BUILDING EORA: A GLOBAL MULTI-REGION INPUT–OUTPUT DATABASE AT HIGH COUNTRY AND SECTOR RESOLUTIONстатья из журнала
Аннотация: Abstract There are a number of initiatives aimed at compiling large-scale global multi-region input–output (MRIO) tables complemented with non-monetary information such as on resource flows and environmental burdens. Depending on purpose or application, MRIO construction and usage has been hampered by a lack of geographical and sectoral detail; at the time of writing, the most advanced initiatives opt for a breakdown into at most 129 regions and 120 sectors. Not all existing global MRIO frameworks feature continuous time series, margins and tax sheets, and information on reliability and uncertainty. Despite these potential limitations, constructing a large MRIO requires significant manual labour and many years of time. This paper describes the results from a project aimed at creating an MRIO account that represents all countries at a detailed sectoral level, allows continuous updating, provides information on data reliability, contains table sheets expressed in basic prices as well as all margins and taxes, and contains a historical time series. We achieve these goals through a high level of procedural standardisation, automation, and data organisation. Keywords: Multi-region input–outputConstrained optimisationData conflictAutomationVisualisation Acknowledgements The Eora project was funded by the Australian Research Council (ARC) under its Discovery Projects DP0985522 and DP130101293. The Réunion Project was funded by the University of Sydney through its International Program Development Fund. The work leading to this Special Issue was funded by IDE-JETRO. Throughout the project, a number of researchers made contributions to various technical aspects of Eora. The basic ideas for Eora's assembly and optimisation procedures were conceived in part by Blanca Gallego (Gallego and Lenzen, Citation2009) and further developed by Ting Yu at the University of Sydney (Yu et al., Citation2009). Julien Ugon from the University of Ballarat and Ting Yu developed the basis of a Quadratic Programming optimisation algorithm, based on earlier ideas by Yalcın Kaya, Regina Burachik, and Jerzy Filar from the University of South Australia (publications forthcoming). The supercomputer facility at the Australian National University through the NCI National Facility granted supercomputer runtime for carrying out some of the optimisation calculations. The authors thank Margaret Kahn from NCI as well as Yasushi Kondo from Waseda University for valuable advice. Sebastian Juraszek from the School of Physics at the University of Sydney expertly grew our high performance computing cluster as Eora got bigger and bigger. Richard Wood and Jessica Dielmann contributed to the data management process at earlier stages of the project. Richard Wood contributed a time series of Australian SUT (Wood, Citation2011), Tommy Wiedmann contributed detailed I–O data for the UK, and Mark Müller made available I–O tables for Central Asian countries (Müller, Citation2006; Müller and Djanibekov, 2009). Mathis Wackernagel at Global Footprint Network kindly shared the National Footprint and Biocapacity Accounts (Global Footprint Network, Citation2010) allowing us to calculate ecological footprints embodied in international trade. Helmut Haberl provided data on human appropriation of net primary productivity (HANPP), another important indicator of the ecological impact of consumption. Pablo Muñoz and Chia-Hao Liu processed data for South America and Taiwan, respectively. Robbie Andrew, Barney Foran, and Tommy Wiedmann gave valuable feedback on construction tools and user interface. Leonardo Souza from the United Nations Statistical Division provided advice on the interpretation of UN National Accounts databases. Charlotte Jarabak from the University of Sydney's Science and Technology Library supplied many CD-ROM-based data compendia. Patrick Jomini extracted the structure for the Hong Kong economy from the Salter database. The statistical agencies of numerous countries as well as international statistical organisations, such as the United Nations Statistical Divisions and Eurostat, assisted this project by supplying data. Finally, we thank three anonymous referees and the Guest Editor Arnold Tukker for their time in providing comments that helped improving this manuscript. Notes 1In the following, we will refer to a MRIO database extended with physical information simply as an MRIO. 2GTAP 8: 57 sectors and 129 regions, extended into a full MRIO by Peters et al. Citation(2011a); EXIOPOL: EU27 and 16 non-EU countries, and about 130 sectors; WIOD: 27 EU countries and 13 other major countries in the world; more than 30 industries and at least 60 products. 3 http://www.globalcarbonfootprint.com/queries/classifications.jsp. 4Instead of four specifiers as in Stelder and Oosterhaven Citation(2009). 5Gross national income (GNI) is GDP less primary incomes (net taxes on production and imports, and compensation of employees and property income) payable to the rest of the world (non-resident units) plus the corresponding items receivable from the rest of the world. 6The UN SNA Main Aggregates and Official Country databases list 252 geographical entities. Amongst the 65 entities excluded in our MRIO are small nations (Vatican, Monaco, Niue, Tokelau, and Nauru), disputed territories (Western Sahara), and small dependencies (Mayotte, American Samoa, Guam, and Gibraltar). For example, one may want to undertake a life-cycle or footprint analysis for a multi-national company, using Leontief's quantity I–O model for a classical demand-pull exercise. It is likely that the expenditure vector of that company exists only in terms of purchasers' prices. Having all margins matrices at hand, such an expenditure vector can readily be converted into basic prices without requiring further assumptions and data. A mathematical formalisation as well as intuitive introduction of these optimization methods is given in Section S2 of Lenzen et al. Citation(2012a). On conflict between balancing rules and raw data, see McDougall Citation(2006) and Wiebe et al. Citation(2012). On conflict within the UN Comtrade database, see Lenzen et al. Citation(2012a) and Bouwmeester and Oosterhaven Citation(2008). I–O and SUT can be compiled according to a range of technology assumptions, with the commodity and industry technology assumptions being the most widespread amongst data sets provided by Statistical Offices around the world (Ten Raa and Rueda-Cantuche, Citation2007). Each assumption has its drawbacks, and there is no definite overall advantage of one over the other assumptions (Kop Jansen and Ten Raa, Citation1990). Supply-use frameworks were suggested previously for use in Life-Cycle Assessment (Heijungs and Suh, Citation2002; Suh et al., Citation2010). Aus, Aut, Bel, Bra, Can, Che, Chl, Cze, Deu, Dnk, Esp, Est, Fin, Fra, Gbr, Grc, Grl, Hun, Irl, Ita, Ltu, Lva, Mkd, Mys, Nld, Nor, Nzl, Per, Pol, Prt, Rou, Svk, Svn, Swe, Tur, Twn, Ven, and Zaf. 10,160 sectors excluding supply-use industries, 15,909 sectors including both supply-use industries and products. The Eora website features colour-coded maps in order to distinguish positive and negative entries by magnitude.
Год издания: 2013
Авторы: Manfred Lenzen, Daniel Moran, Keiichiro Kanemoto, Arne Geschke
Издательство: Taylor & Francis
Источник: Economic Systems Research
Ключевые слова: Environmental Impact and Sustainability, Climate Change Policy and Economics, Global Energy and Sustainability Research
Открытый доступ: closed
Том: 25
Выпуск: 1
Страницы: 20–49