Grantee Research Project Results
Final Report: A Comparison of Direct Methods for Valuing Environmental Policies: A Case Study in New Hampshire's White Mountains
EPA Grant Number: R825824Title: A Comparison of Direct Methods for Valuing Environmental Policies: A Case Study in New Hampshire's White Mountains
Investigators: Halstead, John M. , Hill, L. Bruce , Stevens, Thomas H.
Institution: University of New Hampshire , University of Massachusetts - Amherst
EPA Project Officer: Chung, Serena
Project Period: October 1, 1997 through September 30, 1999 (Extended to December 31, 2000)
Project Amount: $159,071
RFA: Decision-Making and Valuation for Environmental Policy (1997) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
It has been speculated that deregulation of the electric industry in New Hampshire and other states throughout the country may lead to a degradation in air quality levels in the northeastern United States and other regions. This study examines means of determining how one aspect of air quality change?visibility?affects consumer surplus and the regional economy, and provides a direct comparison between two of the primary methods of direct valuation, the contingent valuation method (CVM) and conjoint analysis (CA).The objectives of the study were to: (1) compare and contrast two frequently used methods of valuation of nonmarket commodities such as visibility (the CVM and CA) to provide insight into which (if either) might be the more appropriate technique to address the problem at hand; (2) derive estimates of the impacts of visibility changes in the White Mountains region of New Hampshire on visitors to the region; and (3) use these estimates to determine part of the potential economic impact of deregulation of the electric industry in New Hampshire.
Case Study Methods: A case study of visibility in the Great Gulf Wilderness in New Hampshire was undertaken during 1999?2000. Visibility at the study area, which is about one-quarter mile northeast of the Mt. Washington summit, is commonly impaired by regional haze that is largely a product of fossil fuel energy production ( Hill, et al., 2000). Four surveys were used to measure the value of visibility in the region:
- Onsite by a trained interviewer using a laptop computer to present respondents
with computer-modeled images derived from the WinHaze Visual Air Quality Program.
This program allowed us to hold weather conditions constant while changing
visibility only.
- Offsite to individuals residing in the Northampton/Amherst area in western
Massachusetts (about a 3- to 4-hour drive from the study site).
- Mail survey of a random sample of 1,000 New England residents.
- Mail survey of a random sample of residents of New Hampshire, Vermont, and Maine.
A split sampling approach was employed throughout. In each of the intercept surveys, one-half of the respondents received a CVM question that asked for their Willingness to Accept reduced visibility in exchange for lower electricity bills. The other respondents were asked to rate, on a scale of 1 to 10, the status quo and a scenario with less visibility and lower monthly electricity bills. The first mail survey was modeled after the intercept surveys, except that it was possible to confront respondents with multiple scenarios of visibility degradation in eliciting Willingness to Accept measures via the electric bill vehicle. The second mail survey switched elicitation procedures and sought to estimate respondents' Willingness to Pay (using the CVM) to avoid degradation of visibility.
Conjoint Analysis:
- How would you rate the situation in photograph A on a scale of 0 to 10,
with 0 being totally unacceptable and 10 indicating that you would definitely
be willing to accept this level of visibility along with no change in your
monthly electric bill?
- How would you rate the situation in photograph B on a scale of 0 to 10, with 0 being totally unacceptable and 10 indicating that you would definitely be willing to accept this level of visibility along with a $x decrease in your monthly electric bill?
Contingent Valuation (Willingness to Accept): Would you be willing to accept this new level of visibility (indicated by picture B) in the White Mountain National Forest if your monthly electric bill were reduced by $x?
Contingent Valuation (Willingness to Pay): Would you be willing to pay $x per month more for electricity to avoid this new level of visibility (indicated by picture B) in the White Mountain National Forest?
In all cases, picture A?which represented the base scenario or status quo?described the average visibility level at the site during the summer months. Picture B represented one of four visual range reductions. The electric bill reduction was 20 percent of the respondent's total monthly bill in the personal survey and one-fourth, one-third, or one-half of the monthly bill for the first mail survey respondents (20 percent is the average savings expected from deregulation), while respondents to the second mail survey were confronted with bids ranging from $10 to $50 per month (these values were chosen based on the initial year surveys).
Double-wave mailings with postcard followups were used in each mail survey. Response rates were approximately 36 percent for the Willingness to Accept survey and 39 percent for the Willingness to Pay survey. These response rates are disappointingly low and raise the issue of nonresponse bias.
Summary/Accomplishments (Outputs/Outcomes):
Median Willingness to Accept estimates are provided in Table 1. Regarding the objectives of this research, the findings that emerge from this study can be summarized as follows. First, the CVM and CA models can produce very different results. In this study, the difference seems to be a result of the criterion used to define a "yes" response in the conjoint format. Twenty percent of conjoint respondents were "willing to accept" the tradeoffs presented in this study when a "yes" response was defined as B>A; 9 percent of conjoint respondents were "willing to accept" if the criteria is B>A; and only about 3 percent indicated that they would definitely accept (B=10 and B?A). Results from the Willingness to Pay CVM survey also suggest that median Willingness to Pay value estimates are very sensitive to whether or not respondent uncertainty is incorporated in the analysis. We therefore believe that future studies should include tests for sensitivity to the valuation question format and to respondent uncertainty.
Table 1. Visibility value estimates: median Willingness to Accept per montha | ||
Model 1
(B>A) |
Model 2
(B>A) |
|
Average Respondent |
$924
|
$1,006
|
NH Resident, No Visits Planned |
$36
|
$17
|
Average Respondent, No Visits Planned |
$154
|
$162
|
Average Respondent, Conjoint Model |
$2,790
|
b
|
Average Respondent, CVM Model |
$331
|
b
|
a
Values rounded to nearest dollar. b Conjoint dummy variable not different from zero. |
As noted above, we believe that conjoint responses rating B>A are conceptually consistent with "yes" responses in the traditional CVM. Our empirical estimates suggest no difference between conjoint and CVM in this case. However, the problem of hypothetical bias suggests that "yes" responses should be defined by B>A. When this was done, resulting estimates derived from the conjoint format were much different from those derived via CVM.
Regarding the second (policy-related) objective, most respondents were not willing to accept cheaper electricity in exchange for reduced visibility over the range examined in this study. In fact, the estimated economic value of visibility suggests that compensation for improved visibility via lower priced electricity is simply not feasible; the necessary compensation is likely to be greater than the average respondent's actual electricity bill. If respondents are well informed, we might therefore infer that deregulation will not result in a substantial increase in pollution as a result of greater household demand for the cheapest source of electricity.
Other results of interest to researchers in the stated preference area include:
- Effects of location appear to be more complex than previously imagined.
When asked for Willingness to Accept, respondents living nearby valued
visibility less than those living further away, ceteris paribus. Location
was not a statistically significant factor in the Willingness to Pay
models. Those planning future visits were much less likely to accept reduced
visibility and were more likely to choose Willingness to Pay to avoid
reduced visibility.
- Median value estimates differed dramatically depending on whether a Willingness
to Accept or Willingness to Pay format was employed. Although this
result was expected, the magnitude of the difference was not.
- No differences were associated with whether the valuation question was conducted
by mail or in person. Perhaps the National Oceanic and Atmospheric Administration
(NOAA) guidelines requiring personal interviews should continue to be reevaluated.
- Despite survey pre-tests and careful wording of the valuation question, many respondents valued air pollution in general. Consequently, the value of visibility may be overestimated. A CA that includes several attributes of pollution, including visibility, might clarify this issue, but the problem of sensitivity of this method to the definition of "yes" responses is likely to remain an issue.
References:
Elkstrand E, Loomis J. Estimated Willingness to Pay for protecting critical habitat for threatened and endangered fish with respondent uncertainty. In: Englin J, compiler. Tenth Interim Report, W-133 Benefits and Costs Transfer in Natural Resource Planning, University of Reno, Reno, NV, 1997.
Harper WJ. A comparison of direct methods for valuing environmental amenities: a case study of the White Mountain National Forest. Ph.D. Dissertation, University of New Hampshire, 2000.
Hill LB, Harper WJ, Halstead JM, Stevens TH, Porras I, Kimball K. Visitor perceptions and valuation of visibility in the Great Gulf Wilderness, New Hampshire. In: Proceedings: Wilderness Science in a Time of Change. Ogden, UT, U.S. Department of Agriculture Forest Service, Rocky Mountain Research Station, 2000.
Stevens TH, Belkner R, Dennis D, Kittredge D, Willis C. Comparison of contingent valuation and conjoint analysis in ecosystem management. Ecological Economics 2000;32(1):63-74.
Wang H. Treatment of "don't know" responses in contingent valuation surveys: a random valuation model. Journal of Environmental Economics and Management 1997;32(2):219-232.
Journal Articles:
No journal articles submitted with this report: View all 19 publications for this projectSupplemental Keywords:
air quality, valuation, benefits, conjoint analysis, contingent valuation, visibility, RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Geographic Area, State, Economics, decision-making, Ecology and Ecosystems, Social Science, Economics & Decision Making, Psychology, New Hampshire (NH), air pollution policy, contingent valuation, ecosystem valuation, empirical validation, community involvement, direct valuation method, electric industry deregulation, environmental values, New Hampshire's White Mountains, non-market valuation, nonmarket choice, psychological attitudes, public values, benefits assessment, conjoint analysisProgress 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.