A decision-support model for productivity upgrading in global value chains

Public agencies must decide which firms should receive scarce support for global-value-chain (GVC) productivity upgrading and which constraint should be addressed first. This paper develops a transparent model-based decision-support system (DSS) architecture that combines export/GVC exposure, operational capabilities, innovation, and frictions. Using 171,947 World Bank Enterprise Survey observations from 167 economies during 2006–2025, weighted high-dimensional fixed-effects models estimate an empirical upgrading kernel that is translated into ranking, sorting, and budget-constrained selection rules. Direct export intensity is positively associated with labor-productivity growth, but the most stable signals are digital presence, audited financial statements, product innovation, and low cash-operations friction; dominance analysis assigns more explanatory importance to innovation and frictions than to GVC exposure alone. The DSS therefore helps agencies classify firms into scale-up, capability-building, friction-reduction, or monitoring categories rather than targeting exporters mechanically. It is an exploratory, associational decision aid that requires local validation and recalibration before use in eligibility or funding decisions.