Electricity Outages, Adaptive Self-Generation and Firm Growth in Africa: Doubly Robust Evidence from Seven 2025 World Bank Enterprise Surveys

Reliable electricity is a productive input, yet firms across much of Africa operate under recurrent outages and respond by installing costly self-generation. This article examines whether outage exposure and intensity are associated with firm growth and whether generator ownership moderates the relationship. We harmonize the 2025 World Bank Enterprise Surveys for Burundi, Comoros, Guinea, Guinea-Bissau, Liberia, Malawi and Nigeria, producing a pooled sample of 1,981 formal firms. The empirical design combines country-normalized sampling weights, country fixed effects, rich firm controls, heteroskedasticity-robust inference, country-clustered inference, continuous outage-intensity models and five-fold cross-fitted augmented inverse-probability weighting (AIPW). Approximately 83.1% of firms experienced an outage, 69.7% owned or shared a generator and 15.6% used solar power. In the weighted fixed-effects specification, outage exposure is associated with 1.20 percentage points lower annualized sales growth, although the estimate is imprecise. The doubly robust AIPW estimate is −1.85 percentage points (p = 0.088). A one-unit increase in log monthly outage-hours is associated with 0.50 percentage points lower sales growth; statistical significance is sensitive to the variance estimator. Employment effects are negative but imprecise. Generator ownership neither predicts faster growth nor significantly attenuates the outage penalty, suggesting that private backup is an incomplete and potentially costly substitute for reliable grid supply. The findings are best interpreted as robust conditional associations rather than definitive causal effects because outage exposure is non-random and the surveys are cross-sectional. Policy should prioritize reliability enhancing grid investment while expanding cleaner, financeable resilience options for firms.