Evaluation of Mass Electricity Connections in Kenya

In partnership with the Rural Electrification Authority (REA), our team of researchers will distribute national grid connection subsidies of varying amounts to randomly selected unconnected households in rural parts of Western Kenya. These subsidies will allow households to connect to the national power grid at relatively low prices (compared to the current connection price of $412). There will be three levels of subsidies: (1) high value subsidies equal to $412, (2) mid value subsidies equal to $235, (3) low value subsidies equal to $118.

The randomized encouragement design will allow researchers to generate exogenous variation in electricity connections. In the first stage of the project, the intervention will allow researchers to: (1) trace out the demand curve for electricity connections in order to better understand the price at which the majority of Kenyans would connect to the grid, and (2) trace out the economies of scale in costs resulting from coordinated (grouped) connections. In the second stage of the project, the intervention will allow researchers to measure the social and economic impacts of electrification. The remainder of this text will focus on describing the first stage of this project.

2014-04-28

2014-08-31

The following are the key outcome variables of interest in the first stage of the project:

1. Demand: Different levels of subsidies will create random variation in the effective prices faced by rural households. By comparing take-up rates at different prices, researchers will be able to trace out a demand curve for grid connections.

2. Costs: Different levels of subsidies will create random variation in the number of households applying for power at the same time. By connecting multiple households at the same time, utilities can experience significant economies of scale in transport, design, and low-voltage line costs. We will therefore estimate the effect of increasing scale (i.e. number of applications) on the average connection cost per household.

The study will take place across 150 “transformer communities” in Western Kenya. Each transformer community is defined as the group of all households located within 600 meters of a central electricity distribution transformer. In Kenya, all households within 600 meters of a transformer are eligible to apply for an electricity connection. The study will enroll roughly 15 randomly selected unconnected households from each transformer community. The subject population will consist of roughly 2,550 unconnected households in total.

The total sample of transformer communities will be randomly divided into treatment and control groups of equal size (75 treatment, 75 control). Each of the 75 treatment communities will be randomly assigned to one of three treatment arms (i.e. subsidy groups). Within each treatment community, all enrolled households will receive the same subsidy offer. After receiving the subsidy offer, treatment households will be given 8 weeks to accept the offer and deliver the required payment to REA. REA will then ensure that an electricity connection is delivered within a reasonable timeframe. Within each treatment community, connection applications will be grouped together, allowing REA to install multiple connections at the same time.

Randomization done in office by a computer

Randomization will be performed at the level of the transformer community. There are 150 transformer communities in total, of which 75 will be assigned to the treatment group and 75 will be assigned to the control group.

Yes

150 clusters

2,550 households

75 clusters control, 25 clusters high subsidy, 25 clusters mid subsidy, 25 clusters low subsidy

A note on pre-analysis plans (July 2014)

Pre-analysis plan: The demand for and costs of supplying grid connections in Kenya