Understanding the drivers of many environmental problems requires enumerating the global supply chain. Multi-region input–output analysis (MRIOA) is a well-established technique for this purpose, but constructing a multi-region input–output table (MRIOT) can be a formidable challenge. We constructed a large MRIOT using the Global Trade Analysis Project (GTAP) database of harmonised economic, IO, and trade data. We discuss the historical development of the GTAP-MRIO and describe its efficient construction. We provide updated carbon footprint estimates and analyse several issues relevant for MRIO construction and applications. We demonstrate that differences in environmental satellite accounts may be more important than differences in MRIOTs when calculating national carbon footprints. The GTAP-MRIO is a robust global MRIOT and, given its easy availability and implementation, it should allow the widespread application of global MRIOA by a variety of users.