The economic impact of power outages in developing countries: A quasi-experimental analysis

Power outages disrupt both economic activity (production and services) and household welfare. In April 2024, Ecuador reintroduced rotating blackouts, and Guayaquil experienced repeated interruptions that affected commerce. Restaurants are highly electricity-dependent because refrigeration, cooking equipment, and point-of-sale systems require continuous power. However, the rotation spared circuits adjacent to hospitals and other critical facilities; as a result, some areas faced outages while others remained supplied, creating a quasi-experimental contrast within the city. Therefore, the aim of this paper is to estimate the causal effect of power outages on daily restaurant sales in Guayaquil using a difference-in-differences (DiD) design that compares affected and unaffected areas. Additionally, triple-difference (DDD) checks interact the DiD treatment with indicators for (i) no electric generator, (ii) high electric dependence (exceeds 70 %), and (iii) long operating hours (more than eight hours per day). A separate heterogeneity analysis considers operational and demand-side margins and labor-input differences. Results show that outages reduced sales by about 15.3 %. Establishments without an electric generator experienced an additional decline of about 14.2 %; high electric dependence was associated with a 17.2 % drop; and operating more than eight hours per day was associated with a 10.2 % decline. Results are similar when daily fixed effects are omitted, under leave-one-day-out checks, and with alternative sample restrictions. This study provides the first quasi-experimental, firm-level evidence on the economic effects of power outages in a major Latin American city, complementing earlier approaches based on customer-damage functions, input–output and dynamic inoperability models, and instrumental-variables designs.