Recent studies have evaluated municipal water systems in South Africa using difference-in-differences (DiD) models to assess their cost-effectiveness. However, these evaluations often lack replication and detailed methodological scrutiny. The methodology involves re-analysing data from a comprehensive dataset collected over several years. We employ difference-in-differences models to estimate cost-effectiveness metrics, accounting for potential confounding variables with robust standard errors and confidence intervals. Our replication confirms the initial findings but also identifies an unexpected trend where certain municipalities saw reduced water usage despite investment in infrastructure improvements. This study provides a detailed and validated methodological framework for future research on municipal water systems, highlighting specific insights into cost-effectiveness that were not previously evident. Future studies should consider the identified trends and incorporate additional control variables to enhance model accuracy. Policy makers could leverage these results to inform more targeted interventions in South African municipalities. Difference-in-Differences, Municipal Water Systems, Cost-Effectiveness, Robust Standard Errors Model estimation used $\hat{\theta}=argmin_{\theta}\sum_i\ell(y_i,f_\theta(x_i))+\lambda\lVert\theta\rVert_2^2$, with performance evaluated using out-of-sample error.