Experimental Tests of Provision Rules in Conjoint Analysis for Environmental ValuationEPA Grant Number: R830820
Title: 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
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 how a good will be provided (the provision rule). However, development 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 has 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 non-market values generated by CV and conjoint analysis is due to the absence of decision rules in conjoint surveys.
First, with a private good, conjoint choice data will be collected using an incentive-compatible, individual provision rule involving real payments and purchase. These results will be compared to a treatment using an individual provision rule but with hypothetical payment. This allows exploration of (1) the differences in choices due to hypothetical versus real provision. Next, hypothetical private-good treatments are conducted where either a group provision rule or no provision rule is used. This will allow us to determine (2) the effects of moving away from incentive compatibility and (3) whether the results from the treatment with no decision rule converges on the results from the incentive-compatible decision rule or the group decision rule. Each of these treatments will then be repeated in public good experiments, where the possibility of free riding may alters subjects' incentives relative to the private good treatments. In addition, for the public goods experiments, a hypothetical CV treatment will be included to allow testing of whether the inclusion of provision rules causes convergence of the results of the two different techniques. A total of 1,600 student subjects will be used in the experiments. Data will be analyzed using random utility models.
This study will demonstrate the effects of information regarding provision rules on the incentive-compatibility of conjoint surveys. Preliminary testing indicates that the inclusion and choice of provision rules has a significant and non-trivial effect on subjects' choices in these types of surveys. The results will be used to develop more credible and defensible conjoint surveys for valuation purposes. The beneficiaries of this research will include all stakeholders and decision makers who will be able to produce more accurate and defensible non-market value estimates for cost/benefit analysis involving natural resources as well as natural resource damage assessment. Improving valuation and assessment of environmental and natural resources will also improve the cost-effective allocation of federal and state environmental protection resources.