Project performance-based optimal capital structure for privately financed infrastructure projects

Privately Financed Infrastructure (PFI) projects are characterized by huge and irreversible investments and are faced with various risks. Project performance risks, such as project completion time and costs, affect the project value significantly, particularly in project development phase. This is because a major part of the project investments is made during this phase. Due to high uncertainties in managing the project performance risks, the selection of optimal financial structure is a challenge to Project Company sponsors and Lenders. Conventional project performance measurement and valuation methods cannot capture the dynamics of risk variables and their impact on the project value. Without such dynamic performance information, the decision of capital structure may not only be suboptimal, but lead to erroneous results. This research proposes an uncertainty evolution model, with which the dynamics of the project performance risk variables can be predicted at any desired time over the project development phase. A dynamic capital structure model is proposed that explicitly considers the performance risks and adjusts the capital structure dynamically to counter the impact of performance risks. Numerical results show that such a model can add a significant value to a PFI project. Two risk-sharing mechanisms are incorporated in the capital structure for a PFI project: active project management (self-support) and government support. An active project management method called dynamic crashing is proposed. By dynamically controlling the project performance through dynamic crashing, we show that the project value can be unproved and the chances of potential bankruptcies can be reduced. In addition, the significance of government support as a risk-sharing mechanism is also modeled, which may be viewed as another means to protect the Project Company against the potential bankruptcies and improves the project value. Numerical results are implemented to validate the models. Overall, this research contributes an integrated framework to capital structure decisions for projects with performance uncertainties.