Constructing a Time Series of Nested Multiregion Input–Output Tables

Abstract

We develop a large-scale high-resolution time series of nested multiregion input–output (MRIO) tables, encompassing a range of technical advances that are relevant for MRIO applications worldwide. First, our database is the first ever hierarchically nested system of subnational and international MRIO tables on three independent counts: (a) it features global country-level coverage, (b) it is available as a long annual time series, and (c) it is complemented with matching information on element uncertainty. Second, it is at the time of writing the largest existing MRIO system in the world, and in its creation a number of challenges related to computer storage and run time had to be overcome. The MRIO tables feature complete interregional trade at this level of detail, in combination with detailed regional–international trade with 185 countries. Our experiences with constructing such a large and detailed framework contribute knowledge needed by practitioners wishing to assemble similar databases for other countries, in that our build pipeline can readily be adopted for the integration of subnational MRIO databases, for example, for the United States, China, Australia, Spain, and Germany. We demonstrate our approach by constructing a time series of MRIO tables for the example of the Chinese economy between 1997 and 2011, distinguishing each of the 30 provinces and 135 industry sectors for each province, and linking each province with 185 world countries.