A central principle of risk allocation is that each risk should be allocated to whoever can manage it best. Irwin's book on guarantees and PPP risk [#161, pages 56-62] defines this principle more precisely, stating each risk should be allocated to the party:
• Best able to control the likelihood of the risk occurring-for example, the private party is usually in charge of project construction, because they have the most expertise in that area. This also means they should bear the cost of construction cost over-runs or delays
• Best able to control the impact of the risk on project outcomes, by assessing and anticipating a risk well and responding to it. For example, while no party can control the risk of an earthquake, if the private firm is responsible for project design, it could use techniques to reduce the damage should an earthquake occur
• Able to absorb the risk at lowest cost, if the likelihood and impact of risks cannot be controlled. A party's cost of absorbing a risk depends on several factors, including: the extent to which the risk is correlated with its other assets and liabilities; its ability to pass the risk on (for example, to users of the service through price changes, or to third parties by insuring); and the nature of its ultimate risk bearers. For example, the ability of governments to spread risk among taxpayers means they may have lower risk-bearing cost than private firms, whose ultimate risk-bearers are their shareholders.
As described in the OECD's publication on risk sharing and value for money in PPPs [#194, pages 49-50], applying these principles does not imply transferring the maximum possible risk to the private sector. Transferring to the private party the risks that it is better able to control or mitigate can help lower the overall project cost, and improve value for money. However, the more total risk transferred to the private party, the higher the return-or risk premium-the equity investors will require, and the harder it will be to raise debt finance.
The principles and practice of risk allocation in PPPs is also increasingly the subject of academic research and literature. For example, Ng and Loosemore's article on risk allocation in PPPs [#190] describes PPP risk categories and allocation approach, and provides a case study of risk allocation in the New Southern Railway project (an underground airport--city rail link) in New South Wales, Australia. Bing et al's article on risk allocation in PPP/PFI projects in the United Kingdom [#31] assesses how risks have been allocated in PFI projects in practice, to identify risk allocation preferences. An IDB review of the Spanish PPP experience [#28] includes several examples of risk allocation used in different types of projects, from roads to hospitals.