Beyond manufacturing-led growth: institutional quality and regime-dependent structural transformation

Why do economies that undertake broadly similar industrial and institutional reforms follow markedly different growth trajectories? This study reassesses the manufacturing-led growth hypothesis by examining whether the gains from structural transformation vary across growth regimes and are shaped by institutional quality and resource dependence. Using an annual panel of 217 economies for 1981–2024, the analysis combines MICE imputation, two-state Markov-switching AR(1) regime classification, regime-specific two-step System GMM, observed-WGI validity checks, conservative instrument tests, lambda-sensitivity analysis, and an MI20 consistency check. Given the observational macro-panel design and the use of internal instruments, the estimates are interpreted as dynamic conditional associations, rather than causal effects. Manufacturing intensity does not emerge as a robust predictor of growth performance: in the adjusted baseline model, it is negatively associated with growth in the high-growth regime and weakly negative in the low-growth regime, while losing statistical significance in the observed-WGI, conservative-instrument, post-2000, and MI20 checks. Institutional quality is a more consistent growth-supporting correlate, operating primarily through a direct institutional growth channel rather than through manufacturing-based moderation. This study shows that the contribution of manufacturing to economic growth is not uniform across countries or growth regimes. Using a global panel of 217 economies from 1981 to 2024, the analysis finds that manufacturing intensity is not a consistently robust growth correlate once regime heterogeneity and robustness checks are considered, whereas institutional quality shows a more consistent positive association with growth. These findings qualify one-size-fits-all manufacturing-led development strategies and highlight the importance of aligning structural transformation policies with institutional capacity, growth regime conditions, and resource structure. Therefore, this study provides evidence for more context-sensitive development strategies rather than the mechanical expansion of manufacturing. The findings represent conditional associations rather than causal relationships.