Uncertainty in the electric power industry. Methods and models for decision support (Q1886076)

This monograph applies a large amount of mathematics to modelling the uncertainty facing the electric power industry, mostly in Europe and especially in Germany. The time horizon is 40 years out, and the uncertainties are enormous. The uncertainties covered here include fuels' prices, electricity prices, competitive behavior of firms in the industry, demand, technology, and regulatory and political. That's quite a list, and involves some quite sophisticated mathematics. To pick just one example, models of electricity and fuels' prices apply the latest stochastic process results from finance and financial economics. Among the interesting findings here are that the price of natural gas is the most volatile among the fuels' prices, and GARCH appears to fit best the behavior of electricity prices. With over 400 references, 300 equations, 100 figures, and 40 tables, the overall treatment is nothing less than encyclopedic. Two chapters stand out especially to this reviewer, even though these are least explicitly mathematical in the entire monograph. Chapter 8 on fuel cells is extremely insightful. The overall setting is technological uncertainty. Fuel cells are far from economically viable at present, and fit into only one type of current electric generation mode, Combined Heat and Power (CHP) systems. The hope for fuel cells rests on the learning curve: the more units produced, the lower their unit cost. The learning curve effect, first observed in aircraft production in 1936, has yet to receive an axiomatic foundation: it remains purely empirical. That said, it has been observed over and over again in many fields of production, including in electricity generation. Interestingly, even under the most aggressive of assumptions about steepness of the learning curve, fuel cells at best appear at the household and small business level in the next 40 years. They simply will not be economical in large CHP systems -- which is precisely where their developers currently target them. Chapter 9 on investment decisions encapsulates all the previous chapters (basically writing off fuel cells). It models the industry-level optimization problem at the state level of Germany (Baden-Württemberg) of choosing among gas turbine, gas fired, coal, and nuclear facilities over the next 40 years. The lowest cost solution involves a preponderance of nuclear capacity by 2040. If the German government persists in phasing out nuclear, then coal dominates-at vastly higher cost of production. Even in a book this good, at least one important uncertainty is left out: moral hazard by the management of the energy company (see Enron). Still, that is a small flaw in a book this comprehensive and this convincing.