Within a public sector comparator model, risk refers to financial uncertainty in cash flow forecasts. Risk exists in a project whether or not it is measured and, as such, failure to measure and manage project-specific risk may result in poor cost estimates and sub-optimal service procurement.
The cost of risk is a function of the probability and consequence of occurrence of the risk event. Risks have both direct and indirect impacts where, for example, a construction cost overrun has a direct impact on the cost of a project. The same event may also have an indirect impact in terms of a higher borrowing requirement (to fund the cost overrun), leading to higher debt service costs over the life of the project. Here, the indirect impact may be more significant than the direct impact.
All material risks should be costed according to the expected costs (both direct and indirect) that would be incurred by government were such a risk to occur. The expected cost is the mean of all the probability-weighted cost outcomes for the particular risk event.
The most useful tool for risk identification and quantification is the performance history of similar past and existing projects. Such data should be supported by additional risk analysis as required, including:
∙ a review of similar projects in other Australian jurisdictions
∙ risk evaluation by specialist technical consultants
∙ the use of quantitative techniques to estimate risk probabilities and consequences
∙ the use of insurance premiums as an estimate of defined risks
∙ the use of insurance actuaries to evaluate unusual risks (such as discriminatory change in law)
∙ sensitivity analysis.
Risk valuation should reflect a commonsense approach, where it is important to measure only those risks that are material. It is of note that, in some instances, a number of similar risks may be immaterial by themselves, but may become material when aggregated. The timing of a risk also affects its materiality. For example, an increase in costs in year 25 of a project is less important on a discounted cash flow basis than a similar increase in costs in year 5.
The quantification of risks should take into consideration any risks transferred to the private sector under the reference project. That is, the public sector comparator should not be inflated by the cost of risks not expected to be borne by government. For example, if the reference project assumed a fixed-price turn-key contract, the risk of construction cost overruns will be commensurately lower than it would have been on a project alliance contract where government is exposed to a proportion of any cost increase.
The expected cost of all risks not transferred to the private sector should be added to the costs in the public sector comparator model. Similarly, the cost of shared risks should be pro-rated and added to the public sector comparator model.