Efficient Investment Portfolios for the Swiss Electricity Supply Sector

Abstract

In this paper, we investigate existing and possible future power generation capacities in Switzerland from a risk-return perspective, using the Mean-Variance Portfolio Theory of Markowitz (1952). The study covers power generation technologies currently in operation, such as nuclear power, storage hydro power and run-of-river hydro power plants, and two new renewable energy technologies (photovoltaics and wind). Additionally, natural gas combined cycle (NGCC) technology, a possible extension to the current Swiss portfolio, is assessed. The technology-specific risks considered include electricity spot market price, production capacity and reliability, fuel cost, funding liabilities, and operation and maintenance outlays. These factors are implemented in a Net Present Value (NPV) model and Monte Carlo simulations are applied to assess each investment alternative. The lifetime-adjusted average return, together with the return-speci¯c variance, forms the basis for the portfolio optimization conducted in the second stage of the analysis. The minimum variance (or maximum return) optimization is performed separately for base-load and peak-load technology portfolios. By defining different scenarios for the upper and lower bound for each technology's share, we simulate different situations, enabling us both, to explain the risk-return profile of the current technology mix, and to make predictions for future portfolios. Our NPV calculations are in line with currently observed returns and, by imposing some reasonable restrictions, the model performs sufficiently well in terms of explaining past portfolio compositions. Moreover, our predicted optimal outcome matches quite nicely with the debated options for enlarging power production in Switzerland.

Keywords: Portfolio Optimization, Peak Load Demand, Electricity Supply, Switzerland

JEL Classification: G11, Q42