Utility Investment Planning using Risk Analysis: A Case Study of Tanzania’s Power Sector

Abstract: Investments in infrastructure projects in developing countries are often seen as risky by private investors and local capital markets are not fully developed. Consequently, financial resources and guarantees are often to a large extent provided by financial institutions. Financial and economic analysis is important in assessing infrastructure investment proposals in order to assure that they are financially sustainable and make the best use of scarce resources. Investments in the energy sector are associated with some special risks, such as uncertainty about hydrology variations, volatile oil prices, future demand etc. Many traditional methods for assessing investment programs are not very well equipped to handle risk. Tanzania is a development country, which is receiving support from different donor countries and agencies for the purpose of reforming the energy sector and for increasing the access to electricity within the country. At the same time, Tanzania’s and its power sector has been affected by severe droughts, leading to shortage of productive capacity and load shedding. Objective: The purpose of the thesis is to develop a method which combines quantitative risk analysis with traditional methods for power sector investment appraisal. The method shall be demonstrated by creating a financial model and adapting it to a real case. The model shall be able to consider identified, inherent risk factors and their conceivable impact on the power system. Methodology: In order to fulfil the purpose of this master thesis, a quantitative case study has been conducted. The case studied is a capital investment program for Tanzania’s power sector, including identification of inherent risk variables and their effects on the financial viability of the program. Three risk variables are identified as the most critical for the case; (i) the annual contribution of hydro-power to the system; (ii) the price of oil; and (iii) the specific demand for electricity. The data required for the analysis of this thesis has been collected from multiple sources of information. The main source has been documentation, used for empirical data as well as for the theoretical framework. Further, archival records have been widely used for the study as well as open interviews and personal correspondence with employees at Tanzania’s major power utility, TANESCO. Finally, the author has used direct observational sources of information during her two years of consulting assignment at TANESCO, working with related issues. A financial Excel model has been developed for the case study and is used as a tool i to estimate the probability distribution of financial indicators. For the purpose of performing the quantitative analysis, a risk analysis tool, namely @Risk, has been combined with the financial model. The technique used by @Risk is Monte Carlo simulation. Conclusion: In order to demonstrate the flexibility of the method for quantitative risk analysis, the case was analysed from various angles, i.e. base case analysis, stress analysis and variation of the investment program etc. The overall advantages of the method can be summarised as follows. Firstly, calculating the probability distribution of the financial indicators enhances the basis for decision and quantifies the company’s risk exposure due to the investment program. Further, as a result of the simulation, the full range of expected outcomes of the risk variables is analysed and the result is presented in a lucid way. For a decision maker, the combined influence of all risk variables can easily be observed. Moreover, the method can easily be extended to more than three risk variables as well as provide information on whether a risk variable is negligible or not. Stress analysis provides the decision maker with additional insight. Finally, risk assessment increases the opportunity of identifying effects of changes in investment plans and comparing them with the base case both in terms of expected benefits and in terms of associated risk exposure. However, the method has some drawbacks as well. The method involves a difficulty in the identification of risk variables as well as estimating the distribution functions for the variables. Quantitative risk analysis also makes the basis for the investment decision more complex, as the decision is not based on a single net present value