Household’s energy transition in South Africa: Evidences from regression discontinuity design and multiple discrete-continuous model

Start date

The primary goal of this research project is to better understand the consumption patterns of different fuels in South Africa, and to identify the causal role of income in spurring household energy transitions among the poor.

Households in developing countries display a primarily reliance on biomass fuels such as firewood, charcoal, animal dung, and agricultural residues to meet their major energy needs. Aiming to address the supply aspects of this problem, many developing country governments are now pursuing ambitious electrification programs. Much of the extant literature proposes that household income is a key determinant of the household energy transition toward cleaner energy sources. Unfortunately, the causal connection between income and fuel choice is difficult to pin down, given the obvious endogeneity biases that may arise from reverse causality and omitted (unobserved) confounding variables.

This research aims to address these empirical challenges by:

  1. exploiting exogenous variation in household income resulting from a unique natural experiment in South Africa, which will allow for cleaner identification of the causal effect of income on fuel choices); and
  2. applying a multiple discrete-continuous model to behaviorally and statistically account for the mixture of corner and interior solutions that result from fuel stacking, thereby providing empirical evidence of income and price effects at the extensive and intensive margins of fuel choices. 
Project status
Financed by
Environment for Development initiative
Project | 23 February 2018