A Difference-in-Differences Model for the Cost-Effectiveness Evaluation of Power Distribution Infrastructure in Kenya

{ "background": "The expansion and modernisation of power distribution infrastructure in sub-Saharan Africa requires robust, evidence-based methods for evaluating the cost-effectiveness of competing equipment systems. Current evaluation practices often lack rigorous counterfactual analysis, leading to potential misallocation of capital.", "purpose and objectives": "This study develops and applies a quasi-experimental difference-in-differences (DiD) model to empirically evaluate the comparative cost-effectiveness of two prevalent medium-voltage distribution equipment systems deployed across multiple regions.", "methodology": "A panel dataset of operational and capital expenditure for 42 distribution feeders over multiple years was constructed. The causal effect of equipment type on total cost was estimated using a two-way fixed effects DiD model: Yit = \α + \β (Treati \× Postt) + \ + \ + \εit, where robust standard errors were clustered at the feeder level. Cost-effectiveness was measured via the net present value of differential life-cycle costs.", "findings": "The insulated conductor system demonstrated a statistically significant reduction in average annual operational expenditure of approximately 17% (95% CI: 12% to 22%) compared to the bare conductor alternative, after controlling for feeder length and customer density. This operational saving offset its higher initial capital outlay within the evaluated period.", "conclusion": "The applied DiD model provides a rigorous, transferable framework for infrastructure investment appraisal, moving beyond descriptive comparison. The results indicate that the higher upfront cost of insulated systems can be economically justified through substantial operational savings.", "recommendations": "Utilities and regulators should adopt quasi-experimental evaluation methods for major infrastructure programmes. Procurement policies should consider life-cycle cost models informed by causal analysis, rather than relying solely on initial capital cost.", "key words": "difference-in-differences, cost-effectiveness, power distribution, infrastructure evaluation, quasi-experimental design, Kenya", "contribution statement": "This paper provides the first application of a difference-in-differences model to