This snapshot, taken on 05/09/2007, shows web content selected for preservation by The National Archives. External links, forms and search boxes may not work in archived websites.
It is impossible to predict the future with total accuracy. This means you will have to make assumptions when assessing the impact of an option and when comparing between them. These should be tested to ensure your final recommendation is not driven by a particular assumption.
To deal with risk and uncertainty you should state clearly what assumptions you have made. You should identify any specific risks or areas of uncertainty that may impact on the levels of costs and benefits.
For risks where it is possible to assign a probability to an event happening (e.g. the risk of fire or accident), you should use this to estimate the expected costs and benefits.
Often it is not possible to assign a probability to an event. You should use sensitivity analysis to analyse the impact of a number of different scenarios in which your assumptions are varied.
Differing levels of compliance with your proposal should be investigated. You should not assume that total compliance will always be achieved. It is important to understand what level of compliance is needed for the benefits to justify the costs.
There is evidence that policy-makers are often too optimistic when developing proposals. You should consider the impact of this “optimism bias
” on your benefits and costs. This can be specifically allowed for in your analysis.
Using these techniques you can produce ranges of benefits and costs. Such ranges allow decisions to be taken with consideration of problems that might occur. It is important that this is considered in the decision–making process as there may otherwise be undesired costs, undelivered benefits or costly action needed to rectify a problem later.
See the HM Treasury ‘Green Book: appraisal and evaluation in central Government’ for further discussion on testing your assumptions and dealing with risk and uncertainty).
It is important that this you present this section of your Evidence Base in a clear and non–technical way. Make sure relevant material is included on the Summary: Intervention & Options page and the Summary: Analysis & Evidence page and that any relevant research information, non–monetised impacts (see Specific Impact Tests) as annexes.