Creating A Flexible Large-Scale MRIO Database In A Collaborative Virtual Laboratory

Abstract

The development of a large-scale Multi-Region Input–Output (MRIO) database is often hampered by a number of issues: a) it is labour- and cost-intensive to produce updated database, b) a fixed geographical and sectoral coverage limits comprehensive and detailed analytical assessment, and c) a lack of reliability and uncertainty information as a result of imposed assumptions during MRIO construction restrains the transparency of the published data. Therefore, the main aim of this thesis is to address the listed issues by creating a flexible adaptation of a well-known World Input–Output Database (WIOD) in a virtual laboratory environment, called the ‘WIOD-Lab’. To achieve the core objective, I first present the creation of a cloud-computing platform—the Global MRIO Laboratory—offering advancements to many researchers in a collaborative research environment in development of WIOD-Lab. Second, I detail the MRIO construction of the WIOD database in the virtual laboratory and provide explanations on how WIOD-Lab advances the existing database. To this end, I also draw comparisons between the WIOD-Lab and the original WIOD database to test their closeness and their adherence to primary data sources. Third, I devise a new methodology to support the MRIO comparison analysis in which I directly compare the MRIO datasets and analytical outcomes from the MRIOs. Finally, I conduct an economic study on wage inequality of a selected developing country case using Eora, one of the MRIO databases, combined with a micro dataset. Overall, the Global MRIO Lab innovation reduces MRIO compilation cost and expedite timely updates, as well as encourages new opportunities for more research uptakes using MRIO frameworks within the virtual MRIO laboratory by non-IO stakeholders from governments, statistical offices and the private sectors.