1 / 0

Stated Preference Methods - Introduction

Stated Preference Methods - Introduction. Introduction to stated preference methods This class: Contingent valuation method (CVM) Steps in a CVM study What are the pros and cons of using CVM? Worked example using a case study from Nepal Discussion paper by Loomis et al. (Nelly)

conroy
Download Presentation

Stated Preference Methods - Introduction

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Stated Preference Methods - Introduction Introduction to stated preference methods This class: Contingent valuation method (CVM) Steps in a CVM study What are the pros and cons of using CVM? Worked example using a case study from Nepal Discussion paper by Loomis et al. (Nelly) Next class: Choice experiment method (CE)
  2. Contingent Valuation Method (CVM) Ask people directly about the values they place on environmental resources Elicit willingness to pay(WTP) or to accept compensation (WTA) for changes in environmental quality Use a survey questionnaire and typically one of four elicitation approaches (response formats) Requires a payment vehicle (e.g. tax, membership fee, entry charge, donation, etc.)
  3. Steps in a Contingent Valuation Study Step 1: Setting up the hypothetical market Step 2: Obtaining bids Step 3: Estimating WTP Step 4: Aggregating the data Step 5: Evaluating the CVM exercise (validity) Hanley and Barbier (2009)
  4. Set up the hypothetical market Establish a reason for the payment based on the policy problem of interest Determine the information package to be provided to respondents and the status quo Decide on the payment vehicle: it may be a tax, trust fund contribution, membership fee, entry charge, etc. Design the questionnaire with careful placement of the WTP question (middle?) Pre-test using focus groups, strategic sampling, etc. and revise as needed
  5. Obtain bids Survey is administered using one of the following: face-to-face (most preferred) by mail (low response rate, e.g. 10%) by telephone (poor at conveying contextual information), or internet (increasingly popular). Bids are obtained using one of these response formats: Open-ended payment - difficult for respondents when they have no experience with good or service Payment card - provides a range of payment options and allows respondents to fix their own responses within the range Iterative bidding - individuals are asked about specified payments and then these until the individual choses an amount Dichotomous choice or referendum approach - payments are randomly selected from a list; respondent then "votes" yes or no.
  6. Estimate WTP When the first three bid approaches are used, average WTP/WTA is easy to obtain Bid curves that try to "explain" WTP can be estimated using explanatory variables from the survey, e.g. income, age, education With open ended questions this is especially straightforward If dichotomous choice is used then estimating WTP and bid curves is more difficult For DC questions the logitor probit models are used to estimate the bid curve WTP is then calculated using special formulas
  7. Aggregating Results to the Population Choose the relevant population Adjust from sample mean WTP to population mean WTP; best to use the bid curve if the sample is not representative Discounting may be required if expressing the WTP values over a period of time or space
  8. Evaluating CVM .. CVM has several advantages: Measuring non-use values Including all economic values (TEV, potentially) However, we make assumptions when asking people WTP questions: Hold values in advance or can easily generate them Have sufficient information and understanding of what they are valuing Can decide (alone) on the values they attribute to ecosystems Value consistently Value according to individual rationality
  9. But several disadvantages related to bias: Blue Ribbon panel evaluated CVM and made a number of recommendations for ‘good practice’
  10. Worked Example: Assessing Prospects for Community-based Management in Nepal
  11. Introduction to Nepal Case Study Community-based wildlife management (CBWM) has had mixed success …(Songorwaet al. 2000) More social capital argued to increase prospects for successful CBWM (Pretty & Smith 2004) Others argue that presence of leadership types important as well (Ostrom 1990) We investigate these influences, arguing they are complex and conflated …not simple Use a case study from the Sherpa (Khumbu) region of Nepal, involving Himalayan musk deer, to investigate interest in CBWM and role of leadership
  12. Khumbu Region of Northeastern Nepal
  13. Phortse Village, Khumbu Region, Nepal
  14. Social Capital and Leadership Characteristics of Phortse Villagers Surveyed 45 of 77 households in Phortse (38 usable responses) Asked questions about various social capital and leadership attributes Key social capital variables were used in a factor analysis (PCA) Resulted in 2 distinct groupings of social capital attributes (leaders vs. others) Households were clustered using the two factors
  15. Social Capital Characteristics for Clustered Households, PhortseVillage Legend: *** p  0.01; ** p  0.05; * p  0.1
  16. Demographic and Economic Characteristics of Leadership Cluster, Phortse Village Respondent is younger than 55 years of age Respondent is literate with some education Household has more than four members At least one member lives outside the village Household grows more than one field of grain Household owns a tourism business
  17. Himalayan Musk Deer (Moschos chrysogaster)
  18. Contingent Management Question: Musk Deer Study In this section, I will describe to you a hypothetical community project that could be a possibility in the future but is not planned right now, and then I will ask you several questions about your willingness to participate in this project. Please judge the costs and benefits of this proposed project in relation to your household’s present circumstances. We would appreciate it if your answers are as realistic as possible. The project is concerned with conserving wildlife in the vicinity of your village by allowing your community to receive a direct benefit from it. It involves protecting musk deer and the collection of musk from male deer using a live capture technique. In considering this project please suppose that the government changed the law to make musk extraction legal. To collect the musk, 10 to 20 members of the community would work together to capture each deer by driving it into large mesh nets. The musk would be removed from the live animal without harming it and then sold through a legal marketing agency. Each household that participates would be expected to provide about 15 days of labour per year, mostly from adult males. Most labour would need to be provided between Dawa Guwa and Dawa Chuniwa for collecting the musk so please consider your availability at this time of year. The remaining labour contribution would be needed at other times to patrol the forest and guard against poachers. Each participating household would receive an equal payment from the sale of musk. At this time, we don’t know exactly how much this payment would be so we would like to play a game where you select a possible payment and then answer the following question as truthfully as possible. Assume that no food is provided when responding. Amount selected ____________ Rs/year Given the payment shown on the slip of paper you have chosen, would your household be interested in participating in the musk deer community project described above?
  19. A Model of Household Participation in Community-based Wildlife Management (CBWM) Following Cooper and Keim (1996): where: j is a binary variable indicating willingness to participate in CBWM (yes/no) y is income d is a set of demographic variables s is a set of social capital, leadership and related attributes εj is a stochastic error term
  20. Household’s decision to participate depends on size of the payment (R) from the CBWM project: Expressing this in probability terms, we get: This relationship can be estimated as a binary probit model Used a CVM type question describing a hypothetical CBWM project with a random payment; respondents can choose to participate or not
  21. Variable Definitions
  22. Musk deer probit model results (I)Dependent variable is: “willingness to participate in project” (Yes/No)
  23. Musk deer probit model results (II)Dependent variable is: “willingness to participate in project” (Yes/No)
  24. Actual & Predicted Outcomes for “Willingness to Participate”, Probit Regression (Model II)
  25. Calculating Median WTA for the Average Individual in the Sample
  26. Estimating WTA Use the formula for Median WTP with covariates Calculation is as follows: {-4.489 + (7.141 x 0.384) + (-1.584 x 0.316)}/0.0005 = 4495 Values are in NRs (about 35NRs to 1.00 USD) This compares to a mean offer in the survey of 7500 NRs
  27. Concluding Observations We used a novel ‘willingness to participate’ approach that can be used to assess potential payments required Social Capital and leadership characteristics are important influences on ‘willingness to participate’ Presence of leader types is important for successful CBWM, but must be interested … We might ask whether a musk deer CBWM project might succeed in Phortse … Clearly, economic considerations are important too
  28. Discussion groups Which of the various types of bias associated with CVM do you think is likely to be most serious? Does this depend on the nature and context of the valuation problem being studied? If so, how?
More Related