PDF version

The Impact of Prepaid Electricity Meters on Low-Income Households in South Africa

Location: Cape Town
Sample: 2,394 households
2014 to 2015
Research Initiative: Urban Services Initiative
AEA RCT Registration Number: 

Electricity access and usage is growing in low- and middle-income countries, including South Africa. However, many households struggle to pay their monthly electricity bills, making it difficult for utilities to sustainably supply them with electricity. One possible strategy for overcoming this is to install prepaid electricity meters, which provide consumers flexibility in the timing of their electricity expenditures while also guaranteeing that utilities are paid for the electricity they supply. Researchers are conducting a rigorous evaluation of the effect of prepaid metering on electricity consumption, household welfare, and revenue recovery.

Policy Issue 

Global electricity consumption will increase dramatically in the next few decades, and much of this increase will come from developing countries. At their current capacity, utility companies may struggle to increase both access to electricity and the amount supplied to low- and middle-income households. In most developed countries, electricity is purchased on a credit basis: households consume electricity throughout the month and receive a bill for what they have already consumed. This model, however, is not well suited to the income and expenditure patterns of poorer households, who may struggle to save in order to pay a large, unpredictable bill. High rates of nonpayment may prevent electricity companies from serving these households.

One approach is to sell electricity on a prepaid basis. Prepaid electricity meters function much like prepaid cell phone contracts: consumers can purchase electricity in any amount and at any time, in advance of its use. When the balance on a meter runs out, the household' s electricity shuts off. Prepaid electricity may benefit both consumers, who have greater flexibility in how and when they purchase electricity and are better able to monitor their own consumption, and also utility companies, who recover a larger share of the money they are owed. While many countries across Sub-Saharan Africa have begun to experiment with prepaid meters, there is little rigorous evidence on the effects of prepaid meters on consumption and expenditure patterns and the ability of utility companies to collect payments.

Context of the Evaluation 

South Africa pioneered the use of prepaid electricity meters after apartheid, when the government rapidly expanded access to electricity across the country. While the meters were initially seen as a cost-effective way to provide electricity to very poor and rural households, they are now widespread, even in affluent areas. About 80 percent of residential consumers in Cape Town currently use prepaid meters. The City of Cape Town is looking to convert households that remain on a credit payment system to prepaid meters, including about 2,400 households in the low-income suburb where this evaluation takes place.

Aerial view of Cape Town at night. Photo: Anna Schickele | J-PAL

Details of the Intervention 

Researchers are partnering with the City of Cape Town to test the impact of switching from a credit-based billing model to prepaid meters on household electricity consumption, welfare, and utility company revenues. The City plans to install prepaid meters in 2,394 households in a low-income neighborhood of Cape Town. These households will be divided into 13 geographically contiguous blocks, which will receive the prepaid meters in a random order. Meter conversion is not voluntary, and households are told that their electricity would be shut off if they refused the installation.

Researchers will conduct two surveys: the first after approximately half of the households are converted and the second after all households are converted. In addition, researchers will use administrative data from the City of Cape Town on households' expenditures, consumption, and electricity debt in the analysis.

The randomized component of the evaluation, in a single suburb, will be complemented with analysis on a larger administrative dataset that covers the entire City of Cape Town. Evaluation using the larger dataset will rely on quasi-experimental methods to uncover impacts.

Results and Policy Lessons 

Project ongoing, results forthcoming.

Jack, Kelsey, and Grant Smith. "Charging Ahead: Prepaid Electricity Metering in South Africa." NBER Working Paper, December 2016.