Skill signaling, job mobility and wage dynamics: evidence from Ethiopia’s industrial parks

Purpose This study examines how labor market information interventions influence wage formation and job mobility in emerging industrial economies. Using Ethiopia’s flagship industrial park as a natural experiment, we evaluate the causal impacts of a randomized skill grading policy designed to reduce information asymmetries between workers and firms. The research investigates two critical questions: how do skill assessment interventions affect initial wage determination, and what role does strategic job mobility play in workers’ ability to overcome adverse policy effects in early-stage labor markets? Design/methodology/approach We exploit a unique natural experiment where workers entering Hawassa Industrial Park were randomly assigned to skill assessment versus direct job placement. Combining comprehensive administrative records with original survey data, we employ difference-in-differences and triple-difference designs that leverage the random assignment structure. Our matched employer-employee panel dataset tracks 1,206 workers across 19 garment firms, enabling identification of causal effects through the randomized grading treatment while controlling for selection bias and unobserved heterogeneity. Findings Skill grading creates an initial wage penalty of 5–7 percent for assessed workers relative to non-graded counterparts. However, this penalty is completely eliminated among workers who switch employers, with job transitions yielding 19 percent wage gains. The results demonstrate that information interventions can generate unexpected labor market mismatches, which workers strategically resolve through mobility. First job transitions produce the largest gains, with diminishing returns to subsequent switches, consistent with job search models where workers move to better matches. Research limitations/implications Our analysis focuses on short-term wage effects within a single industrial park setting, potentially limiting generalizability to other contexts or long-term outcomes. The study cannot distinguish whether wage gains represent permanent income increases or transfers through housing costs. Additionally, we lack comprehensive measures of match quality that could provide deeper insights into the mechanisms driving observed patterns. Future research should examine longer-term effects and extend analysis to diverse industrial settings and countries at different development stages. Practical implications The findings challenge conventional wisdom about skill assessment policies in developing labor markets. Policymakers designing industrial park interventions should anticipate that information-based matching may initially reduce wages but can be offset through worker mobility. Labor market policies should facilitate rather than restrict job transitions within industrial zones. The results suggest that infrastructure investments should be coupled with institutions supporting worker mobility to maximize welfare gains. Development practitioners should reconsider the design of formal recruitment systems to minimize initial mismatches. Social implications This research highlights unintended consequences of well-intentioned labor market policies, particularly relevant for industrial development strategies in sub-Saharan Africa. The findings suggest that worker agency through job mobility serves as a crucial mechanism for correcting policy-induced market failures. Understanding these dynamics is essential for designing inclusive growth strategies that avoid penalizing vulnerable workers. The results demonstrate how individual-level adaptations can mitigate adverse policy effects, emphasizing the importance of worker empowerment in industrial development contexts where formal institutions may be weak. Originality/value This study provides the first causal evidence on skill assessment policies in industrial parks using randomized assignment. The paper makes novel theoretical contributions by demonstrating how information interventions can create rather than resolve market inefficiencies, extending job search theory to developing country contexts. Methodologically, we introduce innovative use of administrative-survey matched data in industrial settings. The findings contribute to literatures on place-based policies (Kline and Moretti, 2014), job mobility (Stanton and Thomas, 2016) and labor market information (Pallais, 2014), offering new insights for industrial development policy in emerging economies.