In developing countries in Asia and Africa, nearly a billion people still live without electricity at home. In an effort to solve this problem, governments and the private sector continue to invest a significant amount of time and money with the object of expanding the grid, or developing decentralized renewable electricity solutions. Notwithstanding these efforts, many households continue to live without modern energy at home. To understand how best to reach the goals of universal electrification it is critical to fully understand the nature of household demand for energy, and decision-making in a marketplace with several electricity sources.
Why this study
This study uses a randomized control trial to identify the demand curve for off-grid solar power and understand the preferences of households over different electricity sources (grid, solar microgrids, diesel generation etc). The experiment also seeks to measure short- and mid-term impacts of access to small amount of electricity, provided through a solar micro-grid.
Evidence on the benefits of the first few units of electricity, as well as the nature of demand are critical to designing economically efficient policy instruments that seek to encourage household electrification. The data gathered and the corresponding models that have been estimated also provide a useful means of understanding when, and to what degree, decentralized renewable electricity supply sources are likely to be adopted by consumers.
Researchers did a randomized experiment on pricing and availability of oﬀ-grid solar power to estimate demand for competing electricity sources. The experiment was conducted in rural Bihar—an eastern state in India—from 2013 to 2017 with the help of the implementing partner Husk Power Systems (HPS), a Bihar-based private company that oﬀers solar microgrids.
The study covered roughly 3,000 households in 100 sample villages across three districts. These villages were not listed as ‘electriﬁed’ by the government; they had not been oﬀered HPS micro-grids, and were located reasonably close to existing HPS operating sites so that micro-grid services could be feasibly expanded to these households. Total population of households in all 100 villages was 48,979.
Three rounds of surveys were conducted. The first two surveys—baseline and endline—covered household demographics, income and occupation, time use, electricity access and billing, appliance ownership and use, and health and education measures. The final round of surveys, endline 2, only covered electricity access.
The study developed three treatment groups:
- Pure control group (34 villages) where HPS microgrids were not offered;
- Normal price group (33 villages) where microgrids were oﬀered at the prevailing price;
- Subsidized price group (33 villages) where microgrids were offered at a subsidized rate
Researchers collected data on cost structure and customer base from diesel generator operators, and administrative data from HPS on consumer payments. Data from state utilities on supply conditions and customer payments was also collected.
Experimental estimates suggest that demand for micro-grid solar power is highly price elastic. At a market price of INR 200 per month, the demand for HPS solar micro-grids was near zero. It was also near zero at INR 160 per month. Only about 8 percent households demanded off-grid solar at a market price of INR 100 per month. The result is in contrary to the conventional wisdom that there is substantial demand for solar power in remote areas of developing countries.
Moreover, there was little evidence that the electricity provided by solar microgrids had large impacts on health or education, although the estimates for education outcomes may not be precise. While the households still value off-grid solar for lighting and other energy services, they may perceive greater benefits from different electricity sources, leading to different willingness-to-pay, take-up, and use, according to the findings of the study.
Read the paper: Demand for Electricity on the Global Electrification Frontier
For this project, the Energy Policy Institute gratefully acknowledges generous research support provided by the Tata Centre for Development (TCD) at UChicago.