It is a new thing for New Zealand to have a chief science advisor to the Prime Minister.

Sir Peter Gluckman’s laudable brief is "to promote discourse that will lead New Zealand to better apply evidence-based knowledge and research across all domains of public endeavour".

This is an important but difficult mission, particularly as the quality of public policy analysis is frequently lamentable.

This is despite decades of effort to improve the quality of law-making by those involved in the Legislation Advisory Committee’s work and/or the Regulatory Impact Statement and the Regulatory Responsibility Act initiatives.

No comparable effort has been made to improve the quality of the analysis of government spending, although Dr Graham Scott’s 2010 report was a significant step in the right direction.

The efficacy of the bulk of government spending is not formally and rigorously assessed. It is easy to see why. The four largest functional items: health, social welfare, education, and New Zealand superannuation, are heavily politicised, and defended by entrenched vested interests. There is no agreement of any substance as to what problems with private arrangements justify these programmes in their current form.

To his credit, Sir Peter is making a serious effort to raise the quality of scientific evidence and analysis used in public policy formation. He defines science broadly to include the ‘social sciences’, so his task is really to improve the use of evidence in policy formation generally.

His challenging September 2013 report, The role of evidence in policy formation and implementation, follows two earlier reports: Interpreting science – implications for public understanding, advocacy and policy formation (April 2013) and Towards better use of evidence in policy formation: a discussion paper (April 2011).

For his September 2013 report, he surveyed 17 government departments and agencies to assess the use of evidence in their work. He found that some saw their role as doing what their Minister wanted rather than providing evidence on policy options. Some "were not clear on how research-based evidence could be used to support policy processes".

Indeed, poor quality advice persists because there is a demand for it. Vested interests advocate self-serving policies. Political parties compete for votes from vested interests. Unpalatable evidence can be politically embarrassing. Witness Sir Michael Cullen’s bullying ‘ideological burp’ response in 2005 to the Treasury's mainstream, evidence-based advice on the dis-incentive effects of high marginal tax rates.

Sir Peter Gluckman’s daunting brief is worth supporting.