Last month, federal Opposition assistant treasury spokesman and former ANU economics professor Andrew Leigh backed, in part, Nicholas Gruen’s excellent idea for a federal Evaluator General to oversee the evaluation of Commonwealth programs (see Building a better feedback loop). With such high-profile policy failures in the past as the pink batts and green loans schemes, an agency that can promote best-practice evaluation of policies is much needed. I expect the Evaulator General would promote the adoption of a Monitor-Evaluate-Report-Improve framework of continuous improvement for policy programs (see figure below). It would also promote the running of randomised control trials of government policies, a passion of Andrew Leigh’s.
Regrettably, Leigh’s proposed Evaluator General is a watered-down version of Gruen’s original concept, which was to have the Evaluator General as an independent statutory authority (see Gruen’s Mandarin essay), rather than as a branch of the Treasury, as Leigh has proposed. In my view, it would definitely work better as an independent statutory agency for the reasons Gruen has mentioned (e.g. it would be free from political interference) and also because:
- The Evaluator General should evaluate a range of tax and super policies overseen by the Treasury and it would be better to do that as an independent agency;
- It wouldn’t fit well in the Treasury bureaucracy, which is geared toward responding to the day-to-day needs of the machinery of government (e.g. question time briefs, coordination comments on cabinet submissions, ERC briefings, ministerial replies, speeches, etc) rather than the deep work of policy evaluation;
- There is a risk an Evaluator General branch of the Treasury would become the new Foreign Investment Review Board, the part of Treasury the star performers avoid because it would hold back their careers which depend so much on exposure to the Executive Directors, the Secretary, and the Treasurer’s office; and
- As a separate agency it would develop its own culture, separate from the strong existing Treasury culture which has its pluses (i.e. technical competence and an intolerance for errors) but also its minuses (e.g. risk aversion and a fanatical devotion to prosecuting the agenda of the government of the day, regardless of its deficiencies, leading to heroic efforts such as H.K. Holdaway’s 36 hour shift to finalise the legislation for FuelWatch, which ultimately the Rudd government couldn’t get passed in the Senate).
Also on the last point, I would say federal Treasury is the best public service example of Seth Godin’s mantra “people like us do things like this” (see my This is Marketing post), so I can’t see an Evaluator General within the Treasury developing its own culture and savoir faire which it will need to effectively discharge its duties. It needs to be separate, and I hope the Opposition will rework its Evaluator General proposal before the almost inevitable change of government in May next year.
Finally, I encourage you to read Nicholas’s submission to the Australian Public Service review on his Evaluator General concept: