Energizing Development: Scaling Decentralized Electricity Systems in East Africa.

by Gian Luca Gehwolf

Prof. Williams’ research is focused on trying to understand how universal basic services and infrastructure can be provided in a sustainable manner, with a strong emphasis on the generation and supply of electricity. He emphasised that infrastructural systems are tightly connected to electricity: “Imagine you are a small-scale farmer in Rwanda, and you have access to electricity to increase your productivity. If you have no access to markets, transportation, communication infrastructure or financial services to allow to grow your agricultural business, the electricity is somewhat useless.”

In East Africa, where a lot of his work is focused and where Carnegie Mellon University has a campus (CMU Africa in Kigali, Rwanda), only ten percent of the population has access to electricity. Electrifying the rural population using the traditional approach, i.e., via the grid, would be very expensive and require a lot of time. According to the UN, the population of Africa will double by 2050, meaning electricity demand will only increase. Hence, East Africa is facing a huge elec-trification challenge. Therefore, Prof. Williams focuses on mini-grid technology, which can provide electricity in remote places. He stated that the private sector is already pushing innovation in the off-grid sector, where mobile connectivity and its application to financial transactions are the driving forces. There are, however, a lot of challenges regarding the lack of clear policies, access to finance, and thus, the perception of risk among investors in the off-grid sector. Prof. Williams tries to quantify the technical and financial risks of the mini-grid sector in order to find ways to mitigate these risks. His approach includes the use of a stochastic techno-economic mini-grid model (STEMM), with the end goal of enabling the scale up of mini-grid deployment by facilitating access to finance. He also talked about models that predict electricity consumption from historical load data, including the application of regression models to relate customer survey data to subsequent consumption. The resulting customer load profile segmentation can help developers better understand their customer’s needs and help them improve design and operation decisions.

Towards the end of the talk, Prof. Williams came back to his earlier comments on the link between electricity production and consumption. There is evidence that newly connected customers use very little electricity. In other words, electricity is only useful in so far as it can be applied to useful services. This point led to a small discussion with the audience, where audience mem-bers proposed the market model of lowering the price of electricity to increase consumption. Prof. Williams argued that subsidising certain industries that use a lot of electricity had been considered, but that so far, the governments in East Africa have not been given enough thoughts on how to incentivise an increase on the demand side. He reasoned that energy systems intertwined with other infrastructure systems and services together enable development. He put forward the electricity-agriculture nexus as a key example. Agriculture is a crucial link to development as 54% of sub-Saharan Africans earn their livelihood from it. Nonetheless, productivity is extremely low, with many countries being net importers of food. Access to reliable energy is critical for each step of the agricultural-food chain (e.g., transport and logistics, primary production, food pro-cessing, etc.). Therefore, Prof. Williams’ team mapped agricultural returns on energy for irriga-tion. Their objective was to identify areas and crops with high potential for electricity use for irrigation. Furthermore, they tried to estimate yield response to irrigation and, lastly, develop geospatial datasets mapping irrigation potential at the electricity-agriculture nexus. They found out that the yield increase per unit energy varies spatially and with fertility. Finally, Prof. Williams concluded that we need to think more about how electricity is used in addition to how it is provided.

The ISTP would like to thank Prof. Williams for giving such interesting insight into the topic of energizing development and we hope to welcome him again at ETH Zurich in the near future!