The energy sector is subject to a large number of uncertainties ranging from technological uncertainty, over price uncertainty to uncertainty about the future course of policy and regulation. Studying the effect that such uncertainty can have on energy investments is of paramount importance given the longevity of capacity in this sector and the huge costs involved.

Previous large-scale energy assessments have largely been focusing on deterministic modeling using scenario analysis as a tool for risk assessment. Even though this is useful, insofar as it delineates the scope of possible outcomes and contingencies in the face of uncertainty, such an approach does not capture the impact of uncertainty itself on the decision-making process. Instead, the investor optimizes for a given set of parameters, which are then varied to compare optimization results across scenarios.

With increased computing power allowing for more complex analysis and advances on the theoretical side, where modeling tools from the field of e.g. finance have been adopted, this is no longer the case. Risk preferences of investors and their valuation of flexibility in the face of uncertainty can now readily be accounted for. Even though such methods fail the test of practicality for large-scale bottom-up or top-down modeling, the insights provided are of great importance for both energy sector investors and policymakers.

In this talk, two methodologies currently employed at IIASA will be presented: real options theory and portfolio optimization. After an introduction to the approaches, several examples of recent IIASA work in this area will be given. This should give an impression of the potential usefulness of these methodologies to other applications as well.