Cost Benefit Analysis (CBA)

Primary Source: Armines, A. R.. (2006) Cost Benefit Analysis Swat Evaluation in Report on the SWOT analysis of concepts, methods, and models potentially supporting LCA. Eds. Schepelmann, Ritthoff & Santman (Wuppertal Institute for Climate and Energy) & Jeswani and Azapagic (University of Manchester), pp 38-45

Level of analysis: Micro, meso and macro

Assessed aspects of sustainability:  environmental, economic plus possibly other external costs (time spent by consumers, space required in household, etc.)

Main purpose of the assessment:

Compare costs and benefits of a proposed action (e.g. environmental directive, choice of technology, environmental policy, …)

Description of the methodology:

Description of the methodology: evaluate all the significant costs and benefits entailed by the proposed action. Both direct and indirect or induced effects need to be taken into account. For many choices this involves evaluation of upstream and downstream impacts, using LCA. But by contrast to most conventional LCA, it includes monetary valuation of the impacts. A typical and very important example is the CBA of the CAFE program (Clean Air for Europe) [Holland et al 2005; EC 2006]. Similar CBAs have been carried out in the USA, for example the study by Abt [2000] for the EPA.

Detailed description

Whereas the basic principle of CBA is straightforward, the practical implementation can be extremely difficult. Countless examples of CBA can be found in the literature, ranging in complexity from a simple comparison of two technology choices (e.g. an evaluation of stricter limits for the emission of PM10 from cement kilns that use waste as fuel [Rabl 2000]) to the evaluation of an entire policy (e.g. the CAFE program (Clean Air for Europe) [Holland et al 2005]).

For environmental CBA a crucial tool is the analysis of impact pathways, i.e. of the chain emission – dispersion – impact – cost. For a more complete presentation we refer to the reports of the ExternE Project or to review papers [ExternE 2005, Rabl and Spadaro 2000,]. For many applications upstream and downstream impacts also need to be evaluated, using the inventory phase of LCA.

Over the years, numerous dispersion models have been developed. Usually separate models are used for the local and the regional domains. In the local domain, up to about 50 km from the source, pollutant deposition and aerosol formation by chemical transformation are relatively insignificant and concentrations are influenced primarily by meteorological parameters, such as wind speed and wind direction. Beyond 50 km, one must account for removal of the pollutant from the air by deposition, both dry and wet; for some pollutants chemical reactions must also be taken into account.

For some pollutants the impact pathway involves not only atmospheric dispersion and chemistry but also the subsequent passage of the pollutant through soil, water and the food chain. Here, too numerous models are available [e.g. the RiskPoll model of Spadaro and Rabl 2004, the CalTox model of McKone TE and KG Enoch 2002].

Impacts are quantified using dose-response functions, also known as exposure-effect, exposure-response or concentration-response functions (CRF) in the case of air pollutants. They relate the pollutant concentration to the resulting impact on a receptor (human health, crop, etc.).  Impacts on human health include asthma attacks, hospital admissions, chronic bronchitis, restricted activity days, cancers and mortality. ExternE calculates mortality impacts of air pollution as a reduction in life expectancy, expressed as Years Of Life Lost (YOLL), rather than a number of premature deaths. That is necessary to allow more meaningful comparisons with other causes of death, for instance accidents for which the YOLL per death are much higher than for air pollution (of course such comparisons are not perfect, since the affected individuals differ in age and health status).

For health impacts, the CRFs are derived from a survey of epidemiological studies. In view of the available epidemiological evidence, it is reasonable to assume that the CRFs are approximately linear, without threshold, for the air pollutants of greatest concern, especially particulate matter (PM). Of course, epidemiology is very uncertain at low doses or concentrations and the linear model may not be correct; however, there is no clear evidence why other models would be better. Also, if there is a threshold below current concentrations it has no effect on the calculation of incremental damage costs (i.e. for changes relative to current conditions, which is what is reported in external cost studies and here). Linearity is also commonly assumed for carcinogens.

For crops and building materials, the CRFs have non-linear shapes. For agricultural crops there is even the possibility of a small benefit (fertilizer effect) when the background concentrations of SO2 and NOx are sufficiently low. For crops, one calculates the losses or gains in yield, and for building materials, the increase in cleaning and repair costs due to air pollution.

Monetization is used for aggregating health impacts and environmental burdens with different physical units into a single damage indicator, and it is crucial for CBA. To obtain the damage costs, one multiplies the number of impacts (e.g. cases of asthma attack) by the unit cost per impact (e.g. € per asthma attack). Monetary values also have the great advantage of translating the impacts into the language of the economy; thus they are directly usable for cost-benefit analysis.

For health impacts, the unit costs include the cost of treatment and wage and productivity losses, which are market based, as well as non-market costs that take into account an individual’s Willingness-to-Pay (WTP) to avoid the risk of pain, suffering or loss, or the Willingness-to-Accept (WTA) this risk for an economic compensation. Non-market costs may contain other components, such as altruistic value, bequest value and existence value (e.g. the value of a cultural monument) [see, e.g., Skovgaard et al. 2007]. If the WTP for a non-market good has been determined correctly, it is like a price, consistent with prices paid for market goods. Economists have developed several techniques for valuing non-market goods. In recent years, contingent valuation has become the method of choice; it obtains WTP estimates by asking individuals how much they are willing to pay to achieve a benefit [Mitchell and Carson

1989]. This method is problematic and the uncertainties are large, but for many environmental non-market costs no better alternative is available. Depending on the nature of the good, other techniques for monetization such as hedonic pricing and revealed preferences may also be used or even preferred.



A major strength of CBA is the comprehensive scope and the effective aggregation of results into a single unit that can be easily communicated. If a CBA analysis has been carried out with care, clearly stating the underlying assumptions, it brings hidden consequences into the open and helps focus the debate on the facts. In particular it can indicate whether a proposed decision reflects true preferences or merely a cognitive illusion.


This list of weaknesses/limitations/objections may appear discouragingly long. However, most of the objections that have been raised concern studies that were poorly done or applied without proper care (see the discussion of Threats); the weaknesses and objections can be avoided if the CBA is well done and followed by appropriate utilisation.

Of particular concern are the notoriously large uncertainties, especially of the benefits but often also of the costs [Spadaro and Rabl 2008]. However, they do not render a cost-benefit analysis useless. As illustrated by the examples in Rabl, Spadaro and van der Zwaan [2005], the conclusions of a CBA can often be remarkably robust, despite very large uncertainties. Some environmental decisions involve considerations that defy quantification in monetary terms; such decisions should not be reduced to simple CBA. But even in such cases one should try to quantify as much as possible, treating the non-quantified considerations by multicriteria analysis with the involvement of stakeholders.

Sometimes one finds conflicting results between different studies because of different assumptions or methodologies used by different authors. Often such differences are merely a reflection of the current uncertainties of the underlying science [Spadaro and Rabl 2008]. A good study indicates clearly which choices have been made, to allow detailed comparison with others. In certain studies funded by advocacy groups such as industrial or environmental lobbies choices have been made to obtain a desired result rather than in the spirit of objective inquiry. Biased studies can even resort to such disreputable practices as presenting the results only in terms of a benefit/cost ratio where the denominator has been artificially reduced by counting damage costs on the benefit side as negative benefits.

The fact that sometimes different CBAs come up with conflicting results for the same issue is really not a weakness of CBA in general but can arise in particular situations because of different assumptions or methodologies used by different authors. In many case such differences are reflections of the uncertainties at the current state of scientific knowledge, but sometimes they are due to ideological biases (e.g. studies by environmental advocacy groups versus studies by industry).

Several objections to CBA have been raised (e.g. Pearce 2001; Skovgaard et al. 2007):


The costs and benefits of an option, and their monetary value, are often highly uncertain. Wynne (1992) distinguishes four types of uncertainty; risk, uncertainty, ignorance and indeterminacy. If the possible outcomes can be defined and their probabilities can be assigned in a meaningful way, one is talking of risks. If the possible outcomes are identifiable, but their probabilities cannot be determined, one is faced with uncertainty. Ignorance refers to when we do not know what we do not know. Finally, indeterminacy is used to describe situations in which the complexity of the system is so large and so little is known about the relevant parameters and their relationships that modelling becomes a matter of hit and miss (Mickwitz 2003). Where ignorance and indeterminacy may be at play, and it will often be the case because of the complexity of social and environmental issues, decision-making will have to rely on other tools in addition to the CBA. A CBA can account for risk, and the sensitivity analysis can, in principle, deal with uncertainty. However, if the full uncertainty is properly accounted for, the CBA results might encompass a level of uncertainty that makes them difficult to interpret and use. On the other hand, if the uncertainty is not properly accounted for, the study lacks in credibility.

Moral objections:

CBA is based on utilitarian moral philosophy, which means it is based on the assumption that all types of negative effects can be compensated by positive effects. It can be argued that certain negative effects, e.g., the loss of human life or the extinction of a species, cannot be compensated for by positive effects. Furthermore, individuals that benefit from a policy or project typically do not, in practice, compensate the individuals that lose. As a result, the CBA should be complemented by an identification of negative (and positive) effects that are difficult to compensate (or offset) by other effects; and by an analysis of the distribution of positive and negative effects for various groups in society.

The efficiency focus:

The objective of the CBA is to assess how efficient efficiency options are when they are implemented in the current economic, technological and social context. Policy-makers, however, often have additional objectives such as fairness, equity, longterm sustainability, competitiveness, employment, regional balance, etc. A full basis for a decision might require additional analyses to cover these issues.


Decision-makers may feel that CBA results, by indicating the most efficient option, usurp the freedom of choice from the decision-makers. Here, it is important to remember that the CBA is a decision support tool, and that other effects or political considerations may not be encompassed in the CBA.


The CBA has been accused of not involving relevant stakeholders. By presenting one-dimensional results there is a risk that it closes the door for debate. Stakeholder participation and debates are important to resolve conflicts of interest. Without them, important stakeholder groups might not accept the option selected by the decision-makers. This can be a significant problem for controversial options such as the construction of a waste incinerator or an expansion of the source separation scheme. Since CBA does not resolve conflicts of interest, it cannot replace the decision process but only provide input to this process. However, although Nordic CBAs on waste management do not resolve conflicts of interest, they have highlighted some of them. Early CBAs on recycling highlighted the conflict between the benefits (environmental improvement) and costs (particularly the time spent on source separation). Subsequent CBAs on collection systems highlighted the conflict between the benefits of kerbside collection (less time required from households) and bring systems (less effective, lower collection cost). In this way, CBAs can contribute to an informed debate. The debate can be stimulated during the CBA by involving a steering committee or reference group that includes important stakeholders. The completed CBA report can stimulate further debate, if it highlights important methodological choices and data uncertainties.



Expertise in both economics and natural science is necessary to perform a CBA. Moreover, a certain level of expertise is also necessary among the people using or judging the results, for example to participate in a debate that is based on CBA results. If stakeholders are involved in the steering committee, they get the opportunity to learn about the individual study and about CBA in general. This will make them more able to participate in any debates spurred by the completed CBA.

Hence, CBA does have important limitations. Most of them can be overcome or alleviated through a few careful measures:

–          start by generating and screening ideas for relevant options,

–          involve a steering committee, to achieve learning and acceptance,

–          write a transparent report highlighting important methodological choices and uncertainties, and

–          carry out or recommend complementary analyses to achieve an improved basis for decisions.

Opportunities for broadening and deepening LCA

CBA can provide crucial input for policy-making, especially for the increasingly important sector of environmental protection, for example:

–          Guidance for environmental regulations (for example, determining the optimal level of the limit for the emission of a pollutant);

–          Finding the socially optimal level of a pollution tax;

–          Identifying technologies with the lowest social cost (for example, coal, natural gas or nuclear for the production of electricity?);

–          Evaluating the benefits of improving the pollution abatement of an existing installation such as a waste incinerator;

Whereas in the past environmental CBA has been controversial and most environmental policies were implemented without CBA, the situation has been changing: environmental CBA has become respectable and is increasingly required by many government agencies in Europe and North America. A very important example of a recent application is the cost-benefit analysis of the CAFE (Clean Air for Europe) program of the European Commission DG Environment [EC 2006].

Threats for broadening and deepening LCA

There are three main risks:

1)       The risk that the method is abused, which will hurt its reputation and make it less likely to be used in the long run. Abuse includes, e.g., utilising the large uncertainties in the method to stop good decisions, systematically making assumptions to obtain results that fits to defend decisions already made, special interests and/or products. It can also include using the aggregation to hide tradeoffs that cannot otherwise be easily defended (see weaknesses).

2)       The misconception that the differences one sometimes finds between results of different CBA studies of the same question mean that CBA is inappropriate. In reality such differences are due to different assumptions and methodologies used by particular studies. However, a good CBA will document its assumptions and methodology in sufficient detail to allow verification of its results and comparison with similar studies. In many cases good justifications can be found for different assumptions because the underlying science is uncertain.

3)       The misconception that the uncertainties are so large as to render the results meaningless. In reality a well-conducted CBA is better than the infinite uncertainty in the absence of such analysis. Furthermore, as shown by Rabl, Spadaro and van der Zwaan [2005], the uncertainties are sufficiently small to help avoid erroneous policy choices that would be very costly.

Literature/Internet links

Abt 2000. “The Particulate-Related Health Benefits of Reducing Power Plant Emissions.” October 2000. Prepared for EPA by Abt Associates Inc., 4800 Montgomery Lane, Bethesda, MD 20814-5341.

EC 2006. The CAFE Programme (Clean Air for Europe). European Commission DG Environment.

ExternE 2005. ExternE – Externalities Of Energy: Methodology 2005 Update. Available at

Holland, M., Watkiss, P., Pye, S., de Oliveira, A., van Regemorter, D. 2005. Cost benefit analysis of Policy Option Scenarios for the Clean Air For Europe Programme. .

McKone TE & KG Enoch 2002. “CalTOX™, A Multimedia Total Exposure Model”. Report LBNL – 47399. Lawrence Berkeley National Laboratory, Berkeley, CA. Available at

Mitchell, R.C. & R.T. Carson 1989. Using Surveys to Value Public Goods: the Contingent Valuation Method. Resources for the Future. Washington, DC.

Pearce D 2001. “Integrating cost-benefit analysis into the policy process, annex II in Valuing the benefits of environmental policy”. RIVM report 481505 024, Blithoven.

Rabl A 2000. “Criteria for limits on the emission of dust from cement kilns that burn waste as fuel”. ARMINES/Ecole des Mines de Paris, Paris. March 2000. 10 pp.

Rabl A, J. V. Spadaro & B. van der Zwaan 2005. “Uncertainty of Pollution Damage Cost Estimates: to What Extent does it Matter?”. Environmental Science & Technology, vol.39(2), 399-408 (2005).

Skovgaard M, K Ibenholt & T Ekvall (2007) Nordic guideline for cost-benefit analysis in waste management. TemaNord 2007:574. Nordic Council of Ministers, Copenhagen. /TE

Spadaro JV & A Rabl 2004. “Pathway Analysis for Population-Total Health Impacts of Toxic Metal Emissions”. Risk Analysis, vol.24(5), 1121-1141. The RiskPoll model can be downloaded from

Spadaro JV and Rabl A 2008. “Estimating the Uncertainty of Damage Costs of Pollution: a Simple Transparent Method and Typical Results”. Environmental Impact Assessment Review, vol. 28 (2), 166–183.



Register| Forgot Password?

Policy Tools Wiki