Capital budgeting for interrelated projects in a real options framework

This dissertation addresses capital budgeting for interrelated projects within a real options framework and models the learning-by-doing nature of many projects (such as R&D). First, a simple model is developed and its analytic solution is derived. This model highlights the relevant differences between sequential development and parallel development (simultaneous development of multiple projects). By paying to develop a second project, the firm is buying the option to exchange or replace a potentially low valued project with a higher value alternative project. Under parallel development the firm always purchases the option to exchange in return for a guarantee that it can always implement the best of the alternative projects and so it can accelerate receipt of the option relative to sequential development. Sequential development postpones the decision on the option in order to utilize information of the outcome of earlier projects to make a more informed decision on the purchase of the option. The net value of the option to implement increases (i.e., parallel development becomes more likely) when uncertainty increases, development time increases, development costs decrease or correlation between project benefits decreases.; A more general numerical framework is also presented that studies optimal capital budgeting policy and combined project value for two mutually exclusive projects when the firm uses a dynamic investment policy, can accelerate investment, and faces budget constraint. This model can also be used for other types of project interrelations, more general cost functions and a wide variety of stochastic processes. We show that periodic review and potential alteration of investment policy adds substantial value. Optimal investment policy frequently includes parallel investment, is highly dynamic, and has a combined project value significantly higher than the value derived from the traditional NPV rule. These results have significant managerial implications since previous studies of managerial practices find about 20% of firms do not review and alter their investment decisions. Of the firms that do review their investment policy, most are using traditional NPV techniques that do not explicitly address or value flexibility.