Electricity Demand, Generation Efficiency and Costs in Kenya

Abstract

The government of Kenya has been making commendable efforts towards providing affordable energy to its citizens. However, the cost of electricity has been increasing despite reform programmes aimed at reducing costs. This thesis examined some of the critical considerations in the determination of electricity tariffs. Three essays were undertaken. The first essay examined the demand for electricity and made forecasts with a view to ascertain if the official demand forecast was realistic. Using autoregressive distributed lag (ARDL) model and time series data from 1985-2016 sourced from various sources including Kenya Power and Lighting Company (KPLC) annual reports, Kenya National Bureau of Statistics Economic Surveys and Statistical Abstracts, World Bank statistics and Kenya Electricity Generating Company hydro data. The findings showed that the official demand forecast was overstated and encouraged overinvestment in the generation of electricity. Overinvestment pushes the costs of electricity supply increasing the tariffs. Commercial and industrial consumers were projected to continue being the largest consumers of electricity. The finding indicated the need for the Ministry of Energy to revisit the planned investments and prioritize projects that address supply side constraints. The potential increase in costs arising from overinvestment can be prevented by signing take and pay power purchase agreements instead of take or pay removing the current protection offered to the generators. The second essay investigated the efficiency of thermal power plants. Using stochastic frontier analysis and data for 27 thermal generating power plants for the period July 2015 to December 2017 sourced from the power plants and the Energy Regulatory Commission, the study found the plants to be inefficient. Fuel was found to be a significant factor of production. Grid connected plants were found to be more efficient than isolated power plants. The inefficiency largely stemmed from age and ownership. The Malmquist data envelope analysis, however, found improved performance over the study period. To increase efficiency in generation, there is need for the regulator to revisit the methodology used for fuel oil cost adjustments. Ministry of Energy should also encourage private XVI

investments in generation and extend the grid to the isolated areas. The third essay sought to explain the electricity tariffs by exploring the drivers of KPLC tariffs. Since the tariffs are set using the cost of service regulation, KPLC cost data for the period 1986-2016 was used for the analysis. Average cost function of KPLC was estimated using ARDL model. The findings indicated output, system losses, system load factor and price of labour to be the drivers of average costs. System losses and price of labour were found to be increasing the average cost. The finding indicated the need for the regulator to set stringent loss reduction targets for KPLC. The Ministry of Energy should facilitate competition in the commercial retailing functions of KPLC as proposed in the Energy Act, 2019 to reduce the commercial losses associated with theft, corruption, billing and metering errors. The regulator should also tie allowed staff costs to improved customer service standards as a way of managing the cost of labour. KPLC was found to be enjoying economies of scale and economies of output density, this indicates the need for the electricity market to retain transmission and distribution of power as a natural monopoly. The Ministry of Energy should also continue with interventions and incentives that increase the system load factor such as time of use tariffs. Encouraging industrial parks and special economic zones through special tariffs could also increase the energy consumption and load factor. 1