Counting the cost

  • Authors: Leo Dobes
  • Published Date: 08 May 2019

For economist Dr Leo Dobes and his co-authors Joanne Leung and Dr George Argyrous, all is not well with the evaluation of government programs and projects. Here, Dr Dobes makes his case for boosting the capacity of government to conduct rigorous social cost-benefit analyses of proposals.

Dr Leo Dobes will be delivering ANZSOG's executive education workshop Cost-benefit analysis: concepts and practice in Canberra on 4-5 June 2019. 

Society’s resources – people, land, machines, materials and environmental goods – are limited. Using a resource for one project or policy precludes its use for alternatives. If governments are to increase the wellbeing of their citizens, they must be able to select and implement the most socially beneficial projects and policies.

Government funding must therefore be considered in the context of missed opportunities.

At its most confronting, building a safer road may actually kill people because funds were unavailable for critical diagnostic equipment in hospitals. Since the public purse can’t stretch to everything, governments must make judicious choices.

There is no shortage of decision-making approaches. Depending on where one is in the fashion cycle, fuzzy logic, multi-criteria analysis, triple-bottom lines, balanced scorecards, gap analysis, flipping coins, triangulation, and a multitude of others are available on tap from consultants and public service enthusiasts.

But only social cost-benefit analysis provides a rigorous means for advising governments where to direct resources so that they will most benefit the whole community.

Unfortunately, expertise in economic analysis among Australian public servants, both federal and state, is now largely confined to parts of central and transport agencies. This may partly be due to public service emphasis on managerialism in recruitment over recent decades, but universities have also overseen a marked decline in ‘hard’ economics courses in favour of business-type degrees.

The hollowing out of public service expertise is reflected in the increased use of commercial consultants to carry out cost-benefit studies. Satirical ABC series The Hollowmen captures the outcomes well, with public servants struggling to understand which projects are in fact worth implementing. Politicians often call for a cost-benefit analysis, but few, if any, actually understand what it means.

Calls on both sides of politics for increased spending on large infrastructure projects should always be tested against alternative measures in a cost-benefit analysis.

A frequent flaw is the omission of alternative approaches that can feasibly achieve the primary policy objective, resulting in agenda-driven bias. ‘Roading’ evaluations are a standout example. A reduction in road congestion could be achieved through congestion pricing or by subsidising public transport, but the options are typically limited only to constructing a new road or expanding an existing one.

It has become increasingly common to include in the projected benefits of an urban transport project a component termed Wider Economic Impacts, to reflect benefits flowing from the increased effective density of a city when it becomes more accessible to those living outside it. However, the underlying theoretical aspects are by no means settled, and are likely to produce sizeable overestimates.

Another common failing in Australian cost-benefit studies is that of not defining ‘standing’ – the perspective of the society whose benefits and costs are to be included. The Canberra light rail business case, for example, fails to define standing although it implicitly encompasses the Australian Capital Territory (ACT). It includes as a benefit the additional income tax gained due to higher employment generated by the project. However, income tax is a federal tax that cannot be counted as a benefit to the ACT; an obvious error that would not be immediately apparent to most public servants or politicians.

Given the lack of in-house public service expertise, there is a serious risk that Ministers and their Cabinet colleagues will be misinformed when considering the social value of a proposal, or that the full implications will not be realised. Even worse, Ministers may inadvertently mislead Parliament about the likely social costs and benefits of a policy or project.

The best approach to ensuring that economic appraisals of government projects, programs and policies meet minimal professional standards is to specify a broad framework of methodology and values, but without imposing uniformity or standardisation. For example, analysts would be free to choose a value of statistical life, but would need to explain and justify their choice, and subject it to peer review and public comment. Where analysis is limited to a single project like building a road, analysts would be required to justify their exclusion of feasible alternatives. Over time, the standard and consistency in approach of cost benefit analyses would improve, making their preparation and comprehension easier.

Ultimately, Australian governments have a choice. They can continue to ‘nation-build’ by breeding white elephants like the Darwin-Alice Springs railway, mothballed desalination plants, the National Broadband Network in its original configuration, the Clem Jones tunnel in Brisbane, the Ord River scheme, and countless others. Or they can get serious about promoting credible cost-benefit analysis.


Social cost-benefit analysis in Australia and New Zealand: The state of current practice and what needs to be done, by Leo Dobes, Joanne Leung and George Argyrous, was published by ANU Press in April 2016. It is free to download from ANU Press.

This post originally appeared on