Predictive Preparedness and Recurrent Shock Reduction in Somalia: The Roles of Adaptive Governance and Institutional Fragmentation.

Abstract

Introduction: Somalia faces recurrent shocks including droughts, floods, conflict, and displacement that overwhelm governance systems and sustain reactive humanitarian responses. This quantitative cross-sectional study examines the associations between predictive preparedness, adaptive governance, institutional fragmentation, and recurrent shock reduction in Somalia, testing adaptive governance as a mediating mechanism and institutional fragmentation as a boundary condition.

Methods: Survey data from 315 disaster professionals, government officials, humanitarian workers, and community leaders across Somalia's federal member states were analyzed using partial least squares structural equation modeling (PLS-SEM) with 5,000 bootstraps. Multi-item scales measured predictive preparedness, adaptive governance, institutional fragmentation, and recurrent shock reduction based on self-reported perceptions.

Results: Predictive preparedness was positively associated with recurrent shock reduction (β = 0.28, p < 0.001) and adaptive governance (β = 0.42, p < 0.001). Adaptive governance partially mediated this relationship, with the indirect effect accounting for 35% of the total effect (indirect β = 0.15; 95% CI: 0.07, 0.23) and the direct effect accounting for the remaining 65% (β = 0.28; 95% CI: 0.16, 0.40). This partial mediation indicates that predictive preparedness operates through both governance-mediated and direct pathways. Institutional fragmentation significantly weakened the predictive preparedness-adaptive governance relationship (interaction β = -0.18, p < 0.001) and the adaptive governance-shock reduction relationship (interaction β = -0.22, p < 0.001). The conditional indirect effect was positive at low fragmentation (0.10, 95% CI: 0.04, 0.16, p < 0.001) but non-significant at high fragmentation (-0.10, 95% CI: -0.18, -0.02, p = 0.012). The model explained 48% of variance in shock reduction and 35% in adaptive governance.

Discussion: Predictive preparedness is associated with recurrent shock reduction through both direct and governance-mediated pathways, with governance serving as a force multiplier rather than a prerequisite. Institutional fragmentation severely constrains these associations, fully attenuating the mediated pathway at high fragmentation levels. These findings suggest that preparedness investments should not be withheld pending governance reforms, as direct benefits may accrue even in fragmented contexts, but governance strengthening should proceed in parallel to amplify benefits. However, given the cross-sectional design and self-reported measures, all relationships should be interpreted as associations requiring longitudinal validation.