Valuing Water Quality in Midwestern Lake Ecosystems: Temporal Stability and the Role of Information in Value FormationEPA Grant Number: R830818
Title: Valuing Water Quality in Midwestern Lake Ecosystems: Temporal Stability and the Role of Information in Value Formation
Investigators: Herriges, Joseph A. , Kling, Catherine L. , Zhao, Jinhua , Downing, John
Institution: Iowa State University
EPA Project Officer: Hahn, Intaek
Project Period: September 15, 2002 through September 14, 2005
Project Amount: $444,782
RFA: Decision-Making and Valuation for Environmental Policy (DMVEP) (2002) RFA Text | Recipients Lists
Research Category: Environmental Justice
In this project, economists and ecologists will assemble an extensive panel data set of revealed and stated preferences concerning Iowa lakes in conjunction with an ongoing ecological effort to collect detailed water quality data on these same lakes in overlapping years. The data collected will contain the information needed to estimate both stated and revealed preference models that can be used in turn to value changes in lake water quality.
By gathering recreational use and stated preferences over a four year period from the same individuals, we will generate a unique panel data set to investigate the responsiveness of use patterns to objective and subjective measures of water quality and to test to the consistency of values over time. Variations in the information set provided to survey respondents will also allow us to test the role of expected future information in a subject's willingness-to-pay (WTP) and how it relates to the individual's compensating variation.
Using this data, we will develop and estimate time series mixed logit models (Train, Herriges and Phaneuf) to test both the temporal stability of parameters and to investigate the relationship between objective water quality variables (as collected by ecologists) and usage of the resource. The ability of the mixed logit model to capture individual preference heterogeneity is of significant value in modeling multiple year observations from the same individual. This feature is further exploited in the model used here by explicitly incorporating changing information sets over time into preference formation. Stability of preferences over time and the accuracy of predictions from single-year static models and contingent behavior responses will be assessed in this framework as well.
The same set of surveys will collect stated preference information to test hypotheses about the dynamic formation of values, particularly as these values related to information and the ability to delay and/or reverse the decisions. Stated preference questions that reveal differing amounts and quality of information to respondents will be administered to stratified samples to form the basis for hypothesis testing. By identifying the wedge between the stated WTP and a subject's expected compensating variation, the testing results will aid us in designing surveys that convey appropriate information and in estimating the expected compensating variation from the SP data.
The expected results of this research include: (1) improved understanding of the factors causing wide gaps between WTP and WTA measures of environmental quality, (2) the degree to which recreation demand parameters and preferences are stable across time, (3) improved understanding of how perceptions of water quality relate to physical measures of water clarity and the general health of ecosystems, and (4) how information provided to survey respondents affects their perceptions of water quality and recreational behavior.