Abstract
At the end of 2007, as electricity demand surpassed supply, South Africa began experiencing widespread blackouts. Given the threat this posed to the national grid, ‘load shedding’ was introduced. Losses due to power outages are often associated with industry and business; but households too are end users of electricity, and their welfare is negatively affected by power outages, the result of increased electricity dependence over the years. South African households – like households in many developing countries – are faced with regular power outages. This is a serious problem, since the outages that households experience are both frequent and long in duration. Despite the efforts of all concerned, South African households will continue to face electricity supply challenges for the foreseeable future.
The primary objective of this study is to quantify household’s willingness to pay (WTP) to avoid power outages. The second objective is to estimate households’ WTP for nuclear-generated energy. The third objective is to assess the level of support for renewable energy. In this study, the contingent valuation method (CVM) is used to elicit outage costs, to estimate WTP for South Africa’s proposed second nuclear power plant, and to identify the determinants of support for renewable energy.
Face-to-face surveys were undertaken around Gauteng province, as well as in areas in close proximity to the proposed Thyspunt nuclear power plant in the Eastern Cape province. The surveys were conducted using electronic equipment (devices, or tablets) rather than the orthodox paper method. This analysis of power outages caters for different outage situations, including planned and unplanned outages, summer and winter outages, peak and off-peak outages, and weekday and weekend outages. Models used to assess the determinants of WTP to avoid power outages were the random parameter Tobit model, the standard Tobit model, and the Spike model, consisting of a probit model followed by a truncated regression model.
With a few exceptions, all the outage scenarios’ estimated parameters are significant at the one percent level, except for the off-peak scenario. The coefficients of all the slopes are positive, meaning that on average, costs are higher for planned outages, during winter, during peak times, and on weekdays. Moreover, WTP increases with duration, which was expected. Overall, WTP is driven mainly by duration of power outage, and by seasonality...
M.Com. (Economics)