Hedging Strategy and Electricity Contract Engineering

This thesis studies risk management in the electricity market in general and the interaction between physical production and electricity contracts in particular. From a risk management point of view, a power portfolio differs substantially from a traditional financial portfolio. Electricity is non-storable, which together with the marginal production cost characteristics creates jumps in the spot price. The return of a power portfolio is hence typically heavy-tailed, and a risk measure, such as CVaR, that captures this heavy-taildness is needed. To be able to compare production and contracts on a unified basis, we identify the set of contracts that corresponds to each power plant. These contracts build up a replicating portfolio of the power plant. This engineering of contracts allows us to risk manage these often complex contracts, through production. Further, a producing electricity company can through a simple absence of arbi¬ trage argument assess these contracts by studying the costs associated with the corresponding power plant. Flexible production units, such as a gas turbine, re¬ late to options whereas inflexible units, such as a nuclear plant, relate to futures. The electricity market is heavily incomplete, why perfect hedges are not achievable for a number of contracts. Hence we introduce the concept of best hedge. The best hedge is found through an optimization, where risk, measured as CVaR, is minimized subject to a constraint on the expected profit. It turns out that this problem can be solved with linear programming, allowing us to handle problems of substantial size. When a whole portfolio is considered we try to utilize our risk mandate at the best possible way. This leads us to the well-known problem in finance