.121 0.121 0.121 0.121 0.121 0.122 8 Denmark 1.128 1.112 1.065 0.921 0.895 0.941 1.129 1.242 1.244 9 Finland 1.145 1.113 1.065 0.921 0.895 0.941 1.129 1.242 1.244 10 France 1.124 1.112 1.065 0.921 0.895 0.941 1.129 1.242 1.244 11 Germany 1.128 1.111 1.065 0.921 0.895 0.941 1.129 1.242 1.244 12 Greece 1.248 1.153 1.115 0.933 0.895 0.941 1.129 1.242 1.244 13 India 0.028 0.024 0.023 0.022 0.021 0.021 0.021 0.022 0.023 14 Indonesia 0.0003 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 15 Ireland 1.194 1.121 1.065 0.921 0.895 0.941 1.129 1.242 1.244 16 Israel 0.290 0.263 0.242 0.245 0.238 0.211 0.220 0.223 0.223 17 Italy 1.137 1.115 1.065 0.921 0.895 0.941 1.129 1.242 1.244 18 Japan 0.0083 0.0076 0.0088 0.0093 0.0082 0.0080 0.0086 0.0092 0.0091 19 Korea 0.0011 0.0007 0.0008 0.0009 0.0008 0.0008 0.0008 0.0009 0.001 20 Mexico 0.126 0.109 0.105 0.106 0.107 0.104 0.093 0.089 0.092 21 Netherlands 1.129 1.111 1.065 0.921 0.895 0.941 1.129 1.242 1.244 22 New Zealand 0.661 0.535 0.529 0.454 0.420 0.462 0.581 0.663 0.704 23 Norway 0.141 0.133 0.128 0.114 0.111 0.125 0.141 0.148 0.155 24 Poland 0.305 0.288 0.252 0.230 0.244 0.245 0.257 0.273 0.309 25 Portugal 1.144 1.113 1.065 0.921 0.895 0.941 1.129 1.242 1.244 26 Russia 0.173 0.103 0.041 0.036 0.034 0.032 0.033 0.035 0.035 27 South Africa 0.217 0.181 0.164 0.144 0.116 0.095 0.132 0.155 0.157 28 Spain 1.128 1.112 1.065 0.921 0.895 0.941 1.129 1.242 1.244 29 Sweden 1.128 1.112 1.065 0.921 0.895 0.941 1.129 1.242 1.244 30 Switzerland 0.689 0.690 0.666 0.592 0.593 0.642 0.743 0.804 0.803 31 Turkey 6.585 3.835 2.388 1.599 0.816 0.663 0.666 0.701 0.744 32 UK 1.637 1.656 1.618 1.513 1.440 1.499 1.633 1.831 1.818 33 USA 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000Source: OECD.Stat

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0071891/manifest

Comment

Related Items