Efficiency analysis in the public sector assesses the performance of municipal and government entities in their provision of services to the public. South Africa, in particular, has experienced a large degree of negative feedback with respect to the ability of the government and its municipal departments to provide basic services to citizens. This has led to a number of service delivery protests throughout the country. To remedy this, the ability of the municipality to provide basic services needs to be assessed in order for improvements to be made. A first step in this process would be to determine the efficiency with which municipalities are providing these services. This study focuses on a particular municipal service, namely electricity distribution. Primarily, the efficiency with which municipalities provide electricity to consumers is determined. This is achieved using the parametric cost frontier approach, which is appropriate since municipalities aim to reduce the costs incurred in distributing electricity. The municipalities are compared to a frontier (theoretical best practice) from which inferences on their performances can be made. Those municipalities whose performances are not optimal, deviate from the frontier. The deviations (errors) are then assumed to be caused by two factors, namely stochastic random noise and inefficiency. This composition accounts for effects for which municipalities cannot control (stochastic random noise) and those for which it can (inefficiency). The parametric nature of the cost frontier approach allows for distributional assumptions to be made on these factors. Stochastic random noise is always assumed to be normally distributed, while inefficiency is assumed to be one-sided and positively skewed.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:nmmu/vital:26650 |
| Date | January 2016 |
| Creators | Gqwaka, Aviwe Phelele Sebatian |
| Publisher | Nelson Mandela Metropolitan University, Faculty of Science |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis, Masters, MSc |
| Format | ix, 112 leaves, pdf |
| Rights | Nelson Mandela Metropolitan University |
Page generated in 0.0016 seconds