On the back burner : experimental evidence for energy transitions

A central challenge in the global transition to cleaner energy is how governments can design policies that deliver large social benefits while facing trade-offs in energy security, fiscal costs, and household adoption frictions. We study this question in urban Nepal, where cooking is dominated by imported LPG, but abundant hydropower makes both large-scale electrification and improved energy security feasible. We embed household adoption decisions in a model of a planner balancing fiscal, fuel supply, and energy-security considerations, and estimate its key parameters using a scalable randomized controlled trial in Kathmandu Valley. Subsidies had large effects, increasing electric stove adoption by 23 percentage points and compatible cookware purchases by 41 percentage points. In contrast, information treatments highlighting cost or health benefits alone had little impact. Using detailed survey and electricity billing data, we find substitution away from LPG toward electricity, with meaningful household heterogeneity. Disciplined by these experimental estimates, the model evaluates counterfactual targeting rules, and estimates optimal subsidy levels under different macroeconomic conditions.