Public policy perfection – The Select Committee

By Chris Johnson

Tuesday July 21, 2020


Is achieving perfection in public policy a mere pipe dream? And is being perfect everything we think and hope it is anyway? Could perfection undermine what is actually good when it comes to forming public policy?

All very deep and philosophical indeed. So, we put the whole topic before members of The Select Committee – The Mandarin Brains Trust.

The specific question being:

The paywall is down

Save $220 and receive a free copy of ‘The Nordic Edge‘ book

Our best offer

Would you accept that there is no such thing as a perfect public policy — or, if not, would you agree with the contention that the perfect can sometimes be the enemy of the good?

Their responses, which follow here, are once again highly informative and contain many examples illustrating valid points. Lucy Turnbull AO warns against using the excuse of ‘perfect’ for not trying to improve. Stephen Bartos discusses why policy development involves compromises and how striving for perfection can be counterproductive. And Richard Bolt PSM expounds an argument that the only reliable test of good policy is the extent to which it works better than the policies it replaced.

Lucy Turnbull AO

Lucy Turnbull AO is an urbanist, businesswoman and philanthropist with a longstanding interest in cities, and technological and social innovation. From 2015-20 she was the inaugural Chief Commissioner of the Greater Sydney Commission, tasked by the NSW state government to assist in delivering strong and effective strategic planning for the whole of metropolitan Sydney. Lucy was the first female Lord Mayor of the City of Sydney from 2003-4.

“Is there ever perfect public policy? Not often, if ever.

“A wise person (a very eminent architect) revealed something when he was talking about designing and building: there is a triangle of outcomes with any project, and any policy. The points of that triangle are: Cost; Quality; and Time.

“You can only ever have two of those at the same time. I went away for a long time and thought about it, and he is absolutely right. That formula certainly always works for building and infrastructure projects. To disprove that proposition, please think of a major project that was simultaneously fast to deliver, relatively cheap and high quality? Doesn’t often happen.

Lucy Turnbull

“Projects and policies, like life itself, are always balancing acts. And the biggest enemy of improvement and getting to a better balance is the (usually) unobtainable need for perfection. So yes, the perfect is often the enemy of the good, and even the ‘better’. Opponents of change will always resist change by saying the change is not ‘perfect’ – it will cost more, lead to uncertain outcomes, (aka is politically risky), not do enough (why bother?).

“And in the process, the momentum for and value of incremental and adaptive policy improvement and change is lost. Climate change policy is the perfect example of this.

“Think of any major policy area, to name just a few: climate change, health, education, major infrastructure investment, and certainly in my experience, urban planning. The perfect is often cited as a reason not to improve and not to do anything, just amble along. But in a fast-changing world, this is very dangerous position to take.

“The COVID-19 pandemic has given us a vivid case with the CovidSafe app of having to balance time (because of the risk of doing nothing) and quality. While a lot of effort was made in a short timeframe to try to develop a workable contact tracing app, it turned out not to work when smartphones are in hibernation. It has not worked out as was expected. So, the question is: do we adapt and do something to improve it, or look at other solutions (like those proposed by large tech giants)? We probably should but the political risk of backing down or ‘backflipping’ is too great. Thus, quality suffers. The benefit of wearing of masks is another international policy controversy that is a strong contraindication of rational scientific based policy development in an emergency.

“We need to do better at the same time as we quest for policy ‘Nirvana’. And that means we have to admit to sometimes making mistakes and learning from them.” – Lucy Turnbull AO

Stephen Bartos

Stephen Bartos is the former deputy secretary of the commonwealth Finance Department and a director of Pegasus Economics. He is author of two books Against the Grain – The AWB Scandal and Why it Happened (UNSW Press, 2006) and the reference manual Public Sector Governance – Australia (CCH, 2004). He was also Professor of Governance and Director of the National Institute of Governance at the University of Canberra.

“Policy almost always involves compromises and trade-offs. In economics, there’s even a ‘theory of the second best’, which suggests situations arise where public policy simply cannot achieve an optimum solution (although an economist I once worked with thought governments were lucky if they ended up with fifth best!).

Stephen Bartos

“Striving for perfection can be counterproductive. The Rudd government’s Carbon Pollution Reduction Scheme was comprehensively modelled, tested, amended and embellished, so became too hard for the government to explain. Julia Gillard’s simpler scheme was introduced successfully – albeit it only lasted until the change of government.

“In the reforms of the 1980s and 1990s, although the government and its advisers tried to do their best many of the best policies were introduced with imperfections and then refined in light of experience. A good example was the running costs system for departmental budgets. It was far from perfect and had to be changed many times. Arguably, policies like this can never be perfect, always need adaptation, because the world to which they apply is constantly changing.

“Governments need policies to solve problems or create opportunities: so imperfect policies are better than none. With an important proviso: policies need regular evaluation. An ineffective policy should be abandoned, or if the need still exists, replaced with another. The new policy may also be imperfect – but natural selection, by repeated review and evaluation, will eventually result in the survival of a policy fit for its purpose.” – Stephen Bartos

Richard Bolt PSM

Richard Bolt PSM is a Principal of the Nous Group of management consultants, and an Adjunct Professor of Energy Transformation at Swinburne University. He has degrees in electrical engineering and public policy and management. Richard had a distinguished career in the Victorian Public Service, leading three departments over 12 years with responsibility for primary industries, education, and economic development and transport. 

Every diamond has impurities, every loving relationship has tensions, and all public policy includes unavoidable compromises. Perfection does not exist.

“Indeed, there is no agreed theoretical standard against which the perfection of a policy can even be measured. Policy is the product of competition between values and tribes, so its quality is a contestable perception, not a universal truth.

Richard Bolt

“Even in the apparently objective discipline of mathematics, perfection is unattainable. Although some mathematical propositions have generally agreed proofs, the Austrian Kurt Gödel showed in the 1930s that others cannot be proved at all. Because maths can’t demonstrate its own internal consistency, we wouldn’t recognise perfection if it was staring us in the face.

“In physics, where maths meets the real world, perfection is even further away. Heisenberg’s uncertainty principle establishes that if we know exactly where a particle is, we can’t tell precisely how fast it’s moving – and vice-versa. We still don’t even know what light is, because it behaves like weightless particles in some settings, but like waves in others.

“Comparing policy with maths and physics is not a glib analogy. The rise in mathematical models and computing power has given policy a false aura of science and precision. Socio-economic models are where maths meets humanity, and there are deep limitations on our power to predict the impact of a policy on behaviour – whether of a person, family, community, industry or nation.

“In other words, there are no laws that reliably forecast the socio-economic outcomes of government intervention. We can extract some useful cause-and-effect relationships from burgeoning data pools, as Raj Chetty has done in linking social mobility to the age at which a poor child moves to a rich neighbourhood. However the measured effects are partial, variable and sometimes temporary.

“A case in point is the experimental evidence on early childhood education. In controlled settings involving small groups of skilled educators, impressive gains in cognition arise when vulnerable pre-school children are exposed to play-based learning. However, the benefits are less clear and durable when interventions are scaled up from labs to whole systems and measured over longer periods of children’s lives.

“This and many other examples demonstrate that a policy’s quality can’t be measured against a design standard. The only reliable test of good policy is the extent to which it works better than the policies it replaced. This requires trial and error, with the interplay of research and politics prompting adjustment and sometimes radical reform. Reform in turn leads to further adaptation, and the cycle continues. What works in one community or era may or may not work in others.

“Policymaking is unavoidably imperfect and adaptive. Our focus should not be on finding one policy construct to rule them all, but to improve the system and cycle of research, advice, engagement and decision-making to more rapidly get good results, which will generally be a practical and adaptive mix of regulation, subsidies, pricing and markets.

“There is no golden policy tool for most challenges, just the practical bundling of measures that have the best chance of working…”

“By way of example, a whole-of-economy carbon price is often asserted to be the ideal policy for mitigating climate change. However, recovering carbon costs from energy consumers is less socially progressive and politically tractable than recovering them from taxpayers. Together with the fact that pricing produces partial and sometimes perverse responses, this has led governments to mute or entirely forego carbon prices in favour of regulations and subsidies.

“Carbon prices are also subject to policy risk, because they are created by governments and can be changed by them. This can undermine the private sector’s confidence to invest in emission reductions, which prompts governments to directly underwrite or mandate investments in clean energy.

“And carbon prices can cause coal-fired power stations to become financially distressed, hence less well maintained and increasingly unreliable in the leadup to closure. Although a central purpose of carbon prices is to close high-emission industries, premature closure can cause blackouts, price spikes and a loss of public confidence in decarbonisation. Thus compensating policies may be required, such as gifting emission credits to incumbent power producers and requiring them to give a minimum notice of closure, in the hope that their exit will happen in the right way at the right time.

“Carbon policy is just one of many examples in which governments don’t have the luxury of solving single problems with theoretically satisfying measures. Yet for decades, public policy practice has had a strong preference for the ostensible ideal of pricing public goods and using markets to optimise their consumption and conservation. While in some circumstances markets produce useful policy results (for example renewable energy mandates and tradeable fishing quotas), even those applications have limitations. In other cases they have not worked well (such as uncapped subsidies for skills training which led to gaming by providers and budget blowouts).

“Not only have results been mixed, but political support for marketisation is under challenge not only from the practical difficulties of market design, but from a more fundamental public concern about economic inequality and business ethics.

“Imagine the outrage if governments introduced capped and tradeable permits for interpersonal contact to efficiently optimise COVID-19 infection levels and economic activity. We are more accepting of regulation (lockdown laws) and compensating subsidies (JobKeeper grants) not because they are perfect, but because they are workable and accord with the moral obligation to avoid catching and spreading the virus. However, our attitudes, and governments’ responses, are shifting almost daily in response to infection rates, emerging evidence on how the virus spreads, the state of the economy and our capacity to endure isolation.

“There is no golden policy tool for most challenges, just the practical bundling of measures that have the best chance of working, followed by adaptation in response to evidence in an attempt to pass the test of political and public acceptance. This messy, contested and adaptive process is inevitable.

“However, we can do more to reach better policy answers more quickly. Australian governments still shroud their most important decisions in Cabinet secrecy that has its roots in the once unquestionable power of monarchy. Coupled with the artificial polarisation and ‘celebritisation’ of politics, this has made effective and adaptive solutions to complex challenges needlessly hard to achieve. Equality of access to government decision-making is as important as equality of access to income, wealth and essential services.

“Whether COVID-19’s boost to evidence-based policy and consensual decision-making lasts beyond the pandemic remains to be seen. What’s clear is that applying the art and science of good policy to the system of policy-making itself has become a critical need.

“Policy perfection is a dream, but good policy is needlessly hard. We need it now more than ever.” – Richard Bolt PSM

Our best offer. Save 50%

For two weeks only, we’re making all our Premium content completely free. Sample then subscribe to Premium with our best offer and save 50% ($220). 

Offer ends midnight 2 August 2021. 50% discount available on an annual subscription only.

Chris Johnson
Managing Editor

Subscribe today
About the author
Inline Feedbacks
View all comments

The essential resource for effective
public sector professionals