The emergence of cost-benefit analysis

Individuals make decisions about the use of their limited resources with the aim of maximising personal happiness. The expectation is that governments do the same but with respect to the well-being[2] of the community as a whole. However, the task of governments is more complex, because they inevitably need to balance the conflicting wishes and wants of many different individuals in society. They also need to take account of spill-overs that one particular section of society may inflict on others.

In particular, governments invariably lack readily-available data that reflect adequately the values placed by the individuals on non-marketed goods and services. Estimation of the value of clean air, the preservation of a threatened species, or the social costs of traffic congestion can represent a challenge in this regard. To assist governments in the process of considering the trade-offs involved in resource-allocation decisions, the economics profession has developed ‘cost-benefit analysis’.

The key underlying principle of cost-benefit analysis is the comparison of social benefits with corresponding costs. If benefits exceed costs, decision-makers have a case for proceeding with a project. If costs exceed benefits, and the decision-maker proceeds with the project, then she or he is at least informed of the net cost that society needs to bear.

Despite popular misconceptions, cost-benefit analysis is not founded on market prices. Benefits to society are measured as an aggregation of individuals’ willingness to pay, and social costs reflect opportunities forgone. An individual’s willingness to pay for water, for example, may be several thousands of dollars a year. But the price actually paid for that water by an individual may be far lower (perhaps just several dollars per year). The difference between the individual’s willingness to pay and the price they pay is defined as the benefit generated from the consumption of the water. Thus price alone does not define the benefit. It is the willingness to pay net of the price. The net social benefit of making water available to all the individuals in a community is the present value of the difference between the sum of the residents’ net willingness to pay (their ‘consumers surplus’) and the sum of the opportunity costs of the resources used in supplying the water. Simply because a good or service is not marketed, and hence is not priced, does not mean that it does not generate a benefit to society. So long as there is a willingness to pay for the non-marketed good or service, there will be a benefit enjoyed.[3]

Cost-benefit analysis thus provides a framework that allows governments to assess and compare the social costs and benefits of the full range of impacts of a proposed action, whether they involve marketed goods, environmental impacts, or regulatory controls. This framework brings with it a lineage extending back to writers such as Dupuit, Marshall, and Pigou, who founded the discipline in the nineteenth century (Mishan & Quah 2007: 243). These economists, and their numerous twentieth-century successors, have assiduously debated and refined the underlying concepts within the broader and coherent analytical construct of welfare economics.

North American textbooks on cost-benefit analysis invariably contain some reference to the 1936 US Flood Control Act. Under this legislation, Congress required flood-control projects undertaken by the US Army Corps of Engineers to be preceded by an analysis of costs and benefits.[4] The landmark Flood Control Act contains the famous phrase that the Federal Government should improve streams for flood-control purposes ‘if the benefits to whomsoever they may accrue are in excess of the estimated costs, and if the lives and social security of people are otherwise adversely affected …’. Subsequent presidential Executive Orders (12291: Reagan 1980 and 12866: Clinton 1993) have stipulated government requirements for the analysis of costs and benefits of regulatory proposals prior to their adoption.

Australia too has some history of the application of rigorous assessment of government projects, although the extent of its use in the past is not clear. But the use of analytical tools to fashion input to the political decision-making process has never been institutionalised to the same degree that has been the case in America.

In giving evidence to a committee of the Victorian colonial parliament discussing a Railways Bill in 1871, for example, the Resident Engineer of the Railway Department provided the committee with an illustration of discounted cash-flow estimates of the costs for alternative projects (Evidence taken at the Bar of the Legislative Council ... 1871, appendices K and L). His example demonstrated that it would be cheaper to build a wooden viaduct that would last for only 10 years and would be rebuilt every decade thereafter, than to build a stone structure with steel girders that would last 100 years. He also claimed to have used the method in 1868 when a wooden bridge was in fact built.

But it was only in 2006 that the Council of Australian Governments agreed that the quality of regulatory impact statements should be improved through the use of cost-benefit analysis, and the Commonwealth Government established the Office of Best Practice Regulation to provide training, advice and technical assistance to government agencies (see Harrison in this issue). However, this requirement for economic analysis has not been extended to proposals that are not specifically regulatory in nature.

Throughout its history, cost-benefit analysis has been subjected to critical scrutiny and attack for being overly focused on economic efficiency. Wildavsky (1966: 310), for example, argued that economic perspectives should not ‘swallow up political rationality’, and Peter Self argued (Coleman & Hagger 2001: 120–6) that government-employed town planners, rather than ‘econocrats’, should determine the character and amenity of new towns. As part of this reaction, according to Lichfield (1993: 206), Italian and Dutch researchers in the 1970s began to work increasingly with multi-criteria analysis, and Quinet (1993: 193) reports the same shift away from cost-benefit analysis in France. In the UK, the then Department of Environment, Transport and the Regions (1998) set out a ‘new approach to appraisal’ of road projects that was based on five principal criteria: environmental impact, safety, economy, accessibility, and integration. Considerable effort has been made over the last decade within European Union countries to reconcile cost-benefit analysis with the multi-criteria analysis approach (for example, Sugden 2005; Diakoulaki & Karangelis 2007; and Prokofieva et al. 2008).

Multi-criteria analysis has also become increasingly popular in Australia (Proctor 2009: 74–5) over the last two decades or so. The Resource Assessment Commission (1992) published an overview of the approach, and the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in particular has actively promoted its use in areas such as natural resources management, climate change and adaptation, and water management

(http://www.csiro.au/science/Social-Economic-Sciences.html).

The Commonwealth Government itself has encouraged a simplistic version of multi-criteria analysis in the form of the so-called Triple Bottom Line[5] approach (for example, Environment Australia 2003) that is popular in some quarters of the Public Service. State and local governments and their agencies have also made use of it. For example, the business case developed by Melbourne Water (2008: 4) to justify the diversion of rural water from the Goulburn river to Melbourne was based on the Triple Bottom Line approach, but the study itself is classified as Cabinet-in-Confidence by the Victorian Government and therefore not publicly available. In an amusingly acerbic review, Ergas (2009) notes that a 2008 report by Infrastructure Australia (the body established to advise Australian governments on the relative merits of potential infrastructure projects) employs the Triple Bottom Line approach, despite also advocating the fundamental importance of cost-benefit analysis to rigorous assessment.

A relatively recent development appears to be the injection of prior political considerations as a prelude to undertaking cost-benefit analysis. Guidelines issued by the National Transport Council (NTC 2006), for example, introduced the concept of a Strategic Merit Test that is to be undertaken as a preliminary step in consideration of transport projects. One objective of the test is to ‘identify how well the initiative is expected to contribute to jurisdictional objectives, policies and strategies’ (NTC 2006 volume 3, ‘Appraisal of initiatives’: 15), with one of the rationales being that it ‘provides an efficient means to filter proposals before considerable resources are spent on development’. The concept appears to have now been adopted by Infrastructure Australia (2008, section 4.2: 66) in the form of the ‘strategic fit’ criterion.

Unfortunately, there has been a marked absence of debate or analysis within Australian government circles and academia as to the validity or relative merits of cost-benefit analysis and multi-criteria analysis. We therefore feel that a more open discussion is both timely and necessary. Because both cost-benefit analysis and multi-criteria analysis can be, and often are, subjected to misuse and erroneous application by practitioners, the focus below is on methodology rather than irrelevant straw men.

The first two sections below outline briefly the key features of both cost-benefit analysis and multi-criteria analysis. These are followed by a section that compares key features of the methodologies used, including various misconceptions about them. A penultimate section explores some of the implications for government. Finally, some policy conclusions are drawn in the hope of generating debate among Australian policy analysts and decision-makers about the appropriateness of using multi-criteria analysis.




[2] The terms ‘happiness’ and ‘well-being’ are used here as non-technical synonyms with ‘utility’. In his classic work on welfare economics, Little (1950: 7) discusses the interchangeability of these terms, as well as others such as ‘social welfare’, ‘the happiness of society’, etc, in the context of utilitarian principles.

[3] Markets provide information on consumers’ willingness to pay where goods and services are bought and sold. Where no markets form, as is the case for public goods, economists have developed alternative ‘non-market valuation’ techniques to estimate willingness to pay (see Hanley & Spash 1993).

[4] The antecedents of attempts in the United States to establish an assessment methodology for government projects are probably older. For example, Reuss (1922: 105) cites the 1808 Gallatin report as demonstrating that Congress generally ‘supported public works whose benefits contributed an “annual additional income to the nation” ’.

[5] Sometimes colloquially referred to as People, Planet, Profit (social, environmental and financial aspects), the Triple Bottom Line approach purportedly provides a comprehensive assessment of all aspects of interest to commercial or government decision-makers. The three aspects are variously given numerical scores, allocated ‘traffic light’ (green, yellow, red) categories for risk or importance, or just discussed qualitatively. There is no underlying principle or methodology involved, and the data used invariably depend on what happens to be readily available to the proponent. While this approach may have made some sense in a private-sector context (although not necessarily for shareholders) seeking to broaden the traditional focus on profits alone to a wider social perspective, it is superfluous in a cost-benefit analysis undertaken from a national social perspective by a government agency. In particular, the economic component of a cost-benefit assessment would already include social and environmental effects so that their separate presentation would constitute ‘double-counting’. Common misperceptions that equate ‘economic’ with ‘financial’ aspects undoubtedly contribute to the confusion involved in the Triple Bottom Line approach.