Contingent+Valuation+Method

= Contingent Valuation Method =

= = toc

**Definition**
This method is generally used by economists to assess the maximum willingness-to-pay for goods and services in the future through the means of survey questions that create a hypothetical market. This article is specifically geared towards its use in developing countries.

**Relevant Characteristics**
__History and Context __ Contingent valuation is a method for assessing what is the maximum value that a person is willing to pay for a good or service. It was originally purposed by Ciriacy-Wantrup in 1947 for assessing the value of soil erosion prevention by the public. Since the 1940s, it has been mostly developed by economists to value nonmarket goods, e.g. forests. Since the early 1990s, it has been used by development agencies and donors in third world countries to measuring willingness to pay for different good and services.

 There is a debate in the development community over whether basic goods and services that may directly or indirectly affect health and economic growth (e.g.water pumps, insecticide treated bed nets, latrines, improved cookstoves, etc.) should be offered to developing countries at cost or at a subsidized, or even free, price. Proponents of subsidized or free products state that impoverished people lack the money to provide themselves with hardware necessary for services such as water or sanitation. Since these sorts of services are fundamental to healthy living and are considered a human right it is argued that they should be provided for free or an extremely low cost. Opponents say that subsidized or free products are not fiscally sustainable; they artificially create demand for products that are not wanted and will not be used. They destroy potential for local entrepreneurship since potential customers in developing countries have expectations for artificially low prices for some goods and services.

 Assessing maximum willingness to pay in the developing world is important for both sides of the subsidy and non-subsidy argument. On both sides, it is important to be able to state what the maximum a person is willing to pay to make the argument that a piece of hardware such as a latrine or hand pump should or should not be subsidized. This allows programs to set appropriate subsidies levels and maximize the impacts of their projects by not allocating an excessive amount of money to a particular family or community. Ideally, it should be used as part of the project process and the information from a contingent valuation survey should be used to set prices and tariffs for goods and services.

 __What are we measuring? __  It is necessary to be clear what the contingent valuation method is measuring since it may not be immediately obvious to a layperson. It seeks to measure the //maximum// that a person is willing to pay for a good or service. It does not necessarily mean that a person is “happy” to pay the price, or thinks that they got a good deal. It is also useful to point out that this is what a person is willing to pay under the normal circumstances of their daily lives and not an extreme case.

 Contingent valuation is measuring what a person would pay for a good or service, therefore, it should take into account the ability to pay. A table in Whittington, 1998 (Table 1) explains the different combinations of willing and ability to pay. Contingent valuation essentially breaks the four categories given in the table into two separate ones.

 Willing and able to pay are described as willing to pay. The other three categories are described as not willing to pay. What may confuse a layperson is that some people may be willing to pay a certain price, but they do not have the income to pay. These people would be considered not willing to pay for the good or service since they would //not// pay for the good or service.

 **Table 1 Adapted from Whittington, 1998 **
 * =  ||= Willing to pay ||= Not willing to pay ||
 * = Able to pay ||= Willing and able to pay ||= Not willing but able to pay ||
 * = Not able to pay ||= Willing and not able to pay ||= Not willing and not able to pay ||

**“Method Made Easy”**
At its most fundamental level, the contingent valuation method asks through a standardized survey: What is the maximum a person is willing to pay for a hypothetical good or service. There are four basic types of survey methodologies that may be used in making survey questions. There is no agreement on which one is best. It is generally recommended that the survey method reflect the type of pricing that the respondent is used to. For example, in a country where haggling for a good in a market is common, one’s survey questions should mimic haggling.

The respondent is simply asked to say what they would be willing to pay for a good or service without the surveyor giving a cue what an expected value for the service is. This may be problematic for some people if previously they received the service for free; for example, putting a price on a water system if previously someone was getting water from a stream. Additionally, this sort of open-ended pricing does not occur in daily life under normal circumstances, which may increase hypothetical bias (see Limitations)
 * Open-ended Question

The respondent is asked at incremental levels what they are willing to pay. For example: “Would you pay $1 for this candy bar?”, “Yes”, “Ok, Would you pay $1.25 for this candy bar?” and so on. It is possible to go from a low value and incrementally increase the price until the respondent will no longer pay or start from a price that the respondent would not pay and incrementally decrease the price until they are willing to pay.
 * <span style="font-family: Arial,sans-serif; line-height: 1.5;">Bidding Game

<span style="font-family: Arial,sans-serif;">The respondent is given a list of prices and is asked to mark the prices that they would be willing to pay. It is important to make sure that the correct range is chosen so that the prices listed are above and below the maximum willingness to pay so that it can actually be resolved.
 * <span style="font-family: Arial,sans-serif; line-height: 1.5;">Payment Card

<span style="font-family: Arial,sans-serif;">The respondent is given a choice between two similar goods or services at two different prices. For example: “Would you pay $5 per month for a water system that is in service for 4 hours a day and you have a tap down the street or would you prefer to pay $10 for service 2 hours a day and but with a private tap at your house.” A simple yes or no is given as the response. If a variety of prices are used, it is possible to use statistical methods to find average prices that people are willing to pay. This method also lowers the possibility of strategic bias (see Limitations). It is not uncommon after a set of survey questions to ask a final follow up if the maximum willingness to pay was not determined. For example, if a person is willing to pay all prices they may be asked what their maximum willingness to pay is at the end of the survey.
 * <span style="font-family: Arial,sans-serif; line-height: 1.5;">Dichotomous (Binary) Choice

**Advantages**
<span style="font-family: Arial,sans-serif;">Contingent valuation can be a useful method when the services that one is trying to measure willingness to pay for are not available. It also gives the ability to measure whether, for example, a town would be willing to pay for a greater level of water service that may have greater benefits for the members of that town.

**Limitations**
<span style="font-family: Arial,sans-serif;">The main limitation of contingent valuation is that it is not a direct measure of willingness to pay. One is not directly selling a good or service but rather speculating with a respondent about a potential good or service. This can lead to two broad categorizations of bias.

<span style="font-family: Arial,sans-serif;">Hypothetical Bias arises from the fact that a “hypothetical” situation is present to a respondent. Respondents may have difficulty taking situation seriously or may not understand what the good or service being provided is. This is one of the reasons that it is important to have a well-crafted scenario in the questionnaires that people can take seriously and have enough information in the survey for the respondent to make an informed choice.
 * <span style="font-family: Arial,sans-serif; line-height: 1.5;">Hypothetical Bias

<span style="font-family: Arial,sans-serif;">Strategic Bias is when a respondent purposefully gives a misleading answer to a question in hopes that it will lead to them gaining something in the future. They are strategically giving a wrong answer since they think it may affect the price of the good or service or whether it is provided in the future. It may take the form of under- or over-bidding. Underbidding may occur in the hopes that the good or service will be offered at a lower price in the future. This under-bidder could be a person that knows that a particular NGO has provided water systems in the past for a low cost and would like to get the water system themselves for the same low price. Strategic overbidding may occur in the hopes that a good or service will be provided since the respondent would be willing to pay for it. For example, if the respondent knows that the NGO providing a water system requires everyone to commit to paying a tariff before it is constructed, that person may strategically commit to paying it, even if they cannot afford it, so that they may have a chance of getting a water system.
 * <span style="font-family: Arial,sans-serif; line-height: 1.5;">Strategic Bias

**Analysis**
<span style="font-family: Arial,sans-serif;">Advanced statistics and an understanding of econometrics are required for full analysis of contingent valuation studies. The basic analysis involves statistical tests for difference between choices. Some authors also try to draw correlations between choices made and different social and economic factors such as income level. Meeks 2012 and Whittington et al. 1990 both contain good description of how to thoroughly analyze the data coming from contingent valuation surveys.

**Method in Context**
<span style="font-family: Arial,sans-serif;"> Like any method, surveyors should be very careful in the study design and question wording within the survey. Often, when a survey is being translated from one language to another, nuances in the languages may change the meaning of questions. Surveyors should also be wary of cultural context in question responses. Well-crafted, realistic contingent value scenarios in the surveys are important to engage the respondent and get reliable answers. Enough information should be given to allow the respondent to make an informed choice about their options. It is considered good practice to simulate normal pricing practices for products, for example, in the developed world people often compare similar items with a fixed price, so the dichotomous choice method would be a good method to choose for a survey.

<span style="font-family: Arial,sans-serif;"> It is considered good practice to give different versions of the same survey to different samples of the same population (split-sampling) to validate that the results are reproducible. This may take the form of starting your first price for a bidding game from different starting points to ensure that there is not a bias due to the starting point. This could also take the form of giving different prices for the dichotomous choice method to different portions of the population. Additionally, as with any survey methodology, it is important to pre-test the survey. Training of interviewers/surveyors is very important in reducing errors from a given survey. They should very clearly understand the purpose of the survey as well as the context of the survey, so as to be able to relate to the situations of the respondents. Training will also allow the survey to be administered in a consistent way, which, in turn, will lower the bias due to differences between the surveyers.

**Online Resources**
<span style="color: windowtext; font-family: Arial,sans-serif;">Ecosystem Valuation: <span style="color: windowtext; font-family: Arial,sans-serif;"> []

<span style="color: windowtext; font-family: Arial,sans-serif;"> What is Contingent Valuation Method?: []

<span style="color: windowtext; font-family: Arial,sans-serif;"> EJOLT: Environmental Justice Organisations, Liabilities and Trade <span style="color: windowtext; font-family: Arial,sans-serif;"> Mapping Environmental Justice: <span style="color: windowtext; font-family: Arial,sans-serif;"> http://www.ejolt.org/2012/12/contingent-valuation/

**Further Reading**
<span style="font-family: Arial,sans-serif; font-size: 10pt;">Carson, Richard T. <span style="font-family: Arial,sans-serif; font-size: 10pt;"> 2012 Contingent Valuation: A Practical Alternative when Prices Aren’t Available. //The Journal of Economic Perspective// 26(4):27-42.

<span style="font-family: Arial,sans-serif;"> Meeks, Justin <span style="font-family: Arial,sans-serif;"> 2012 “Willingness-to-Pay for Maintenance and Improvements to Existing Sanitation Infrastructure: Assessing Community-Led Total Sanitation in Mopti, Mali” MS Thesis, Department of Civil and Environmental Engineering, University of South Florida

<span style="font-family: Arial,sans-serif; font-size: 10pt;"> Mitchell, Robert Cameron and Richard T. Carson <span style="font-family: Arial,sans-serif; font-size: 10pt;"> 1989 Using Surveys to Value Public Goods: The Contingent Valuation Method. New York: Resources for the Future.

<span style="font-family: Arial,sans-serif;"> Venkatachalam, L <span style="font-family: Arial,sans-serif;"> 2004 The contingent valuation method: a review. Environmental Impact Assessment Review 24: 89-124

<span style="font-family: Arial,sans-serif;"> Wedgewood, Alison and Kevin Sansom <span style="font-family: Arial,sans-serif;"> 2003 Willingness-to-pay surveys: A streamlined approach. Water, Engineering and Development Centre, Loughborough University

<span style="font-family: Arial,sans-serif;"> Whittington, Dale <span style="font-family: Arial,sans-serif;"> 1998 Administering Contingent Valuation Survey in Developing Countries. World Development <span style="font-family: Arial,sans-serif;"> 26(1):21-30

<span style="font-family: Arial,sans-serif;"> Whittington, Dale, John Briscoe, Xinming Mu, William Barron. <span style="font-family: Arial,sans-serif;"> 1990 Estimating the Willingness to Pay for Water Services in Developing Countries: A Case Study of the Use of Contingent Valuation Surveys