Drawing from a structuralist analytical perspective on the external constraints and imbalances associated with late post-war industrialization, this paper explores the spatial dimensions of China's trade balances by disaggregating these according to available data across its 31 province-level administrative units and tracing these from the early 1990s onwards. Accounting problems aside, the results reveal an important distinction within these imbalances, in that the trade surpluses that have characterized China's overall imbalances since the early 2000s have been entirely generated by seven or eight coastal provinces, while the rest of the country ran sustained and relatively deep deficits, which in many cases descended into extremely deep deficits, particularly in the poorest central and western provinces. This distinction needs to be understood in the context of China's regional economic development strategies. These were facilitated by the centralizing fiscal reforms from 1994 onwards, which permitted an intensive reorientation towards interior industrial restructuring and the build-up of industrial hubs in the centre and west of the country, which had similarly served as industrial hubs during the erstwhile period of Maoist industrialization strategies, e.g. Chengdu in Sichuan, Lanzhou in Gansu, Xining in Qinghai. 

Two very different patterns of industrialization and associated imbalances have emerged from these strategies. Interior industrialization has followed a traditional association with deep and sustained trade deficits, as was common with East Asian industrialisation based on mixes of ISI and EOI, as well as with cases more commonly caricatured as ISI. In this sense, the understanding of China's balance of payments exceptionalism – at least since the early 2000s – really needs to located at a smaller spatial scale, around the Pearl Delta and Yangtze Delta regions, which have been specialised in generating the surpluses needed to relieve the external constraints faced by the domestically-oriented strategies of industrialization in the rest of the country. 

Hence, contrary to the idea that domestic rebalancing requires a shift away from external imbalances, this regional distinction highlights the dependence of domestic rebalancing on the maintenance external imbalances, insofar as the former are to be investment- and industrial- rather than consumption-driven. Nonetheless, the ability of the country to generate such surpluses at a national scale appears to be petering out, to the extent that the current account fell into deficit in 2018 for the first time since 1994, which puts into question the ability to sustain such strategies of domestic rebalancing.