Making informed decisions in government

The National Audit Office (NAO) published a report in May 2011 titled “Option Appraisal: Making informed decisions in government”. Some of the observations are quite shocking (or perhaps it’s more accurate to say that they should be shocking, and in some environments would result in heads rolling):

  • “We have regularly concluded in our value for money reports that departments could either not demonstrate the value for money of government interventions or that interventions provided poor value for money due to an absence of option appraisals.”
  • “…three-fifths of arm’s length bodies did not conduct investment appraisals to compare expected costs and benefits of alternative options before taking a decision to reorganise.”
  • “The main weaknesses we found were:
    • Inadequate development of options against which to judge the preferred course of action;
    • Lack of monetisation of burdens and benefits;
    • Unstructured qualitative analysis – while qualitative arguments were influential in a large proportion of cases, few followed guidance on ways to structure that analysis, or applied a qualitative structure consistently to all options considered.”
  • “In our interviews, Chief Economists acknowledged that departments often consider a narrow range of options and noted that promising options are often dismissed too early or discarded options not revisited when a change in scope would again make them viable.”

I think it is reasonable to ask why we tolerate this state of affairs. It clearly isn’t because solutions to these challenges are not known (in fact, the report points to government guidance from the Treasury Green Book that is not being followed). It seems more likely that individuals or teams are not aware of the guidance, or lack experience or training in how to implement it. Presumably there is also no sanction for failing to apply due process – many private companies would simply reject (or not even consider) any proposal for investment that was ‘out of process’.

Four years on it might be reasonable to expect that the situation has improved, but government machinery is vast and does not have a good track record of learning from past mistakes. It seems likely that significant decisions involving substantial public expenditure are still being taken using processes that are some way from best practice.

So what should we do about that? I think we need to look for every opportunity to draw attention to the issue; to offer information and training; to implement best practice; and to capture and publicise exemplar case studies. All of these are part of our plan.