Existing energy demand forecasts could grossly underestimate future energy use in the developing world.
About 1.5 billion people across the world live without electricity, mainly in developing countries. Observers believe that many of them eventually will obtain higher incomes through anti-poverty programs and economic growth, thus boosting their ability to acquire energy-using assets and increasing their energy consumption.
In How Pro-poor Growth Affects the Demand for Energy (NBER Working Paper 19092), Paul Gertler, Orie Shelef, Catherine Wolfram, and Alan Fuchs analyze data from Mexico after the government introduced a conditional cash transfer, Oportunidades, in the late 1990s as part of an ambitious anti-poverty initiative. The program subsidized families' income under the conditions that they meet certain health-care and educational requirements for their children. The cash transfers, that could increase for families as their children achieved certain milestone goals, totaled about $4 billion annually to 5 million households.
The authors find that this pro-poor transfer program had a large effect on the purchase of energy-using assets. Using both household data and country-level panel data, they conclude that existing energy demand forecasts could grossly underestimate future energy use in the developing world.
Government data and survey results from 1998, when the Oportunidades program began, through 2007 provide the authors with information on participants' purchases of household durable assets, such as refrigerators, gas stoves, TVs, and washing machines. Their actual energy expenditures also are tracked in later-year program surveys. The authors show that both the level and frequency of anti-poverty payments affect the rate of acquisition of assets. Asset accumulation is substantially greater when transfer payments are larger, and the effect is greater for households that are closer to middle class. Also, households are more likely to acquire assets when cash is transferred over a shorter time period.
Based on the findings from Mexico's conditional cash transfer program, the authors anticipate that there likely will be a surge in the demand for energy as families gain from economic development and anti-poverty programs. They also argue that the speed at which the poor emerge from poverty can affect their increased asset accumulation and energy demand. Thus, two countries with the same level of income per capita could have varying rates of asset ownership, resulting in different levels of per capita energy use.
Extrapolating the lessons learned from the household analysis to models of aggregate energy needs, the authors show that in countries with pro-poor growth, the income elasticity of energy is nearly double that of countries with GDP growth that was less favorable to the poor. "These results suggest that not accounting for pro-poor growth would grossly underestimate future energy use," the authors conclude.