NOU 2012: 16

Cost-Benefit Analysis

To table of content

2 Cost-benefit analysis – Main features

2.1 Introduction

The present Green Paper supplements the NOU 1997: 27 Green Paper and the NOU 1998: 16 Green Paper. The present Chapter briefly explains the basic cost-benefit analysis approach and principles relating to the valuation of cost and benefit elements that remain fixed and have not been subjected to renewed discussion by the Committee. The Chapter is intended to offer readers an overview of cost-benefit analysis as a topic, but does not provide a complete discussion of existing recommendations. Reference is made to the discussions in the NOU 1997: 27 Green Paper; Cost-Benefit Analysis, the NOU 1998: 16 Green Paper; Guidance on Using Profitability Assessments in the Public Sector, and the Ministry of Finance cost-benefit analysis guide from 2005 for a more comprehensive review of such issues. Reference is also made to the NOU 2009: 16 Green Paper; Global Environmental Challenges – Norwegian Policy, which examines environmental goods in a cost-benefit analysis context. In the subsequent chapters, the Committee will discuss issues that merit revisiting in view of recent theoretical developments, cf. the terms of reference of the Committee.

2.2 Main forms of cost-benefit analysis

We distinguish between various forms of cost-benefit analysis. Cost-benefit analysis is used as a common term for the three different alternative analysis methods: cost-benefit analysis, cost-effectiveness analysis and cost-effect analysis (cf. Chapter 1, footnote 1). In cases where it is important to distinguish between the different methods of analysis it is specified which method is used

Cost-benefit analysis, as a specific method of analysis, involves the valuation, to the extent feasible, of all positive and negative effects of a measure, based on the fundamental principle that the value of a consequence is what the population as a whole would be willing to pay to achieve it. If the willingness to pay for all benefits emanating from the measure exceeds its total costs, such measure is defined to be economically profitable (NOU 2009: 16 Green Paper). In principle, the cost of a project shall reflect of what one must forgo to implement such project, whilst its benefits shall reflect how much one is willing to give up (NOU 1997: 27 Green Paper).

If the various measures to be compared entail the same benefits, or benefits that can be measured on a uniform scale, it is not necessary to attach a monetary value to such benefits for purposes of ranking the economic profitability of the said measures. The ranking will in any case depend on the costs of the measure only. A cost-effectiveness analysis involves the ranking of measures on the basis of costs, in order to identify the measure that will realise a given objective at the minimum cost.

In some contexts, it will be difficult or undesirable to attach a monetary value to the benefits, although the benefits resulting from various measures differ from each other. One may in such case calculate the costs of the measures in the usual manner, whilst describing the resulting benefits in the best possible way, but not necessarily in monetary terms or on a uniform scale. This is termed cost-effect analysis (NOU 1998: 16 Green Paper). Cost-effect analyses do not enable the measures to be ranked on the basis of economic profitability, but nevertheless provide decision makers with valuable information.

2.3 The purpose of cost-benefit analysis

The main purpose of cost-benefit analysis is to identify and elucidate the consequences of alternative measures prior to making a decision on which measure to implement. Consequently, cost-benefit analysis is a method for organising information (NOU 1998: 16 Green Paper). The analysis shall provide part of the basis for making decisions, without thereby amounting to a decision-making rule.

Cost-benefit analysis and cost-effectiveness analysis, as specific methods of analysis, enable the uniform ranking of measures based on their economic profitability (defined as in Chapter 2.2 above). However, economic profitability is not necessarily the only consideration of relevance to decision makers. In addition to focusing on the economic profitability of measures, the analysis should also aim at describing all consequences assumed to be of importance to the assessment to be made by decision makers, including non-priced effects and the distributional implications of measures. The final discretionary assessment should then be left to decision makers. The handling of distributional considerations in cost-benefit analysis is examined in more detail in Chapter 4.5 of the NOU 1997: 27 Green Paper and is discussed in Chapter 3 of the present Green Paper.

2.4 Valuation principles

A monetary value is attributed to costs and benefits to the extent theoretically justifiable and expedient. Whilst business analysis would use market prices in the valuation of a measure, cost-benefit analysis uses calculation prices. These shall reflect the value of the resources or inputs devoted to the measure used under the best alternative option. In perfectly competitive markets and with no direct or indirect taxes, the calculation prices are given by the market prices. In practice, market prices have to be adjusted for direct and indirect taxes, as well as, if possible, for various forms of market failure; see Chapter 2.5. We have no prices in certain areas of relevance to cost-benefit analysis, such as for many health and environmental goods. These issues are discussed in more detail in Chapters 5 and 6 of the NOU 1997: 27 Green Paper and are summarised in Chapter 2.6.

The cost and benefit implications of a measure often materialise at different points in time. The net present value method is a calculation method that enables the comparison of benefits and/or costs which occur at different points in time. Estimated costs and benefits are discounted to a specific reference year by applying a discount rate. By adding together the discounted future benefits and costs, we then arrive at the net present value of a measure. The measure is defined to be economically profitable if the net present value is positive. Hence, the discount rate may be interpreted as a required rate of return. Principles for determining the discount rate are discussed in more detail in Chapter 5 of the present Green Paper.

2.5 Rules for determining optimal calculation prices

This section summarises the rules that may be applied in the pricing of market goods for cost-benefit analysis purposes. We refer to the NOU 1997: 27 Green Paper and the NOU 1998: 16 Green Paper for thorough discussion of each rule. The present summary1 largely mirrors the discussion in Chapter 7 of the NOU 2009: 16 Green Paper.

Tax effects

Most goods and services are subject to taxation, and hence there is almost always a need for examining how to deal with taxes in a cost-benefit analysis context. The underlying principle is that the resources devoted to a project shall, to the maximum extent possible, be assessed at their value under the best alternative option. It can be shown that if the authorities have sufficient tax instruments available in an economy with free competition, the inputs considered in a cost-benefit analysis shall be valued in the same manner as they would be valued by a private enterprise.2 Such use of producer prices means that labour is valued at the market wage before tax (inclusive of social costs), whilst goods and services that are subject to a value added tax are valued at the market price exclusive of the value added tax.3

Externalities

Externalities arise if the activities of a person or an enterprise have a direct impact on the welfare of other persons or the profitability of other enterprises, which impact is not conveyed through market prices, for example the pollution of a river used by a number of users. It is typically the case that many persons and enterprises are exposed to the externalities caused by each individual user. This makes it difficult for the polluter and those affected to reach agreement on compensation.4

In order for such negative externalities to be taken into account, one may introduce a tax reflecting the value of the marginal detrimental effect, or alternatively use tradable quotas. Taxes and quotas as policy instruments in the regulation of greenhouse gas emissions are discussed in more detail in Chapter 9. In other cases one may consider using prohibition or other quantitative restrictions to deal with an externality, for example if it is administratively challenging to establish taxes or tradable quotas, or when minor emission changes have major environmental impacts, cf. Weitzman (1974). These issues are discussed in more detail in the NOU 2009: 16 Green Paper.

Some forms of economic activity may also entail positive externalities. A standard example is research and development activities. Such activities on the part of an enterprise may, for example, generate knowledge that can also be used by other enterprises over time, or give rise to new products with a higher overall value in the market than can be captured by the enterprise itself through the product price.5 Positive externalities in production or consumption constitute an argument in favour of public support for research and development activities, as well as other activities involving corresponding externalities, in the same way that negative externalities should be curtailed through taxes, etc.

The use of taxes or tradable quotas means, provided the appropriate level is chosen, that polluters take the externalities into account in making their decisions. For cost-benefit analysis purposes this implies that we deal with the externality by valuing taxed goods inclusive of environmental taxes or allowance prices, cf. the discussion in Chapter 6 of the NOU 1997: 27 Green Paper. When calculation prices include taxes or allowance prices that reflect the cost of emissions to society, no further corrections shall be made in the cost-benefit analysis with regard to the relevant externality. This Committee addresses adjustment for externalities from greenhouse gas emissions in Chapter 9. The valuation of externalities that are not related to goods traded in markets is discussed in more detail in Chapter 2.6.

Other types of market failure

Various types of market failure, such as monopolistic pricing, unemployment and asymmetric access to information, imply that market prices should in principle be corrected to arrive at the appropriate calculation prices, cf., inter alia, the discussion in Chapter 5 of the NOU 1997: 27 Green Paper. One issue is the extent to which we should in practice take market failure into consideration when carrying out cost-benefit analysis. One example may be valuing inputs at a price below market price because these are manufactured by an enterprise that exercises market power. Such a correction of market prices is in conformity with the social opportunity cost principle, and is therefore theoretically correct. However, it is challenging to determine what degree of market failure is of relevance to each individual project. In addition, the authorities pursue projects and measures specifically targeting various types of market failure, for example in the form of labour market measures or competition policy measures. The NOU 1997: 27 Green Paper therefore recommended the use of actual producer prices for analysis purposes, unless otherwise indicated by special considerations.

2.6 Valuation of goods not traded in markets

Many effects may be difficult or undesirable to value in cost-benefit analyses. In Chapters 9 and 10, the Committee discusses the valuation of greenhouse gas emissions and the concept of the value of a statistical life. The terms of reference instruct the Committee to examine how the valuation of environmental goods, health effects and time savings may be expected to develop over time, which is discussed in Chapter 4. Apart from that, the Committee does not embark on any renewed assessment of the valuation principles underpinning cost-benefit analysis.

There are two main approaches to measuring the willingness to pay for goods that are not traded in markets. The first approach, which is often discussed under the general term indirect methods, is based on the actual behaviour of individuals in existing markets. This approach can be useful when the use of the non-market good is closely related to market-traded goods, for example by way of housing prices being influenced by external noise, or when the use of a wilderness area involves travel expenses to access such area. The second approach, often termed direct methods, is based on asking individuals how much they are willing to pay. For a more comprehensive discussion, see the NOU 1997: 27 Green Paper or e.g. Freeman (1993), Perman et al. (2003) or Pearce et al. (2006).

No definite answer can be given with regard to how far one should go in attributing monetary values to, for example, environmental goods or good health. Some economists have argued that cost-benefit analysis should normally only value goods with a readily definable market value, whilst other goods should normally be excluded from the analysis of valued elements. Others have noted, inter alia, that the choices eventually made will in any case reflect a form of implicit valuation, and that it may therefore be desirable to perform an explicit (albeit uncertain) valuation in a cost-benefit analysis context. This has previously been discussed in the NOU 1997: 27 Green Paper, the NOU 1998: 16 Green Paper and the NOU 2009: 16 Green Paper. The monetary valuation of environmental and nature goods for cost-benefit analysis purposes is a major, and at times disputed, theme in economic theory; see Hanley and Barbier (2009) for en overview and examples of practical uses.

In October 2010, the Government appointed a designated expert committee on ecosystem services, charged with examining, inter alia, the valuation of the diversity of nature in Norway. The Expert Committee appointed to examine the cost-benefit analysis framework has not addressed the issues to be discussed by the said committee.

2.7 Tax costs and user payments

Many public measures concern public services that may often be difficult to fund in the market. In such cases, these measures need to be funded through taxes or user payments.6 Taxes will generally result in consumers and producers facing different prices. Such tax wedges will distort production and consumption decisions, thus imposing an efficiency loss on the economy.7 For all projects to be funded via public budgets, one should therefore calculate a tax funding cost, which is the marginal cost of collecting one additional krone in tax.8 The tax funding of public projects is discussed in more detail in the NOU 1997: 27 Green Paper. As discussed above, the tax system is also characterised by a number of taxes intended to correct so-called externalities, like environmental and health costs, in the consumption of certain goods. If these taxes are designed correctly, they do not give rise to such an economic efficiency loss. Taxes intended to correct for externalities are discussed in more detail in the NOU 2007: 8 Green Paper; An Evaluation of Excise Duties, and the NOU 2009: 16 Green Paper; Global Environmental Challenges - Norwegian Policy.

User payments will, unlike general taxation, only affect those individuals who make use of the relevant goods or services. However, the economic effects of certain forms of user payments may in many cases have similarities with the effects of general taxation. If central government introduces user payments, for example tolls to meet a need for funding an investment in a road with no queuing problems, and such tolls exceed the use-contingent operational and maintenance costs of the road, users will be faced with a price exceeding the economic cost of using the said road. Consequently, the road will be used less than would be economically desirable. The difference between the economic cost and the user payment corresponds to the tax wedge arising upon general taxation. When making a trade-off between tax funding and user payment, we need to compare the welfare loss arising through user payment with the welfare loss arising through general taxation. In addition, we need to take the cost of collecting the tolls into account.

In the case of toll funding of a road investment that involves, generally speaking, no queuing or other externalities, including no wear and tear, etc., the optimal user price will be zero if we disregard funding needs.

When analysing a road with queuing problems, user payments may be considered a charge introduced to correct for externalities. If there is queuing on the road it will therefore be optimal to make use of road pricing9 to allocate the limited road capacity unless the collection costs are excessive. This example serves to remind us that we should always make use of taxes that improve resource allocation before resorting to distorting taxes. The use of road pricing to allocate the limited road capacity does, when taken in isolation, reduce the need for investing in additional road capacity. Consequently, a cost-benefit analysis of road projects in an area characterised by queuing problems should be based on a situation with optimal pricing of road capacity, instead of assuming that the capacity problem as a whole must be solved through expanded road building. The relationship between road pricing and any wider impacts of transportation projects in urban areas is discussed in Chapter 7 on wider impacts.

2.8 Partial and general equilibrium models

The use of calculation prices offers a simple and expedient method for evaluating the economic profitability of a project. However, if a project is sufficiently large to have a significant impact on market prices, fixed calculation prices cannot be used. Consequently, we need to know when a project is of such a magnitude as to require other methods than calculation prices. In an open economy like Norway, most projects can be considered relatively small. Large projects may induce interactions between price and income changes requiring that these effects be examined simultaneously. This will normally necessitate the use of a general equilibrium model. International measures against climate changes are examples of large projects, because such measures will influence the prices of energy goods and transport-intensive goods. The Committee reverts in more detail to the use of general equilibrium models in Chapter 4 on real price adjustment over time and in Chapter 7 on wider impacts.

2.9 Bibliography

Coase, R. H. (1960). The Problem of Social Cost. Journal of Law and Economics, 3.

Diamond, P. and J. A. Mirrlees (1971a and b). Optimal Taxation and Public Production I: Production Efficiency and II: Tax Rules. American Economic Review, 61.

Dreze, J. and N. Stern (1987). Theory of Cost-Benefit Analysis. In Auerbach, A. J. and M. Feldstein, Handbook in Public Economics, Vol. 2. Elsevier Science Publishers.

Freeman, M. A. III (1993). The Benefits of Environmental Improvement: Theory and Practice. Resources for the Future.

Hagen, Kåre P. (2005). Economic Policy and Economic Profitability. Cappelen Academic Publishers. (In Norwegian only. Norwegian title: Økonomisk politikk og samfunnsøkonomisk lønnsomhet.)

Hanley, N. and E. B. Barbier (2009). Pricing nature: cost-benefit analysis and environmental policy. Edward Elgar Publishing Limited, Cheltenham, United Kingdom.

Ministry of Finance (2005). Cost-Benefit Analysis Guide. (In Norwegian only. Norwegian title: Veileder i samfunnsøkonomiske analyser.)

NOU 1997: 27 Green Paper, Cost-Benefit Analysis.(In Norwegian only. Norwegian title: Nytte-kostnadsanalyser.)

NOU 1998: 16 Green Paper, Guidance on Using Profitability Assessments in the Public Sector. (In Norwegian only. Norwegian title: Veiledning i bruk av lønnsomhetsvurderinger i offentlig sektor.)

NOU 2007: 8 Green Paper, An Evaluation of Excise Duties. (In Norwegian only. Norwegian title: En vurdering av særavgiftene.)

NOU 2009: 16 Green Paper, Global Environmental Challenges – Norwegian Policy. (In Norwegian only. Norwegian title: Globale miljøutfordringer – norsk politikk.)

Pearce, D., G. Atkinson, P. Mourato (2006). Cost-Benefit Analysis and the Environment. Recent Developments. OECD 2006.

Perman, R., Y. Ma, M. Common, D. Maddison and J. Mcgilvray (2003). Natural Resource and Environmental Economics (3rd Edition). Addison Wesley

Tirole, J. (1988). The Theory of Industrial Organization. The MIT Press.

Weitzman, M. L. (1974). Prices vs. quantities. Review of Economic Studies, 41.

Footnotes

1.

A somewhat more rigorous, but nonetheless readily accessible, presentation is provided in Hagen (2005), which also discusses other aspects of cost-benefit analysis and public decisions. See for example Dreze and Stern (1987) for a more technical analysis.

2.

This is a consequence of the so-called Diamond-Mirrlees Theorem, cf. the NOU 1997: 27 Green Paper and Hagen (2005). A technical analysis is presented in Diamond and Mirrlees (1971a and 1971b).

3.

Valuation in the same manner as would be done by private enterprises is a special instance of what is often termed the weighted average rule, and the resources are still valued at their value under the best alternative option, cf. the discussion in Chapter 6 of the NOU 1997: 27 Green Paper.

4.

The scope of private players for solving the problem depends on clearly defined property rights to the relevant resource, cf. Coase (1960). In addition, the level of transaction costs needs to be moderate (the number of agents affected should, for example, be small).

5.

In some cases the market may also generate excessive research and development, because the enterprises are competing for the distribution of a fixed profit, cf. e.g. the overview in Tirole (1988).

6.

The text in Section 2.7 is in large part obtained from Chapter 7 of the NOU 1997: 27 Green Paper.

7.

A tax on labour will for example insert a wedge between the net wage received by workers and the gross wage paid by enterprises. The amount of work performed is less than the economically optimal amount, thus giving rise to an efficiency loss; see for example Hagen (2005).

8.

The NOU 1997: 27 Green Paper refers to calculations that estimate this cost to be about 20 percent on average.

9.

Road pricing (also termed road user charges) is discussed in more detail in the NOU 2007: 8 Green Paper.

Go to front page