Preference Formation and Elicitation in Valuing Non-Market Goods
The general research objective of this interdisciplinary project is to investigate the interaction between value formation and value elicitation. The basic premise is that an understanding of how individuals form environmental values (e.g., purchase versus contribution model) cannot be decoupled from the value statement problem and the choice of elicitation mechanism. The corollary premise is that social context is an important, but relatively unexplored, determinant of both value formation and value statement.
Social context effects may be at work at both a larger level (in the attitudes and beliefs of respondents concerning particular environmental policies) and within the highly structured stated preference communication process. Once recognized, a variety of testable hypotheses can be generated. It is hypothesized that the multi-dimensional nature of many environmental policies may generate both positive and negative passive use values, that there may be significant social desirability response effects for some environmental policy changes, and that the relative magnitude of these effects may be influenced by individual characteristics, attitudes, and the methods employed for eliciting values. Likewise, it is hypothesized that nonresponse and protest response behavior may be influenced by individual characteristics, attitudes, and the elicitation mechanism.
The methods to be employed in this research will include a unique combination and sequencing of field surveys and laboratory experiments. The field research will be a series of telephone surveys, with some sample treatments augmented by informational mailings, to address valuation questions and to collect detailed demographic and attitudinal information. The field research will be integrated with a series of laboratory experiments on public goods provision. Both field research and lab experiments yield unique insights (the richness and variability of field data versus the highly controlled nature of the lab). The structure of this research is designed to combine these insights. For example, a comparison of actual versus hypothetical payment for an environmental public good will be made in the lab and validated against the econometric estimates of stated preferences in the field. The application of these methods will be targeted to environmental issues of southwestern ecosystems. While not tied to a particular policy case, the focus on protection of riparian areas and instream flows, and ecological restoration of rangelands will provide general policy relevance. Pilot studies have been initiated on these topics, and will be complemented by a set of focus groups to identify all important dimensions of the environmental goods.
The results of this research will aid in the assessment of current stated preference methods and guidelines, and expand the knowledge of the interaction between value formation and value elicitation. Improved understanding of social context will provide a clear advancement in the use and design of stated preference approaches. Important incremental gains will also be provided by empirical validation of the multidimensional nature of environmental values, identification of the presence and relative magnitude of social desirability effects, and the characterization and treatment of nonresponse and protest behavior.
|Brookshire, David S.|
|New Mexico, University of|
|October, 1995 - December, 1997|
Cost to Funding Agency:
|Project Status Reports:
The objective of this research project is to investigate the interaction between value formation and value elicitation. The project is based on the premise that an understanding of how individuals form environmental values cannot be decoupled from value statement problems and the choice of elicitation mechanism. The methods we are employing include a combination of focus groups, laboratory experiments, and telephone surveys.
The focus of the survey effort is to measure the nonmarket benefits of protecting instream flows. In February 1995, we conducted a dichotomous choice contingent valuation (CV) telephone survey using a voluntary contribution trust fund format, and replicated it in February 1996. Our analysis of the data involved sensitivity tests to detect a change in the scope of the good, and corollary tests for sensitivity to information about the collective nature of providing the good and the temporal reliability of the results. Using the pooled data, we test sensitivity to scope and the group-size reminder under alternative modeling assumptions. We use four parametric models and evaluate the results of estimates of mean and median willingness to pay (WTP) and interquartile ranges.
The evidence compiled to date supports sensitivity to scope. One policy caveat that should be noted is that estimates of mean WTP are extremely sensitive to the distributional assumption, while estimates of median WTP are much more conservative and stable. The evidence also supports insensitivity to the group-size reminder.
Of significance is the absence of evidence supporting the "contribution model." Further, our results suggest that telephone surveys may be credible as an alternative to in-person interviews for investigating particular issues in contingent valuation studies. In the future, we will perform a side-by-side comparison of the group-size reminder for open-ended and dichotomous choice formats. We have collected all necessary data, including an additional split-sample cross treatment.
The behavioral laboratory effort for this project is focused on understanding the individual decision process in valuation. Our first investigation found that the disparity between willingness to accept (WTA) and WTP is due to uncertainty concerning the payoff from the good. This uncertainty can arise from a variety of sources. Value uncertainty is when the good is unfamiliar to the respondent. Outcome uncertainty concerns whether the agency in question will be able to provide the good with the funds generated. We found that the disparity is a function of the level of uncertainty and can be mitigated when the purchase is reversible.
A second issue that we are investigating is the role of provision mechanisms on stated WTP values. Our laboratory work will investigate WTP under different public good provision mechanisms that can be implemented in surveys. Although data collection is not yet complete, the evidence suggests that individuals give different responses for the same good depending on the mechanism. Future laboratory work will investigate the role of uncertainty concerning payoffs on public good provision and stated WTP.
David S. Brookshire, Robert Berrens, Alok Bohara, Philip Ganderton, Hank Jenkins-Smith, Hillard Kaplan, Michael McKee, and Carol Silva, Progress Report: Preference Formain and Elicitaion in Valuing Non-Market Goods, Report to EPA, October 28, 1996.