ISSUES IN NONMARKET VALUATION AND POLICY APPLICATION: A RETROSPECTIVE GLANCE

Walsh, Johnson, and McKean

Issues in Nonmarket Valuation 183

Table 2. Description of Variables in the Analysis

Name	Definition OfVariable
Dependent Variable	Consumer surplus estimated by each study, standardized to average values per activity day, adjusted to third quarter 1987 dollars.
Site Quality	Qualitative Variable = 1 if site was rated by each study as uniquely high quality; O if medium or low.
Forest Service Administered Mixed Public & Private Sites	Qualitative Variable = 1 if the study sites were Forest Service administered; O if otherwise. Qualitative Variable = 1 if household survey of participants in an activity at public and private sites; O if otherwise (the omitted categories were other wholly public and wholly private).
Specialized Activity	Continuous variable = percent. Proportion of total recreation use of U.S. Forest Service resources in the activity category. Proxy of taste and preference for specialized vs. gener- alized activities.
Inflationary Adjust- ment Sample Coverage	Qualitative Variable = 1 if data were collected for each study prior to 1980; O if 1980-1988. Qualitative Variable = 1 if only in-state residents were included in the sample of users; O if out-of-state residents were also included.
Method Substitution	Qualitative Variable = 1 if CVM; O if TCM or other method. Qualitative Variable = 1 if a substitute price term was included in the TCM demand speci- fication; O if otherwise.
Travel Time	Qualitative Variable = 1 if travel time cost was omitted in the TCM demand specification; O if time was included.
Individual Observation Household Production & Hedonic Price Open-ended Question	Qualitative Variable = 1 if TCM sample units were individual observations; O if otherwise. Qualitative Variable = 1 if household production or hedonic price TCM procedure; O if otherwise (the omitted category was the zonal group approach). Qualitative Variable = 1 if noniterative open-ended question was asked in a CVM; O if otherwise.
Dichotomous Choice Question Socioeconomic Characteristics Recreation Activity	Qualitative Variable = 1 if dichotomous choice CVM question was used; O if otherwise (the omitted category was the iterative question). Proxy for socioeconomic characteristics of participants in the service area of the study site. The nine Forest Regions are qualitative variables. Alaska is the omitted region. The 19 national recreation use categories are potential qualitative variables for activities. Omitted categories include activities with limited representation in the studies, i.e., resorts, cabins, and organized camps.

whether public or private. A specialized activ-
ity variable tests the hypothesis that benefits
are lower for general activities than for spe-
cialized activities. This may be interpreted as
a proxy for taste and preference. The federal
guidelines (Water Resources Council 1983) dif-
ferentiate between general recreation activities
engaged in by a large number of persons and
specialized recreation limited to fewer partic-
ipants with unique preference patterns. The
guidelines associate specialized recreation with
higher unit-day values than general recreation.

An inflationary adjustment variable is in-
tended to begin examining the question of
whether recreation values increase at the same
rate as changes in the purchasing power of the
dollar. For comparison purposes, the reported
values must be adjusted for inflation. How-
ever, this is equivalent to assuming constant
real prices which would not be consistent with
increased crowding and relative scarcity of
natural resources available for resource-based
recreation activities (President’s Commission
on Americans Outdoors). Moreover, the pro-
cedure assumes an equal proportional change
in the reported values for any given year which
tends to dampen (enlarge) the absolute dollar
adjustment for studies reporting low (high)
values. This is evident for surveys from 1968-
79 when the inflation rate was 6.9^o∕o compared
to 4.8^o∕o from 1980-87. Finally, willingness to
pay is, in part, a function of ability to pay
which suggests that secular adjustments for per
capita real income would be useful.

A method variable is included to test the
hypothesis that intended willingness-to-pay
estimates of the CVM are lower than behavior-
based TCM. This would be consistent with the
observation that TCM values the entire trip
including the primary activity and secondary
activities while the CVM usually values the
primary activity alone. For example, TCM al-
ways values the entire time on site per calendar
day of a trip while CVM usually values only

More intriguing information

1. Top-Down Mass Analysis of Protein Tyrosine Nitration: Comparison of Electron Capture Dissociation with “Slow-Heating” Tandem Mass Spectrometry Methods
2. The name is absent
3. A Critical Examination of the Beliefs about Learning a Foreign Language at Primary School
4. Brauchen wir ein Konjunkturprogramm?: Kommentar
5. Gender and headship in the twenty-first century
6. Knowledge, Innovation and Agglomeration - regionalized multiple indicators and evidence from Brazil
7. Who is missing from higher education?
8. Nurses' retention and hospital characteristics in New South Wales, CHERE Discussion Paper No 52
9. The name is absent
10. On the Relation between Robust and Bayesian Decision Making