Table of Contents
In this paper I compare cost-benefit analysis (CBA) to multi-criteria analysis (MCA). I review the nature of the two approaches and consider the criticisms that have been made of CBA. I conclude that these criticisms largely lack merit, and that even to the extent to which they are meritorious, they provide no justification for relying on MCA. I conclude by expressing my concerns about the growing role of MCA in Australian project appraisal, which is symptomatic of a broader move away from sound policy evaluation.
Cost-benefit analysis (CBA) is a technique for evaluating collective decisions that hinges on the comparisons of the costs of a proposal to its benefits, where costs and benefits are valued in monetary terms. In essence (and abstracting from the relevant technicalities), cost-benefit analysis asks whether the sum of the amounts the individuals who comprise the community at issue would be willing to pay for the project to proceed exceeds the costs of that project. Generally, a project enhances wealth — in the sense of the aggregate monetary valuation of the community’s resources — if it meets a properly specified cost-benefit test. Whether enhancing wealth in this sense is either necessary or sufficient for a project to be worthwhile is a complex issue. Without going into the details of that discussion, it seems reasonable to suggest that projects that fail properly specified cost-benefit tests should be looked at very carefully, and be found to have other, significant, redeeming features, before they are allowed to proceed. By the same token, if a project has benefits that (evaluated in monetary terms) clearly exceed its costs, it seems reasonable to presume that, in the absence of compelling reasons to the contrary, society would gain were it to proceed.
Multi-criteria analysis (MCA), rather than seeking an overall monetary valuation of project effects, identifies salient elements of those effects (be they costs or benefits), scales them and then places subjective weights upon them. In that sense, it is a technique for scoring project attributes. The project is expected to be approved if the weighted sum of the desirable effects (that is, the scaled value of the various dimensions of project benefits) exceeds the weighted sum of the project’s undesirable effects (the scaled value of its costs and harmful outcomes). Elsewhere in this issue, Dobes and Bennett provide a worked, hypothetical, example of MCA as it is typically implemented in Australian policy evaluation.
[1] Concept Economics, henryergas@concepteconomics.com.au. I am very grateful to Alex Robson, Eric Ralph, an anonymous referee, Jonathan Pincus, Mark Harvey, William Coleman and, especially, Mark Harrison for many, very helpful comments and discussions. Responsibility for the views expressed here, and for any remaining errors, is of course entirely my own.