Beyond the Average Effect: Heterogeneous Adjustments in East Asian Production Chains

Abstract

This study examines exposure-linked adjustments in East Asian production-chains outcomes following the 2018 US–China trade conflict and during the post-2021 period. Using a technology-tiered panel of economy–sector cells and multi-regional input–output (MRIO)-based indicators—value-added absorption (VAX) and backward/forward average propagation lengths (APLB/APLF)—this paper assesses how pre-shock export exposure relates to post-period changes in network structure. To capture heterogeneity beyond mean responses, it estimates recentered influence function (RIF) regressions with fixed effects for unconditional quantiles and random-coefficients mixed models that recover unit-specific post-2021 exposure slopes. The results indicate that post-2021 adjustments are heterogeneous and systematically structured. Distributional estimates show that exposure-linked effects are most detectable in the lower-to-middle portions of the unconditional outcome distribution rather than concentrated exclusively in the upper tail. Unit-level predictions further reveal pronounced technology-tier divergence: primary-processing (PP) sectors are substantially more likely to exhibit positive post-2021 exposure slopes, whereas resource-based, low-tech, and especially high-tech (HT) sectors are predominantly negative. These findings suggest that post-2021 adjustment is characterized by tier-specific divergence rather than a uniform shift, implying that resilience and industrial policies should be differentiated by technology tier— emphasizing trade facilitation and input-flow reliability for upstream processing segments while supporting diversification and capability upgrading in technology-intensive manufacturing.