Social Cost Benefit Analysis

The broad purpose of SCBA is to help decision making to make the decision that is beneficial to society. More specifically, the objective is to facilitate the more efficient allocation of society´s resources. Societal Cost Benefit Analysis is not a new gimmick. It is a tool with a long international history , specifically with respect to project appraisal in developing countries, the evaluation of infrastructural works in the US and Europe, and more recently with respect to more qualitative issues and themes such as environmental issues, sediment- and water quality issues. The method contributes, at least potentially, to transparent and better decision making processes. This implies however a good understanding of the relevant system, the definition of the problems and the specification of the alternative solutions to these problems.

As policy and decision making with respect to the Water Frame work Directive (WFD) including priority substances is within the public realm, SCBA provides the appropriate structure for the Decision Support System (DSS).
 
Steps to be taken in an SCBA:
In general the SCBA includes several steps. In order to ease application of the SCBA each step is discussed and examples (models, methods and data) are included based on a Swedish SCBA related to acidification as well as the pollutants impacts on health, biodiversity, base cation and corrosion.
 
Step 1: Analyze the system and define the problem
A starting point in the SCBA is a system and a problem definition to base the analysis upon. In the case of Swedish example the analysis relates to the socioeconomic and environmental effects of the emissions of acidifying pollutants such as sulphur dioxide and nitrogen oxides following the implementation of the CAFE (Clean Air For Europe) programme´s Climate Protocol Current Legislation Baseline scenario (CP_CLE_Aug04(Nov04)), developed by IIASA using RAINS.
 
Boundaries: To start with the derived effects of NOx via ozone formation are not included in the analysis. The studied effects are health and the environmental impacts related to the emissions of SO2, NOx, NH3 as well as secondary sulphate and nitrate particles. Since air pollution is transboundary the cost side is about all Source Contributors to emissions i.e. European countries and the benefits are estimated for the Swedish population during the period 2000-2020.
 
Step 2: Make an analysis of the evolution of the problem if no actions are taken (the so called 0 — alternative or baseline scenario)
In this step identification of the project´s impacts is a prerequisite for the analysis. A distinction is to be made between a base line scenario and a scenario including other measures to be taken.

Table 1 presents a series of scenarios i.e. Current Legislation (CL), A (low), B (medium) and C (high) as well as the Maximum Feasible Reduction scenario for the period 2020 compared to 2000 for ecosystems area with nitrogen deposition above the critical loads for eutrophication. As shown, compliance with current legislation using different measures would lead in the case of Sweden to 16% of ecosystem areas with nitrogen deposition above the critical load for eutrophication.

Furthermore, Table 2 depicts kton pollutants emission and deposition in Sweden where contribution from other countries constitutes the lion share of the depositions. Hence an important part of the measures to be taken to reduce deposition in Sweden are to be taken elsewhere in Europe.

When it comes to the effects of the actions the analysis are taken in the following way: Although the relation sulphur deposition and water pH is dependent on a number of variable ecosystem specific factors, regression analysis are used in this case to give an idea of the impact of air pollution on acidification in the case of the north and the south of Sweden, respectively. The data is for the period 1984-2003 where the water pH level is measured during the four seasons of the year. The dependent variable is pH levels while the independent is sulphur deposition in logarithmic form.

On the other hand, except the correlation between air pollution and pH values, a further correlation is identified between pH values in water and fish stock such as:

Step 5: Indicate the level of uncertainty when predicting these effects (needed for sensitivity analysis)
Although the effects of acidification on freshwater ecosystems are better understood than impacts on terrestrial ecosystems the uncertainty issue cannot be underestimated. The uncertainties may be based on several factors such as the models used and the available data etc. In the case of the results based on regression analysis although the coefficients are highly significant, uncertainty may be the result of using linear models instead of non linear ones. Therefore a sensitivity analysis is required to highlight the ranges of variations in the results in the case for example when other models are used for the estimation.

Step 6: Quantify the effects of the actions and valuate them (NOT weighing) 
Table 6 shows the effects of the actions related to different scenarios (Holland et al, 2005) where reductions of acidifying emissions are showing benefits in the case of Swedish ecosystem related to catchments area with acid deposition bellow the critical loads. Improvements are not uniform, however, and recovery will take some time, first for water chemistry to stabilize and improve, and then for biological recovery (WGE, 2004). As shown, comparison of the CLE scenario to the other ones and especially the MTFR scenario highlights the importance of stringent measures to achieve maximum acidification reduction.

Step 7: Welfare valuation should be made of all effects irrespective of their nature and irrespective of the place where they occur. Always in physical entities, (less sediments, more water for irrigation, more nature etc) and wherever possible (also) in monetary values. A whole range of different methods exists.
In order for physical measures of impacts to be commonly measurable, they must be valued in monetary units. The monetary valuation of different effects is not a straightforward procedure since many of the effects have no market value. In general we often talk about the total value of something. As shown in Figure 1, this total value is often composed of both use and non-use values. The use value is the value derived from actual use of a good or service. The non-use values, also referred to as “passive use" values, are values that are not associated with actual use, or even the option to use a good or service.

The use value includes direct, non-direct and option values. The direct use value is the value attributed to direct utilisation of ecosystem services. Non-direct-use values or "functional" values relate to the ecological functions performed for example by forests, such as the protection of soils and the regulation of watersheds. Option value is the value that people place on having the option to enjoy something in the future, although they may not currently use it. On the other hand, the non-use values include both bequest and existence value. Bequest value is the value that people place on knowing that future generations will have the option to enjoy something. Existence value is the value that people place on simply knowing that something exists, even if they will never see it or use it. In order to assess these values, environmental economics uses several methods.

These methods may be based on stated preferences involving studies including questionnaires asking respondents for their willingness to pay such as in the case of contingent valuation method (CVM) and choice experiment method. Other methods are based on revealed preferences that are often based on consumers´ or producers' behaviour or actions such as: The hedonic price method is used to estimate the value of environmental effects on properties such as the effect of noise or air pollution on house prices; the production function method is used to estimate the value of the environmental effects on production such as the effect of ground-level ozone on the production of wheat or timber.

Figure 1: The total economic value

The willingness to pay for ?s (the change in the risk to die) leads to the value of statistical life such as:

In both the cost case and the benefit case it is sometimes difficult to find values.
One way to solve the problem is to calculate them. However, this solution may be very costly. If there are not enough resources, published values may be adapted to the problem being studied. The method is called benefit transfer and it is well known and used in many studies.

In the case of acidification and biodiversity the benefit transfer method has been used. In the Nordic countries and especially Sweden many studies have been conducted in order to estimate the benefits derived from conservation and/or recovery of aquatic biodiversity which can be achieved through reduction of acidification in the ecosystems.
Table 7 is a review of some studies conducted in order to estimate the Swedish willingness to pay related to biodiversity. Common to these studies is the use of CVM to estimate the total value or the marginal values of benefits. In Toivonen et al. (2000) where the CVM was conducted in all Nordic countries, the WTP to preserve fish stock in Sweden is estimated to 408 SEK (1999 prices). In both Laitila et al. (2002) and Paulrud et al. (2003) CVM studies emphasized the marginal WTP to increase fish stock in Swedish lakes. For doubling the number of fish in Swedish lakes the estimate values by Paulrud et al (2003) range between SEK 24 and SEK 160.

On the other hand there are extreme cases where it is not possible to even make a benefit transfer depending on the fact that no valuation studies have been conducted.
The reasons for the low number of studies may be several such as the general public low awareness of the effects of some pollutants on health and especially on the environment. Depending on the scarcity of these studies, alternative methods may be used to estimate proxies for the damage cost. There are especially two methods that may be used:
The first method is called the standard price method (Vermoote and De Nocker (2003) and the second one is called the Ecotax method (Johansson (1999) and Finnveden et al (2006).

The standard price approach estimates the revealed preferences of policy makers. It calculates the benefits of emission reduction — as perceived by policy makers - based on the abatement costs to reach a well-defined emission reduction target (Vermoote and De Nocker (2003). These costs are a proxy for the benefits that policy makers attribute to these reductions, as we assume that policy makers act as rational decision makers who carefully balance (their perception of) abatement costs of emission reductions with (their perception of) the benefits of these emissions. However, as the standard price approach is based on the current preferences of policy makers it cannot be used for cost-benefit analysis or policy advices.

The ecotax method also called the valuation weighting method is a monetarisation method based on a tax system. This method has its origin in Life Cycle Analysis where in order to create a valuation method different existing environmental taxes and fees are connected to the appropriate impact category. The method relies on two basic assumptions. The first is that the members of parliament represent the will of the people, and the second is that the environmental tax system represents the priorities of the parliament.
Furthermore, there are other effects related to the implementation of measures (and compliance with policy instruments) that are relevant to a SCBA. These effects may take the form of product price increase as well as the effect on employment. The product price increase due to the implementation of some measures are to be found in all sectors i.e. energy, transport, industry and agriculture. Compliance of these sectors with the policy instruments e.g. taxes is not cost free and some of these costs may be passed on to consumers. Unfortunately, depending both on lack of data and on limited resources to analyze this issue, these cost are often not captured in SCBA analysis. When it comes to employment for instance, changes may be based on policies, projects or programmes and should be included into a SCBA. Since the effect measures would have some impacts on cost and thereby on prices, it is probable that it will imply some effect on employment. In order to analyze these hypotheses a general equilibrium model should be used. A general equilibrium model (also called CGE (Computed General Equilibrium) model) gives an understanding of the whole economy, a whole market system where the prices and production of all goods, including the prices of money and interest are related.

Step 8: Some effects may occur during a length of time. Discounting rules should be applied to calculate the NPV (net present value; it may be useful to use more than one discounting rate)
Once all relevant cost and benefit flows that can be expressed in monetary amounts have been defined, it is necessary to convert them all into present value (PV) terms. This necessity arises out of the time value of money, or time preference. The question arises how the time effect can be taken into account, and how future cost and benefit flows made comparable regardless of when they occur. The answer is that all cost and benefit flows are discounted using a discount rate. For example, discounting (of natural assets) is a process of determining the present value (net worth) of assets by applying a discount rate to the expected net benefits from future uses of those assets (OECD (2007)). Hence the PV of a cost or benefit (x) received in time t using the discount rate r is calculated as follows:

Step 9: Select and decide (with the relevant stakeholders) what the most appropriate actions are to be taken.
Compile a SCBA balance sheet with costs on one side and benefits on the other.
The main purpose of the CBA is to assist the selection of policies that are efficient in terms of their use of resources. The criterion applied is the Net Present Value (NPV) test. This simply asks whether the sum of discounted gains exceeds the sum of discounted costs. If so the project can be said to be efficient given the data used in the SCBA. In other words, the NPV of a project is:

Furthermore, except the NPV test a sensitivity analysis is also recommended. The main reason for conducting a sensitivity analysis is to check whether or not the results change and what is the magnitude of that change when the values of some key parameters have changed. These key parameters may be the discount rate and the chosen scenario as shown in Table 10. Of course the value of other key parameters may also change. These parameters may be physical quantities and qualities of input as well as outputs and the time span.

Step 10: Decide on extra research questions that have to be answered, a monitoring system and a possible re evaluation later on. 
The reader may have noticed the strong resemblance with the basic structure of the DSS. Once it comes to the selection of alternative (clusters of) measures that may be taken to improve the water quality by lowering concentrations of priority substances step 2.6 explicitly deals with the specific methodology to be used
In contrast with REACH the guidelines of the EC do not imply that a Social Cost Benefit Analysis (SCBA) has to be carried out in the implementation of the WFD.
It may very well be possible that the problem — and solution framing is very straightforward and no SCBA may be needed. For example if there are no other effects of the measures than the costs of compliance and the reduction in concentrations. In this case more simple selection methods can be applied and the work that has to be done in steps 5 and 6 may then be relatively simple. In other words: only from step 5 on it will be decided whether or not to conduct a SCBA. However, steps 1 through 4 always have to be taken, regardless of the of the selection methods that may follow from the analysis in step 5. The work that has to be done in step 1 through 4 is not carried out in vain and is consistent with the guidelines of the WFD.
As the priority substances have to be dealt with in relation with the WFD, they will be part of the River Basin Management Plans that have to be compiled. This implies that the PSs should not be dealt with in isolation but in relation with the RBMPs and that a lot of information that is needed to make decisions about the PSs will have to be gathered any way.