![]()
Part I
Basics
![]()
1 Introduction
Aim of this Book
In this book I discuss ethical and political controversies related to environmental cost-benefit analysis: why controversies must be expected, why these controversies need to be taken seriously, and how they can be handled in practice. While my focus is on evaluation of projects with environmental impacts, substantial parts of the discussion are relevant for other projects as well.
Most textbooks in cost-benefit analysis assume, implicitly or explicitly, that the purpose of the analysis is to rank projects according to their social preferability. My starting point is different. I assume, instead, that the primary role of an economic project analysis is to provide factual input to a democratic decision-making process. I will assume that several decision-makers, whose ethical and/or political views differ, take part in this process. Consequently, the purpose of the analysis is not to determine one final ranking of projects, but rather to enable each decision-maker to arrive at a well-founded ranking in accordance with her own normative views.
As it turns out, the latter purpose faces the analyst with quite different challenges from the former. It is thus important to be aware of the distinction between these two purposes of project analysis, and to adjust oneâs methodological choices, interpretation of results, and form of presentation accordingly.
A main reason why a project analysis should be approached differently depending on its purpose has to do with the fundamental distinction between normative and positive analysis. Ranking projects according to their social desirability is an explicitly normative task. Pure logic demands that normative conclusions must be based on normative premises. Thus, no analysis performing this task can claim objectivity. If one aims at providing background information to a democratic process, however, the analysis itself is not necessarily normative. With this purpose, the concern for objectivity may indeed be crucial: If a decision-maker finds that the analysis is based on normative views she does not share, she will (and should) be critical of its results. If, moreover, she does not know how to adjust results in order to account for the normative disagreement, she may have to disregard the analysis altogether. Consequently, even if decision-makers have the best intentions, the analysis may end up having little or no influence.
One problem with applied cost-benefit analysis is precisely that it can be hard or even impossible for users to adjust its results to account for normative disagreement. This problem might have been a minor one if the normative assumptions employed were uncontroversial; however, as I will discuss below, this cannot be expected to be the case.
On the other hand, no analysis can be perfectly objective. There are many reasons for this, but one is the need for aggregation and simplification. Decision-makers with different normative views typically require different information to arrive at their own well-founded ranking of projects. Thus, if normative views differ, and each decision-maker is to be given the same information, the total amount of information may simply become too extensive. Busy decision-makers do not mainly need details; they need overview and understanding. In practice, there will be a trade-off between objectivity â requiring possibly very disaggregated information â and simplicity and brevity, requiring aggregation. The emphasis placed on each of these concerns should depend on the purpose of the analysis.
When making these trade-offs, the analyst will to a considerable extent need to resort to his own subjective judgment. In a normative analysis, such judgment is facilitated by the possibility of resorting to the normative premises already chosen for the analysis. In a positive analysis, the required judgment on the analystâs part may be more difficult â which information to report or suppress, the level of aggregation, whether to use possibly controversial indicators, and, if so, which. I hope this book, by pointing out and perhaps clarifying some of the factors to be considered, can be of assistance when making those judgments.
In applied project analysis, the most pressing difficulties may well be related to basic issues such as factual project description, clarification of projectsâ purpose, or defining the base case scenario (the âstatus quoâ situation to compare projects with). These matters are not, however, my concern here, and I will thus simply assume that they have already been sorted out. Instead, I will concentrate my discussion on issues of valuation, aggregation, interpretation, and presentation; for example how far one should go in the direction of valuing all consequences in monetary terms.
Several countries have guidelines requiring cost-benefit analysis or similar analyses for new project or reform proposals. Nevertheless, some research has indicated that the economic analysis tends to be poorly integrated in the political decision-making process, and that politicians hardly, or to a rather limited extent, prioritize in accordance with the recommendations from these analyses.1 While this might of course be driven by policy-makers who care less about social welfare than re-election and personal power, it could also possibly be related to the problem pointed out above: even with the best intentions on policy-makersâ part, analyses based on controversial normative premises may be poorly suited as input into a democratic decision-making process. Since it may be interesting for the reader to hear what policy-makers themselves have to say about this, I have included here some results from an interview survey among members of the Norwegian parliament concerning their use of and attitudes to cost-benefit analysis.2
My reasoning below will be based on economic theory, but with a strong emphasis on readability and applicability. It is aimed at those â economists and non-economists alike â who use or are faced with cost-benefit analysis and environmental valuation in their work: politicians, employees of ministries and regulatory agencies, students, journalists, consultants, and researchers. No particular prior knowledge of economics is required, although familiarity with a few very basic and standard concepts of economics will facilitate the reading. Also, those who have previously enjoyed, despaired over or conducted a cost-benefit analysis will of course have an advantage. A couple of chapters include slightly more technical material, which I believe to be clarifying for some readers; these parts are clearly marked, and can be skipped without missing the main ideas.
I should stress that this is not a do-it-yourself handbook of how to conduct cost-benefit analysis in practice. Nor is my purpose to survey the field or to provide new research insights. Rather, I wish to present and sum up a hopefully applicable and coherent approach to environmental cost-benefit analysis, developed through 20 years of my own research, discussions with other researchers, practitioners and decision-makers, and through my participation in several government-appointed expert committees concerned with designing principles and guidelines for public project evaluation. I hope that the book will stimulate debate, but also be helpful to those who are faced with environmental cost-benefit analysis in their everyday work.
Cost-Benefit Analysis: Background
There exists by now a huge literature on cost-benefit analysis in general, and environmental cost-benefit analysis in particular, concerned with theoretical foundations as well as practical application. I have no intention of competing with these books and papers, since my interest lies elsewhere. Those interested in learning more about the method in general are referred to the existing literature.3 A very brief summary of some fundamental aspects of cost-benefit methodology may still be in place, though.
Cost-benefit analysis (CBA) is a frequently used method for evaluation of public project or reform proposals. The methodology is suitable only for marginal projects; âmarginalâ in this context means that the project can be regarded as small enough not to influence equilibrium prices or other marginal values. If the project is not marginal, a less partial method, such as computable general equilibrium modeling, is needed. Note that while this book focuses on marginal projects, some researchers use the term âcost-benefit analysisâ more generally, including even studies of profound non-marginal changes such as whether or not to combat global warming.
In a CBA, each projectâs consequences are valued in monetary terms, and then aggregated to yield an indicator of the projectâs net impact on social welfare. The main principle for the valuation is that a consequence is worth what the population is, in total, willing to pay for it.
For valuation of goods traded in markets, market prices provide the starting point for a CBA. Market prices must often, however, be adjusted due to market imperfections, taxes, and regulations. To value environmental and non-market goods, the relevant populationâs willingness to pay (WTP) needs to be estimated, since no market prices exist.4
Several methods are available for non-market valuation. With direct methods, one simply asks people about their willingness to pay. Indirect methods explore, instead, the relationship between non-market and market goods to estimate consumersâ valuation of the latter. For example, the extent to which property prices vary with outdoor noise levels may provide an indication of consumersâ valuation of noise reductions. Again, other scholars have treated these matters thoroughly and skillfully, and the reader is referred to the existing literature for better and more elaborate explanations.5
With a few notable exceptions, I will largely disregard the problems associated with eliciting willingness to pay. I am mainly concerned with questions such as: even if we did, in fact, know individualsâ willingness to pay, what would those numbers mean in a normative sense?6
In cost-benefit analysis, consequences occurring at different points in time are made comparable through discounting, yielding estimates of present values. Since discounting determines the weight placed on future generationsâ interests versus those of the present, this is an issue associated with substantial controversy, and many discussions on ethics and CBA are primarily concerned with discounting practices. Nevertheless, although the problems treated below are important in dynamic as well as static contexts, these problems are not specifically related to the time dimension. To express my arguments as simply and straightforwardly as possible, I am going to disregard the fact that costs and benefits occur at different times, and ignore discounting.7 In the same vein, I will not distinguish between an individualâs income and his wealth, but use the term âincomeâ as a measure of the total purchase of private goods an individual can afford.
When all benefits, valued and discounted according to the above-mentioned principles, are added, and all costs, valued and discounted by the same principles, are subtracted, one gets the projectâs net present value, or aggregate discounted net willingness to pay. This measure is often used as an indicator of the projectâs net contribution to social welfare. Since I will disregard discounting in this book, net present value and aggregate net willingness to pay will here have the same meaning; however, to abstract from the time dimension altogether, I will stick to the term aggregate net willingness to pay.8
CBA methodology is based on economic theory. Nevertheless, it should be noted at the outset that CBA takes a substantial step beyond standard neoclassical economic theory when it comes to measurement of utility and welfare. Neoclassical consumer theory does not suggest any method for measuring utility in a way that can be compared between individuals; nor does standard consumer theory require that such comparison is possible. Normative interpretation of CBA, however, is based on measuring and comparing welfare changes between individuals.9
Economists have long been aware of the difficulties in measuring and comparing utility in CBA. In most CBA textbooks these issues are clearly pointed out, but the most common conclusion still seems to be that CBA can still be used normatively â although with care. The main problem with this view is, as I see it, that it is usually not clear what it means to use care.
![]()
2 The Purpose of a Project Analysis
Textbooks on CBA usually assume, implicitly or explicitly, that the purpose of the analysis is to answer the question of which public project(s) should be implemented, based on projectsâ contributions to efficiency or social welfare. For later reference, let us call this purpose A: ranking.
However, a final answer to which projects should be given priority is not necessarily what bureaucrats and politicians primarily look for when considering which policies to support. After all, even when the economic analysis is summarized by a specific recommendation, the final decision is usually not determined by that recommendation. The final decision is taken in the course of some administrative and/or political process to follow, in which the economic analysis is only one of several pieces of background information. Usually, policy decisions are made on the basis of facts as well as subjective deliberation on the decision-makersâ part, and actual decisions are rarely delegated to economic analysis as such. Indeed, choosing costly public projects based on an economic analysis only would presumably conflict with widely acknowledged principles for democratic decision-making.1
The main contribution of a project analysis, then, may not be to provide a final ranking, but rather to make the actual decision-makers sufficiently informed to make their own well-founded judgments about projectsâ social desirability. This latter task is quite distinct from the former one â if participants in the decision-making process hold different ethical and/or political views. An important requirement for democratic decision-making is that each participant in the process should have âequal and effective opportunities for learning about the relevant alternative policies and their likely consequencesâ (Dahl 2006, p. 9). A ranking is a conclusion about which projects should be preferred. What is needed in a democratic process with a diversity of normative views are factual foundations for a conclusion â not the conclusion itself. Purpose B, thus, can be formulated as providing informational input into a democratic decision-making process. Although the distinction between the two purposes is rarely discussed in the economic literature on CBA, they do face the analyst with quite different challenges.2,3
Purpose A: Ranking
To rank projects it is, first of all, necessary to choose normative premises for the analysis. Also, all consequences must somehow be made comparable.
The normative premise invoked in most of economic theory is the Pareto principle. This principle says that if a change leaves at least one individual strictly better off, while leaving no individual strictly worse off, the change is socially preferable. However, this criterion tells us only how to rank projects where one is a Pareto improvement compared to the other; it does not tell us how to rank projects involving conflict of interest, no matter how small or unimportant that conflict may be. Most public pr...
