Grantee Research Project Results
Final Report: Experimental Tests of Provision Rules in Conjoint Analysis for Environmental Valuation
EPA Grant Number: R830820Title: Experimental Tests of Provision Rules in Conjoint Analysis for Environmental Valuation
Investigators: Taylor, Laura , Boyle, Kevin , Morrison, Mark
Institution: Georgia State University
EPA Project Officer: Hahn, Intaek
Project Period: May 1, 2003 through August 31, 2006
Project Amount: $126,804
RFA: Decision-Making and Valuation for Environmental Policy (DMVEP) (2002) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
In recent years there has been increasing use of conjoint analysis for environmental valuation. With conjoint analysis applications, respondents are simply asked to reveal their preferences – no published conjoint applications explicitly or implicitly describe a provision rule. However, developments of an incentive compatible provision rule has been an important part of the contingent valuation (CV) literature. Information describing the “market” or provision rules for public goods have proven to be an important element of survey design affecting value estimates. This research has three main objectives: to (1) examine the impacts of including individual and group provision rules in conjoint surveys involving both public and private goods; (2) examine whether provision rules reduce the divergence between stated and revealed preference value estimates and (3) explore whether the reason for the observed divergence between nonmarket values generated by CV and conjoint analysis is due to the absence of decision rules in conjoint surveys.
Summary/Accomplishments (Outputs/Outcomes):
This research reports results from 2,165 choice surveys administered in 10 different treatments. Results indicate that explicit discussions of provision rules do have a significant impact on choices subjects make in attribute-based choice surveys. We examine the impacts of provision rules in choice survey using both public and private goods. Four provision rules are considered: one in which choices are binding and an incentive-compatible provision rule is used (PPM_B); one which describes the incentive-compatible provision rule, but choices are not binding (PPM); one in which a non-incentive compatible provision rule is described that mimics a plurality rule mechanism (PRV); and one in which no provision rule is described (NPR).
Overall, the results indicate that the rules of exchange have a significant impact on the revelation of preferences through choice surveys. The Plurality Rule Voting (PRV) provision rule is of particular interest because it is the mechanism that choice surveys involving public goods might naturally mimic. Unfortunately, choices made when a PRV mechanism is in place are significantly different than those in potentially incentive compatible circumstances. This is perhaps not surprising given a plurality rule vote is known to be non-incentive compatible. Importantly, subjects were less price responsive in the PRV treatment as compared to potentially incentive compatible treatments, indicating that marginal values for environmental attributes derived from these surveys would be larger.
Interestingly, the results in the PRV treatments are not significantly than in the treatment in which no rules of exchange are described (no provision rule, or NPR). This latter treatment is clearly inconsequential; subjects are simply asked to make choices and reveal their most preferred alternative, with no other information other than being told they were participating in “a study about the decisions people make regarding the environment” or “an experiment in consumer decision making.” Because there are no descriptions of how an exchange or provision of the good might occur in the NPR treatment, we cannot even speculate on the incentive properties of such a survey. We simply note the curiosity that there are not significant differences in the behavior in PRV versus NPR, and note that this result is not positive for nonmarket valuation of public goods with choice-based surveys. For many public goods, a public voting rule is a natural provision rule. Theory suggests that a survey with majority rule voting is not incentive compatible, and our empirical results support this contention. Both market shares and marginal values are likely to be overestimated in this type of survey.
Our results are more promising for nonmarket valuation surveys design to value public goods which are divisible and might lend themselves to a private market-like provision rule. For public goods in which there can be established a credible provision mechanism in which survey respondents can individually increment the public good through their choices, hypothetical surveys may be capable of eliciting unbiased marginal values for attributes of the goods. This result is tenuous, as it is sensitive to model choice and object of choice in the survey. Lusk and Schroeder (2004) find similar results which are also not unequivocal. These are nonetheless provocative results worth further investigation. It could indicate that, under some circumstance, unbiased values for nonmarket good attributes may be available through the use of conjoint analysis with explicit discussions of provision rules. Market shares will not be unbiased however, which is important if this information is used to determine aggregate WTP.
References:
Lusk JL, Schroeder TC. Are Choice Experiments Incentive Compatible? A Test with Quality Differentiated Beef Steaks. American Journal of Agricultural Economics 2004;86(2):467-482.
Journal Articles:
No journal articles submitted with this report: View all 10 publications for this projectSupplemental Keywords:
Decision making; conjoint analysis; nonmarket valuation; survey; preferences; public good; willingness to pay,, RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Economics, decision-making, Social Science, Economics & Decision Making, contingent valuation, multi-objective decision making, policy analysis, surveys, decision analysis, cost benefit, environmental values, market valuation models, non-market valuation, standards of value, public policy, willingness to pay, conjoint analysisRelevant Websites:
http://www4.ncsu.edu/~lotaylor/ Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.