4 The role for non-market valuation
Non-market values are regularly included in cost-benefit analyses to indicate the value of changes in environmental quality. Sometimes the value of environmental quality is evident in market data. For example, the cost of reduced water quality might in part be indicated by the value of declining commercial fish production. However, this usually comprises only a part of the total economic value of a change in environmental quality. Non-market valuation is regularly used by government agencies to estimate the community wide value of changes in environmental quality for cost-benefit analyses. It can also be used in any other context where there is a need to identify the marginal value of a change in environmental quality - such as in the application of economic instruments.
4.1 Non-market values
When applying non-market valuation techniques, several different kinds of non-market values can be estimated (Pearce and Markandya, 1989).
- Use value refers to the value that individuals receive from the actual use of a resource. This includes any kind of consumption, benefits from amenity or aesthetics, vicarious use and recreational use.
- Option value refers to the value of maintaining opportunities into the future and reflects the value of uncertainty. In theory it can have either a positive or a negative value. Uncertainty can arise about the provision of a resource (supply-side option value) and if individuals are willing to pay a premium to remove this uncertainty, then the option value is positive. However, if the uncertainty arises about whether individuals will have a demand for a resource (demand-side option value), then the value is negative.
- Existence value refers to the value individuals have for a resource apart from any in situ use of a resource. It arises for various reasons, including altruism, bequest motives and stewardship. This value is sometimes described as non-use or passive-use value.
Total economic value is the sum of use value, option value and existence value. It is this value that one seeks to estimate using non-market valuation techniques. [The two main databases of non-market values in New Zealand and Australia are http://learn.lincoln.ac.nz/markval/ and http://www.epa.nsw.gov.au/envalue/.]
4.2 Techniques for valuing changes in environmental quality
As mentioned above, part or all of total economic value can sometimes be derived from existing market data. This typically involves looking at changes in damage costs or preventative expenditures, or estimating changes in producer and consumer surplus. [Changes in consumer and producer surplus indicate the value to consumers and producers of changes in production. Consumer surplus is the difference between the price paid for a good and the amount that people would be willing to pay, while producer surplus refers to the difference between what producers are paid for a good and their costs.] For example, information about the reduced cost of treating potable water provides an indication of the benefits of reduced turbidity (Moore and McCarl, 1986). Or it may be possible to estimate the costs to tourism from an outbreak of blue-green algae (Walker and Greer, 1992). However, many of the benefits of improved water quality can only be estimated by using related or hypothetical market data. In these cases it may be possible to estimate non-market values using either revealed or stated preference non-market valuation techniques.
4.2.1 Revealed preference techniques
Revealed preference techniques use information from related markets to impute a value for non-market goods (Mäler, 1974). A related market is one that indirectly reveals values for environmental goods. The most commonly used revealed preference techniques are the hedonic price method and the travel cost method. The hedonic price method, a variant is the hedonic wage model, uses differences in property prices or wages to impute a value for changes in environmental quality such as noise, air quality, water quality, etc. In most hedonic price studies a regression equation is estimated where the price or wage is a function of all of the attributes, including environmental quality. The effect of changes in environmental quality on property prices or wages can then be quantified.
A second revealed preference technique is the travel cost method. The travel cost method is used to value recreational quality or changes in recreational quality. It uses information on travel costs and reductions in the number of site visits at greater distances from a site to estimate recreational use value. This technique is regularly used by State government agencies such as the National Parks and Wildlife Service to estimate recreational values (Gillespie, 1997). Where changes in environmental quality affect visitation rates, the travel cost method can also be used to value changes in recreational value. For example, Bockstael, Hannemann and Kling (1987) estimated that people were willing to pay $27 per household per season for a 30% reduction in oil, chemical oxygen demand, turbidity and faecal coliform pollution at downtown Boston beach sites.
However, revealed preference techniques such as the hedonic price and travel cost methods can only be used in limited situations where there are existing related market data. They can only be used to estimate use values and they are retrospective which limits their usefulness for cost-benefit analyses which typically require valuation of options which don't yet exist and of non-use values.
4.2.2 Stated preference techniques
A second class of techniques that can be used to estimate non-market values are those based on the stated preferences of individuals. Stated preference techniques involve the use of surveys from which estimates are derived of the non-market benefits of different resource use alternatives. Because they rely on the use of surveys, stated preference techniques can be used in more applications than revealed preference techniques. They can be prospective and used where no related market data are available for estimating use values. They can also be used to estimate non-use values.
The most widely used stated preference technique for estimating non-market values is the contingent valuation method (CVM). CVM questionnaires contain several well defined elements including a description of the study site, details of the proposed changes (including a method of payment), an elicitation question and a series of socioeconomic and attitudinal debrief questions. State-of-the-art applications of the CVM generally utilise the 'referenda' format for the elicitation question, an example of which is shown in Text Box 9.
Text Box 9: The referenda CVM format
Would you vote to support the proposal to improve water quality at a cost of $50 per household, or do you oppose the proposal? (Tick one box.)
I would vote to support the proposal at a cost of $50
I would vote against the proposal at a cost of $50
Under this format, respondents are asked whether they support a project given that they are required to pay a certain amount towards it, with the payment amounts being varied between respondents. The responses to the elicitation question are then regressed against several variables including the payment amount, respondents' attitudes, and socioeconomic characteristics such as income, age, education etc. This equation is then used to estimate mean and median willingness to pay.
The CVM has the advantage of being recognised by respondents as a standard public choice instrument (as it is similar to a referendum). However, despite its wide usage, the CVM has several limitations. It is relatively costly to use, provides limited information about people's preferences and is arguably prone to various biases (Kahneman and Knetsch, 1992; Diamond and Hausman, 1994).
A second stated preference technique that has been increasingly used to estimate the value of improved environmental quality and could be applied to valuing improved water quality is choice modelling (CM). CM has been used in a small, but growing, number of studies to estimate non-market values. It has grown in popularity because of its informational efficiency, ability to generate values for resource attributes (see below) and because of concerns regarding the validity of contingent valuation. Given its increasing usage in non-market valuation application, we have provided additional detail in this report about this technique, including about the value estimates that can be generated using it (see Bennett and Blamey, 2001).
CM questionnaires are similar to CVM questionnaires in that they contain background information about the non-market good, an elicitation question, and debrief questions. The main difference between the two methods is in the form of the elicitation question. In CM questionnaires, respondents are presented with a series of choice sets, each containing usually three or more resource use alternatives. An example of a choice set is shown in Table 3. From each choice set, respondents are asked to choose their preferred alternative. The alternatives in the choice sets are defined using a common set of attributes (eg characteristics such as water quality, % native fish, health of aquatic plants etc.), the levels of which vary from one alternative to another.
Text Box 10: Example of one choice set in a choice modelling questionnaire
Please indicate the alternatives you prefer most by ticking one of the boxes below:
| Alternative 1 |
Alternative 2 |
Alternative 3 (the status quo) |
|
|---|---|---|---|
| Water quality | Good | Fair | Poor |
| % native fish | 50% | 80% | 5% |
| Health of aquatic plants | Fair | Good | Poor |
| Increased water rates | $10 | $40 | $0 |
In CM applications goods are decomposed into a set of 'attributes' or characteristics. For example, a car could be considered to be simply the sum of its component parts i.e. 4 wheels, a chassis, an engine etc. Or in the case of valuing river quality, the characteristics might be water quality, the percentage of native fish and the health of aquatic plants. The trade-offs respondents make when choosing between alternatives are quantified using statistical techniques. Where one of the attributes involves a monetary payment, the resulting trade-offs can be used to estimate the value of each of the environmental quality attributes. This can be conceptualised in the case of purchasing a car. Existing market data might show that on average people may be willing to pay $1000 extra for air conditioning - this implies that air conditioning is worth this amount of money. Similarly, CM survey data may indicate that respondents are, on average, willing to pay $5 extra for an additional fish species to be present in a certain river.
Two different types of value estimates are derived in choice modelling applications. The first is known as implicit prices. These are point estimates of the value of a unit change in an attribute. They are useful for management decisions where information is required about the value of marginal changes in environmental quality, such as the value of an extra 100 km2 of wetland preserved. They are also useful for identifying the relative importance people place on different attributes. Implicit prices are calculated as follows, if utility is a linear function of all attributes:
IP = βA/βM
where IP is the implicit price, βA represents the coefficient of the Ath environmental attribute, and βM represents a monetary attribute.
The second type of value estimate derived using choice modelling is compensating surplus (or variation). This corresponds to what is estimated using contingent valuation. Compensating surplus is the value of a discrete change in environmental quality. This can be different from the sum of the changes in the implicit prices if the value that respondents have for an environmental improvement is not totally explained by the changes in the attributes. These value estimates are appropriate for use in cost-benefit analysis. Assuming that the only personal characteristic is income (m), and prices and other environmental goods are constant, compensating surplus is defined as follows:
V(x0, m) = V(x1, m-cs)
where V is an indirect utility function, x0 is the original vector of attributes, x1 is the vector of attributes after the change in environmental quality and cs is compensating surplus.
In practice, for a standard conditional logit model, this is estimated as follows:
where βM is the coefficient for the monetary attribute and is interpreted as the marginal utility of income, V0 represents the utility of the initial state, and V1 represents the utility of the subsequent state.
4.3 The use of non-market valuation with the application of economic instruments
Non-market valuation techniques are frequently used in cost-benefit analysis where projects have an environmental dimension. They have been used to a much lesser extent with economic instruments, but theoretically at least non-market valuation has an integral role in the application of many economic instruments. The use of economic instruments does not preclude the use of non-market valuation. Rather non-market valuation can be used to improve the use of economic instruments.
Many economic and traditional regulatory instruments require that optimal levels of environmental quality be established independently. For example, traditional command and control approaches, trading schemes and offset schemes all require the setting of environmental standards. Whenever standards are set there is a potential need for non-market valuation. Theoretically, approaches are optimal when the marginal benefit from increasing the standard equals the marginal cost of doing so.
Non-market valuation is typically needed to measure non-market benefits. For example, consider a proposed policy to restrict nutrient loadings in a catchment - deciding whether or not this policy is appropriate would require estimates of both marginal benefits and marginal costs of implementation. If the marginal benefits are less than the marginal costs, then this new policy would decrease social welfare. If the opposite is true, the policy will be welfare enhancing. Similar to cost-benefit analysis, the determination of marginal benefits in this context would involve estimating Total Economic Value. In some cases it may be possible to use only revealed preference techniques. However, where non-use values need to be estimated, or where the change in environmental quality cannot be related in any way to existing market data, this would require the use of stated preference techniques.
Other economic instruments can also benefit from the use of non-market valuation techniques. For example:
- For an optimally set environmental charge it is necessary that the charges be set equal to the damage cost of the polluting activity. Estimation of these damage costs will normally require the use of non-market valuation. For some damages (e.g. due to effects on drinking water costs, or where any preventive actions are required), market data could be used to estimate this value. However, where recreational values are affected, techniques such as the travel cost method (a revealed preference technique that is commonly used to estimate recreation values) would be needed. And if existence values are affected - which is often the case when environmental quality changes - stated preference techniques would be used.
- For tendering, non-market valuation can be used to reveal whether or not the marginal benefits of any environmental improvements achieved exceed the costs of implementing the scheme and hence, if the budget allocated to this instrument should be increased.
- Offsets involve reductions in environmental quality in one area, but improvements in other area(s). As discussed previously, the community may have different value for these changes in environmental quality, even if the changes in environmental quality are ecologically similar. Non-market valuation provides one method of determining whether marginal benefits to the community from these multiple changes are equivalent, and whether an offset programme would indeed enhance community welfare. Choice modelling would be a particularly useful non-market valuation technique in this context as it provides estimates of the value of the attributes of different resources. It can therefore readily be used to indicate whether an offset proposal would increase, decrease or have no effect on community welfare. Studies of this kind are already being undertaken in New Zealand (Kerr, 2003).
Thus non-market valuation should be seen as a means to increase the effectiveness of economic instruments. It is not the case that non-market valuation is not needed if economic instruments are being used; rather non-market valuation is needed for the optimal operation of economic instruments.
