Introduction and definition
The contingent valuation method (CVM) is a widely used method for estimating economic values for all kinds of ecosystem services and environmental goods which are not traded in the market and hence have no market price. CVM is typically used to estimate the benefits (or costs) of a change in the level of provision (or in the level of quality) of a public good. This information can then be used in a cost-benefit analysis, which assesses the impacts of government project or policy. For instance, imagine an increase in public investment out of current taxes for improving the quality of water in a river or sea by treating sewage. It is easy to count the costs: the amount of money spent. However, what are the benefits? We could try to count them one by one in their respective units, for instance better public health, less odour, availability of non-contaminated fish. On the other hand, we could ask a representative sample of the concerned local population how much they would have been willing to pay (in the forms of taxes for instance) in order to improve the water quality. Adding these results over the whole population, we would have a monetary representation of the benefits obtained.
Use and non-use values
CVM can be used to estimate both use and non-use economic values, and it is the most widely used method for estimating non-use values. Use values are those values which are derived from actual use of a good or service, such as visiting a national park or using a beach for recreation. They also involve non-consumptive uses like basic life-support functions associated with ecosystem, health or biodiversity, the enjoyment of a scenic vista, or event having an option to fish or watch birds in the future. The non-use values do not involve direct use of a resource or ecosystem service. They comprise several forms of philanthropic relation to nature, e.g. existence value, the value people place on simply knowing that giant pandas, whales, a certain protected area or a beach exists, even though they will never see or visit them. It also includes the bequest value, the satisfaction of preserving the natural environment for future generations.
Willingness to pay and willingness to accept
The contingent valuation method is applied through conducting a survey in which people are directly asked how much they would be willing to pay (WTP) for a (change in) specific environmental service. It is also possible to ask people the amount of compensation that they would be willing to accept (WTA) to give up an environmental service. The first approach is called ‘contingent’ valuation, because people are asked to state their willingness to pay, contingent on a particular scenario and the environmental service described to the respondent.
The first step is to define a (change in) a good or service being valued (e.g. improving a lake water quality that would lead to a 20 percent increase in fish stock). Then decisions about the survey itself are made, such as whether it will be conducted by mail, phone or in person, how large the sample size will be and who will be surveyed (e.g. only visitors or both visitors and non-visitors; individuals at the local, national or international scale). Answers to questions regarding survey method and sample size depend mainly on the size of the research budget, while the choice of subjects will depend on (1) whether one decides to estimate only use or both use and non-use values, and (2) on the uniqueness of goods or services being valued (resources with unique characteristics are likely to have higher non-use value and thus the geographical scope of the survey should be larger). In-person interviews are generally the most effective for complex questions, because it is often easier to explain the information to respondents in person. In addition, people are more likely to complete a long survey when they are interviewed in person. However, these are also the most expensive type of surveys. The survey sample should be a randomly selected sample of the relevant population (e.g. every tenth visitor of a national park).
A contingent valuation survey should include (1) a detailed description of a good or service being valued and the hypothetical change regarding the good or service, (2) questions about willingness to pay for a good or service being valued and (3) questions about respondents’ characteristics (age, income, education, etc.) and preferably also their preferences regarding the good or service. The willingness-to-pay question should also define a way in which payment would be made (a general tax, a voluntary donation or an entrance fee). For example, a question can be formulated in the following way: ‘Are you willing to pay EUR __ for the previously described improvement of the river water quality in the form of a voluntary donation per year?’. The valuation question is usually followed by a question which identifies the motivation of those respondents who state that they are not willing to pay anything. This enables distinguishing between the so-called protest votes (respondents who are not willing to pay anything, because they protest against a scenario presented or a payment method) and the people for whom the good indeed has no value. Protest votes are in most cases excluded from the statistical analysis, as they do not reveal people’s real economic value for the good.
Some authors (such as Sagoff (1988), in The Economy of the Earth) have interpreted protest answers as refusals to act as consumers when deciding public policies. He believes that the economic approach, which treats individuals as consumers with certain preferences, is limited and that policy issues should be decided by rational deliberation among citizens. Thus, he plausibly argues that people refuse to give a price, because they want to act as ‘citizens’ (deciding upon policy matters by voting in parliament or in a referendum, demonstrating and/or debating) and not as ‘consumers’ in a fictitious market.
The next step is to conduct the survey, which is followed by statistical analysis and reporting of the results. The main result derived from the CVM is the average willingness to pay per person. This figure is then multiplied by the relevant population (all visitors of a beach or all residents of a country, for instance) in order to derive total economic value of a good or service. For example, if the average willingness to pay of surveyed people for establishing a protected marine area is EUR 20 per person per year, and the relevant population amounts to 200,000 (e.g. annual visitors of the site), then the total benefits of such a project would be EUR 4 million.
Although CVM has been widely used in cost-benefit analysis and environmental impact assessment for several decades, it has been subject to many critiques. The main concern relates to the reliability and validity of its results due to a number of errors or biases that can occur when applying CVM. The most important biases are:
• When respondents are asked about their willingness to pay hypothetically, they tend to give higher values than what they would actually pay in a real situation.
• Rather than expressing value for the good or service, the respondents might sometimes actually be expressing their feelings about the scenario or the valuation exercise itself (they do not believe that a described change is feasible or that it will really take place).
• Respondents may give different willingness-to-pay amounts, depending on the specific form of payment chosen (e.g. if the form of payment is voluntary donation respondents may give higher values than if asked to pay through higher taxes).
• Starting value in the willingness-to-pay question tends to imply a value for the good (e.g. ‘Are you willing to pay EUR 5 for…?’), so that a starting value well above the respondent’s true willingness-to-pay amount will increase the stated willingness-to-pay amount, while starting value well below it will tend to decrease it.
• Strategic bias arises when the respondent does not provide a true answer in order to influence a particular outcome, i.e. provision of a good.
• Non-response bias is a concern, because individuals who do not participate in the survey are likely to have different values than individuals who do take part in it.
Sagoff, M. (1988)The Economy of the Earth: Philosophy, Law, and the Environment, Cambridge University Press.
For further reading:
For a short description of CVM as well as several examples of its application: www.ecosystemvaluation.org/contingent_valuation.htm.
This glossary entry is based on a contribution by Ivana Logar
EJOLT glossary editors: Hali Healy, Sylvia Lorek and Beatriz Rodríguez-Labajos