Original article




                      A meta-analysis of wetland
                      contingent valuation studies
                      R. Brouwer á I.H. Langford á I.J. Bateman á R.K. Turner




                                    asked for either their willingness to pay (WTP) or their
                                    willingness to accept (WTA) compensation for the gains or
  Abstract There is growing interest in the potential
                                    losses involved (Mitchell and Carson 1989). Of these op-
  for producing generally applicable models for
                                    tions the WTP approach has become the most frequently
  valuing non-market environmental services which
                                    applied and has been given peer review endorsement
  do not rely upon expensive and time-consuming
                                    through a variety of studies (see, for example, Arrow et al.
  survey work, but rather extrapolate results from
                                    1993). When aggregated across those who will be affected
  previous studies. This paper presents a meta-
                                    by the suggested environmental changes, this stated WTP
  analysis for the use and non-use values generated by
                                    amount is used as a socio-economic indicator of the
  wetlands across North America and Europe. The
                                    environmental values involved.
  study assesses the socio-economic values
                                    Given the substantial indirect, often off-site, use and non-
  attributable to the hydrological, biogeochemical and
                                    use values involved, wetlands have been the focus of at-
  ecological functions provided by such complex
                                    tention in several CV studies (Crowards and Turner 1996).
  environmental assets. The clustering of multiple
                                    Many of these studies try to estimate the total economic
  values derived from single studies is examined
                                    value of wetlands. Total economic value, not to be
  through the application of multilevel modelling
                                    confused with total ecosystem value, consists of use and
  methods allowing for the hierarchical structure of
                                    non-use values (Pearce and Turner 1990). CV is the only
  such data.
                                    economic method to date that is able, in principle, to
                                    account for possible non-use motivations underlying
  Key words Meta-analysis á Contingent valuation á
                                    people's value statements. Whereas use values refer to the
  Wetlands á Ecosystem functions
                                    values associated with the actual use of the various goods
                                    and services wetlands provide, non-use values are unre-
                                    lated to any actual or potential use of these goods and
                                    services.
      Introduction                        Wetlands are complex hydro-ecological systems, whose
                                    structure provides us with goods or products involving
                                    some direct utilisation of one or more wetland character-
This paper addresses the socio-economic values of the
                                    istics, while wetland ecosystem processes provide us with
various functions performed by wetland ecosystems.
                                    hydrological and ecological services, supporting or pro-
Environmental economists have developed a variety of
techniques for measuring such values, of which the con- tecting human activities or human properties without
tingent valuation (CV) method is probably the most widely being used directly. The stock of wetlands is a multi-
applied in contemporary research. CV is a survey method functional resource with signi®cant economic value, as
                                    also has been suggested by Costanza et al. (1997). Fifteen
where individuals are presented with information about
                                    percent of the value of the world's ecosystem services and
speci®c environmental changes, and their perception,
                                    natural capital is generated by wetlands (Costanza et al.
attitudes and preferences regarding these changes are
                                    1997). However, all over the world countries have experi-
elicited. In order to measure the effect of the suggested
                                    enced severe wetland losses (Tolba and El-Kholy 1992;
changes on people's welfare, respondents are typically
                                    Turner 1992). Sustainable management of these assets is
                                    highly relevant. Since this management process is not
                                    costless, they require accurate and meaningful valuation in
                                    order to be able to weigh the costs and bene®ts of their
      Received: 24 February 1999 á Accepted: 6 June 1999
                                    conservation.
      R. Brouwer (8) á I.H. Langford á I.J. Bateman á R.K. Turner In this paper, the main ®ndings of CV studies of
      Centre for Social and Economic Research on the Global
                                    wetlands in temperate climate zones in developed
      Environment (CSERGE), School of Environmental Sciences,
                                    economies will be investigated. The main objective is
      University of East Anglia, Norwich, UK
                                    to quantify the socio-economic values associated with
      Tel.: 01603 593741; fax: 01603 593739,
                                    wetland ecosystem functioning in a meta-analysis of
      e-mail: r.brouwer@uea.ac.uk




                                                                        47
                                   Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
    Original article



                                        to increase policymaker demand for transferable valuation
   wetland CV studies, supplementing qualitative analyses
                      È                  results.
   provided, for example, by Gren and Soderqvist (1994)
                                        Meta-analysis enables researchers to identify criteria for
   or Crowards and Turner (1996). Natural and social
                                        valid environmental value transfer1 or to test the conver-
   science are brought together by relating the various
   hydrological, biogeochemical and ecological wetland            gent validity of value estimates. In the ®rst case the data-
   functions to the societal bene®ts derived from these            set is entirely used to determine the factors that help to
   functions and the socio-economic values attached              signi®cantly explain variances in valuation outcomes. In
   to these bene®ts.                             the second case the data-set can be split, for example into
                                        two parts, one of which is used for the ®rst purpose and
                                        the other to test whether the value estimates based on the
                                        signi®cant factors fall within the con®dence interval of the
         Approach                           other half's estimates.
                                        Environmental value transfer is commonly de®ned as the
                                        transposition of monetary environmental values estimated
   The results from 30 different CV studies of wetlands in
                                        at one site (study site) through market-based or non-
   temperate climate zones in developed economies were
                                        market-based economic valuation techniques to another
   compared and synthesised in a meta-analysis. Only very
                                        site (policy site). The most important reason for using
   few tropical wetland valuation studies exist (Barbier 1993).
                                        previous research results in new policy contexts is cost
   Tropical wetland studies are excluded from the
                                        effectiveness. Applying previous research ®ndings to
   meta-analysis presented here because of the enormous
                                        similar decision situations is a very attractive alternative
   differences between population samples in developed and
                                        to expensive and time-consuming original research to
   developing countries regarding socio-cultural and demo-
                                        quickly inform decision making.
   graphic-economic characteristics.
                                        The criteria for selecting studies for environmental value
   Since the beginning of the 1990s, meta-analysis has been
                                        transfer suggested in the literature focus on the environ-
   playing an increasingly important role in environmental
                                        mental goods involved, the sites in which the goods are
   economics research (van den Bergh et al. 1997). Originally
                                        found, the stakeholders and the study quality (Desvousges
   a technique used in experimental medical treatment and
                                        et al. 1992). However, very little published evidence exists
   psychotherapy, meta-analysis is the statistical evaluation
                                        of studies that test the validity of environmental value
   of the summary ®ndings of empirical studies, helping to
                                        transfer. Moreover, in the few studies that have been
   extract information from large masses of data in order to
                                        carried out, the transfer errors are substantial (Brouwer
   quantify a more comprehensive assessment. It enables
                                        1998).
   researchers to explain differences in outcomes found in
                                        As more information about factors in¯uencing environ-
   single studies on the basis of differences in underlying
                                        mental valuation outcomes becomes available, for in-
   assumptions, standards of design and/or measurement. As
                                        stance through the meta-analysis presented here,
   such, meta-analysis is an important extension of quanti-
                                        transfers across populations and sites become more
   tative analyses and can be seen as a supplement to quali-
                                        practicable.
   tative analysis.
   Compared to qualitative analysis, important advantages
   of meta-analysis are that on the `input' side it does not
   prejudge research ®ndings on the basis of the original
                                              Dataset and study characteristics
   study's quality, while it avoids a differential subjective
   weighting of studies in the interpretation of a set of
   ®ndings on the `output' side (Glass et al. 1981). How-           The list of wetland CV studies included in the meta-
   ever, one drawback is that it may be biased towards            analysis is presented in Table 1. Most studies have been
   including signi®cant study results only, since it may well         published in journals. Half of all studies were carried
   be that insigni®cant study results will not be published.         out between 1985 and 1989, with most being published in
   Furthermore, multiple results from the same study are           the ®rst 3 years of the 1990s. One study was carried
   often treated as individual, independent observations           out in the 1970s, 19 in the 1980s and 10 in the 1990s.
   without explicit testing for intra-study correlation (Wolf         Besides the inclusion of published signi®cant results,
   1986).                                   Table 1 illustrates two other problems in this
   In the ®eld of environmental valuation, meta-analyses
   have focused on a range of environmental issues from
   outdoor recreation to urban air pollution, based on single
                                        1
                                         The term `environmental value transfer' is used here instead of
   or multiple valuation techniques. The increase in meta-
                                        the popular term `bene®ts transfer', because CV can also measure
   analytical research seems to be triggered principally by (1)
                                        the bene®ts foregone, which makes the estimated values costs
   increases in the available number of environmental valu-
                                        instead of bene®ts. WTP is the conventional economic approach
   ation studies, (2) the seemingly large differences in valu-        to measure environmental values in money and hence make them
   ation outcomes as a result of the use of different research        commensurable with other market values ± costs and bene®ts ±
   designs (Carson et al. 1996), and (3) the high costs of          associated with decisions that have been made, are made or have
   carrying out environmental valuation studies which tend          to be made in the face of limited human and natural resources



48   Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
                                                               Original article



  Table 1
Studies (with year of publication in parentheses) included in the meta-analysis

                              Type of publicationa                               nb
   Authors                                                     Study year

1  Bateman et al. (1995)                 Journal article (EE)                   1991           3
2  Bergstrom et al. (1990)                Journal article (EE)                   1986           1
3  Bishop and Boyle (1985)                Consultancy report                    1985           2
4  Bishop et al. (1987)                  Journal article (TAFS)                  1985           7
5  Brouwer and Slangen (1998)               Journal article (ERAE)                  1994           3
6  Carson and Mitchell (1993)               Journal article (WRR)                  1983           3
7  Cooper and Loomis (1991)                Book chapter                       1987           3
8  Cummings et al. (1994)                 Journal article (AJAE)                  1992           2
9  Desvousges et al. (1987)                Journal article (JEEM)                  1981           21
10  Farber (1988)                     Journal article (JEM)                  1984           1
11  Garrod and Willis (1996)                Journal article (JEPM)                  1993           4
12  Green and Tunstall (1991)               Journal article (AE)                   1986           1
13  Greenley et al. (1981)                 Journal article (QJE)                  1976           4
14  Silvander (1991)                    Dissertation                       1989           2
15  Jordan and Elnagheeb (1993)              Journal article (WRR)                  1991           2
16  Kaoru (1993)                      Journal article (ERE)                  1989           1
17  Kosz (1996)                      Journal article (EE)                   1993           1
18  Lant and Roberts (1990)                Journal article (EPA)                  1987           6
19  Loomis et al. (1991)                  Book chapter                       1989           10
20  Loomis (1987)                     Journal article (WRR)                  1985           1
21  Olsen et al. (1991)                  Journal article (Rivers)                 1989           3
22  Phillips et al. (1993)                 Journal article (CJAE)                  1991           2
23  Sanders et al. (1990)                 Journal article (WRR)                  1983           2
24  Schultz and Lindsay (1990)               Journal article (WRR)                  1988           1
25  Spaninks (1993)                    MSc thesis                        1993           3
26  Spaninks et al. (1996)                 Scienti®c report                     1995           2
27  Sutherland and Walsh (1985)              Journal article (LE)                   1981           2
28  Whitehead and Blomquist (1991)             Journal article (WRR)                  1989           6
29  Willis (1990)                     Journal article (AE)                   1986           2
30  Willis et al. (1995)                  Journal article (JEM)                  1992           2
a
Abbreviations: AE Applied Economics; AJAE American Journal    Journal of Environmental Management; JEPM Journal of
of Agricultural Economics; CJAE Canadian Journal of Agri-     Environmental Planning and Management; LE Land Economics;
cultural Economics; EE Ecological Economics; EPA Environment   QJE Quarterly Journal of Economics; TAFS Transactions of the
and Planning A, ERAE European Review of Agricultural Eco-     American Fisheries Society; WRR Water Resources Research
                                 b
                                  Number of observations taken from each study
nomics; ERE Environmental and Resource Economics; JEEM
Journal of Environmental Economics and Management; JEM



                                 The studies included in the analysis focus primarily on
meta-analysis. First, a number of people have been
                                 wetlands or wetland-type areas.2 The speci®c WTP
involved in several studies and related publications. This
may result in an `authorship' effect. Learning from pre-     questions addressed in each study cover a large con-
vious studies, authors may use similar, perhaps slightly     tinuum of activities, actions or projects related to wet-
adapted survey designs in subsequent studies. Secondly,      lands, but in some cases (approximately a third of all
103 data points (observations) were extracted from 30       studies) also to water resources in general. These values
studies. This corresponds, on average, to three or four      were kept in the analysis because they referred directly
observations per study. More than half of all studies       to the hydrological wetland functions distinguished in
provided one or two observations. Outliers are the        the analysis and were considered reliable estimates for
studies by Loomis et al. (1991) and Desvousges et al.       these functions. The WTP questions range from outdoor
(1987), providing 10 and 21 observations respectively.      recreational activities like birdwatching or ®shing, to
Studies provided more than one observation mainly
because of the use of split survey samples targeting
different wetland user and non-user groups and testing      2
                                  Although there is little agreement among scientists on what
different survey designs. The possibility that results from
                                 constitutes a wetland, a workable de®nition is given by the so-
the same study cluster together, for example as a result of
                                 called Ramsar Convention (1975, article 1): `areas of marsh fen,
identical survey design or sample population, and that      peatland or water, whether natural or arti®cial, permanent or
results from some studies may be more variable than        temporary, with water that is static or ¯owing, fresh, brackish or
others was tested and accounted for in the meta-analysis     salt including areas of marine water, the depth of which at low
(see `The model').                        tide does not exceed 6 m'



                                                                          49
                                   Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
    Original article



                                        in all other cases wetland functions provided multiple
   ground-water protection and complete wildlife habitat
                                        bene®ts.
   preservation. Two-thirds of the studies are carried out in
                                        Most studies (70%) asked respondents for the use and
   the USA, the rest in Europe. Half of the European
                                        non-use values attached to the bene®ts derived from
   studies were carried out in the UK.
                                        wetland functions. One study (Schultz and Lindsay
   Separating the heterogeneous complex of hydrological,
                                        1990), elicited future use value only (option value),
   biogeochemical and ecological functions performed by the
                                        while another (Greenley et al. 1981) focused, among
   wetlands considered in each study in the meta-analysis
                                        others, on the ex ante `option price' of being able to
   was very dif®cult. Based on the various functions ad-
                                        make a better informed judgement in the future based
   dressed in the reviewed studies, a simple distinction was
                                        on more information becoming available regarding wa-
   made therefore ®rst of all between four main wetland
                                        ter quality.
   ecosystem functions: ¯ood control, water generation, wa-
                                        In eight studies an attempt was made to break down the
   ter quality support and wildlife habitat provision (Fig. 1).
                                        stated total economic value ex post in the questionnaire
   Secondly, the main function valued in each study was as-
                                        into the various components distinguished in the litera-
   signed to one of each of these four groups. Hence, each
                                        ture, e.g. use, option, philantrophic, bequest, stewardship
   study was categorised as addressing one of these four main
                                        and existence value. In two water quality studies (De-
   wetland functions, unless a study explicitly generated
                                        svousges et al. 1987; Carson and Mitchell 1993), respon-
   distinct values for different wetland functions (for more
                                        dents were presented ex ante with a `value card' which
   details, see Brouwer et al. 1997).
                                        described the main reasons why water quality might be
   Obviously, wetland ecosystem structures and processes
                                        valued. In another two studies use and non-use values
   and the functions they provide are highly interrelated,
                                        were elicited separately, by the use of either different
   making it very hard, and in some cases impossible, to
                                        questionnaires (Bishop et al. 1987) or separate questions
   distinguish between individual functions. They often go
                                        for use and non-use values in the same questionnaire
   hand in hand and attempts to separate them, for example
                                        (Greenley et al. 1981).
   for economic valuation purposes in order to avoid double
                                        Finally, two study quality indicators were included in
   counting, are liable to be arbitrary. This implies that
                                        the analysis: one for the quality of the studies included in
   double counting is a real problem and attempts to aggre-
                                        the meta-analysis and one for the quality of the meta-
   gate up to system-level values are fraught with dif®culties.
                                        analysis itself. The quality of individual studies is indicated
   Also, in the case of the human bene®ts derived from the
                                        by the study response rate and the quality of the meta-
   wetland functions involved, complete separation of direct
                                        analysis by the so-called scope test. Both indicators are
   and indirect use and non-use bene®ts is dif®cult. Only in a
                                        found back in the National Oceanic and Atmospheric
   third of all studies could a single bene®t ¯ow be identi®ed;




                   Wetland ecosystem
                 structure and processes



                      Functions




   Hydrological           Biogeochemical        Ecological
   1) Flood water retention     3) Nutrient retention and   4) Nursery and habitat for plants,
   2) Surface and groundwater      export            animals and micro-organisms
    recharge                            and landscape structural
                                                       Fig. 1
                                    diversity
                                                     Main wetland ecosystem functions
                                                     identi®ed in the meta-analysis and their
                                                     derived socio-economic bene®ts. The
                                                     analysis is based on stated WTP (will-
                    Socio-economic benefits
                                                     ingness to pay) for goods and services for
   - Natural flood protection    - Improved water quality   - Fishing
                                                     which no market exists. The value of
    alternative           - Waste disposal       - Wildfowl hunting
                                                     marketed products such as reed or ®sh is
   - Reduced damage to infra-                   - Other recreational amenities
                                                     excluded from the analysis to avoid
    structure, property and crops
                                                     double counting with the stated use and
   - Water supply
                                                     non-use values
   - Habitat maintenance



50   Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
                                                            Original article



                                gression technique called multilevel modelling was used
Administration's (NOAA's) `burden of proof' requirements
(Arrow et al. 1993).3                      (Langford 1994; Langford et al. 1998). [The GLS regression
                                was carried out using the package MLN (Rasbash and
A high non-response, either to the entire survey instru-
                                Woodhouse 1995).]
ment or to the valuation question, raises concern regard-
                                Given the shape of the distribution of the WTP amounts,
ing the study's representativeness, and questions the
                                a logarithmic transformation was used. Hence, for the
validity of the survey design employed and the extent to
                                log(WTP) amounts of the studies Y, the GLS model is:
which the valuation scenario in the questionnaire was
comprehensible and credible. The scope test refers to the
                                              Y ˆ Xb ‡ ZH                 …1†
sensitivity of WTP measures to changes in the provision
level of the goods and services being valued, i.e. the     where Xb consists of the design matrix X and associated
difference between reference and target provision levels.    parameters b represent the mean or ®xed effects of the
Conforming to the strong monotonicity assumption in       explanatory variables on the dependent variable Y. How-
neo-classical consumer theory, responses should reveal a    ever, whereas in OLS regression there is a single vector of
smaller WTP for smaller amounts of an environmental       error terms or residuals, here a more complex variance
commodity provided by an environmental programme.        structure may be modelled where the values of residuals
Very few studies reported the extent of protest bids and    are dependent on explanatory variables included in the
other questionable responses in the survey. Although most    design matrix Z for the random part of the model. For
studies mention the survey response rates, it is in many    example, using one explanatory variable for simplicity,
cases not clear what these response rates actually represent  Eq. (1) can be written as:
or which criteria have been used to exclude responses
from further analysis. Where such information was            yi ˆ b0 ‡ b1 x1i ‡ ui
                                                !
                                             !                 !
available, protest bids and questionable responses were
                                                  r2
                                            ui    0           rum
excluded from the response rates.                                  u
                                                 Y
                                       ‡ m1i Y    $N                      …2†
                                                            r2
                                            mi    0  rum
In order to carry out a scope test, the size of the affected                               m
study site and the difference between the reference and the
                                where ui is the residual associated with the intercept b0,
target levels of environmental service provision in the CV
                                and vi is the residual associated with the slope parameter
scenarios should ideally be considered. However, in two-
                                b1 of x1. While the variance of the responses in OLS is
thirds of all studies no information is provided about the
                                determined by a single residual term, in the basic GLS
size of the area involved. In about one-third of all studies,
                                model the variance is dependent on the explanatory
the study site size was estimated using geographical maps.
                                variable:
Problems accumulate when aiming to also include the
difference between the reference and target levels of the             var…yi † ˆ r2 ‡ 2rum x1 ‡ r2 x2            …3†
                                               u       m1
various wetland functions distinguished in the environ-
mental scenarios in each study. The multi-dimensional      This can be done for any number of variables, hence
nature of these functions makes a comparison between      making the variance of the responses a complex function
studies impossible. Hence, instead a `relative size' variable  of the explanatory variables, accounting for he-
was compiled, referring to the share of each study site in   teroscedasticity. This turns out to be highly relevant, as
the country's total stock of wetlands.             there are signi®cant differences, for example, in the vari-
                                ance of responses within different studies (intra-study
                                effects). Using the subscript j to label different studies,
     The model                       the basic previous GLS model can be rewritten as:
                                  yij ˆ b0 ‡ b1 x1ij ‡ uij ‡ m1ij x1ij
The structure of the data used in the meta-analysis is               !   !         !
                                                    r2 rum
complex. WTP values are generated by different studies,             ui      0     u
                                     ‡ sj Y    $N      Y       Y
carried out in different geographical locations using dif-                       rum r2
                                         mi      0       m
ferent valuation formats. Using the summary statistics of
                                     sj ˆ N…0Y r2 †               …4†
these different studies in a pooled sample, the usual con-              s
ditions required for ordinary least squares (OLS) regres- where r2 is the variance parameter that describes the
                                  s
sion are likely to be violated. In order to account for  differing variability of estimates within different studies.
heteroscedasticity, a generalised least squares (GLS) re- This latter model will be referred to as the `extended'
                              model in the results section.
3
 In 1992, the NOAA commissioned a prestigious `Blue-Ribbon
Panel' of economists and survey specialists, co-chaired by Nobel
                                      Results
laureates Kenneth Arrow and Robert Solow, to investigate the CV
method. After carefully considering a wide range of issues, the
panel's report gave the method a quali®ed bill of health, but only
                                       Summary statistics
if studies were conducted to a rigorous set of guidelines. The
                                  A ®rst step in the meta-analysis was to make stated aver-
panel identi®ed a subset of issues which it called `burden of
                                  age WTP amounts in each study comparable. The response
proof ' requirements



                                                                        51
                                 Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
    Original article



                                        surveys were carried out is used as an indicator for income
   variable is average WTP per household per year for the
                                        differences to account for people's capacity to pay. Income
   preservation of speci®c wetland aspects. After expressing
                                        taxation as a payment vehicle generates the highest aver-
   WTP in national currencies in terms of their 1990 pur-
                                        age WTP value, followed by stated WTP over and above
   chasing power, these national currencies were converted in
                                        actual trip expenditures to visit a wetland site (use value).
   the International Monetary Fund's (IMF's) Special Draw-
                                        The latter is part of a set of payment mechanisms that
   ing Rights (SDRs), which is the Fund's of®cial monetary
                                        present wetland functions to respondents as a private
   unit of account (IMF 1996). Average WTP for wetland
                                        good, i.e. to be `consumed' by the individual who is being
   function preservation found in all studies taken together is
                                        interviewed by asking him or her to pay, for example, an
   62 SDRs (by the end of 1995, 1 SDR approximately
                                        entrance fee. On the other hand, general income taxation is
   equalled 1.5 US$). The median is considerably lower,
                                        expected to prompt responses that consider the implica-
   namely 34 SDRs.
                                        tions of wetland preservation for society at large, not just
   The breakdown of WTP values according to a number of
                                        for the individual (Mitchell and Carson 1989; Blamey
   possible explanatory factors is presented in Table 2. Mean
                                        1995). Hence, the higher WTP value elicited through this
   WTP values have been calculated for (1) wetland types as
                                        payment mechanism is expected to re¯ect more than
   identi®ed by Dugan (1990); (2) main wetland functions;
                                        private use values only. (Obviously, `users' of speci®c
   (3) relative wetland size; (4) the different value types
                                        wetland functions, for instance people who visit a wetland
   elicited in the studies (use and/or non-use values); (5) the
                                        site for recreational ®shing or boating, may hold values
   continent where the wetland sites are found; (6) the way
                                        related to their non-use, e.g. preservation for future gen-
   people were asked to pay for wetland function preserva-
                                        erations, as well.) The high value for non-speci®ed pay-
   tion in the CV survey as part of the institutional setting of
                                        ment modes is due to outliers and the very low number of
   the wetland conservation programmes (e.g. through
                                        observations. Calculating through the value of wetland
   general income taxation or otherwise); and (7) the way
                                        function preservation in existing product prices yields a
   the WTP question was elicited in the CV survey (e.g. in
                                        signi®cantly higher mean WTP than the establishment of a
   an open-ended question or otherwise).
                                        private fund or raising entrance fees.
   The calculated differences in mean WTP for each of these
                                        Finally, corresponding to previous research results (e.g.
   categories are statistically signi®cant at the 5% signi®-
                                        McFadden 1994; Bateman et al. 1995; Willis et al. 1995),
   cance level or stronger (see the outcomes of the Chi-
                                        the open-ended (OE) elicitation format yields a signi®-
   square test statistic in the last column of Table 2). The
                                        cantly lower WTP than other formats. The dichotomous
   range of values (minimum and maximum) found for the
                                        choice format (yes or no to a given bid amount) yields the
   factor levels across studies is considerable. Mean WTP
                                        highest average WTP, followed by the iterative bidding
   per household is more or less the same for salt- and
                                        procedure (yes or no to a sequence of bid amounts).
   fresh-water wetlands. However, the number of observa-
                                        Possible explanations are the larger numbers of non-
   tions for salt-water wetlands is very low. Almost all ob-
                                        responses or protest responses OE elicitation tends to
   servations refer to fresh-water wetlands. Within fresh-
                                        produce (Desvousges et al. 1983) or the uncertainty
   water wetlands, the value of wetlands fed by rivers (riv-
                                        experienced in answering the unfamiliar WTP question for
   erine) is twice as high as the value of lakes and ponds
                                        non-market goods and services in an OE format (Bateman
   (lacustrine) or marshes and swamps (palustrine).
                                        et al. 1995).
   Ground-water is valued highest, although the number of
   observations is again low.
                                              Regression results
   The wetland function ¯ood control generates the highest
                                        The ®ndings for the basic and extended GLS model in
   mean WTP, followed by wildlife habitat provision and
                                        which we account for study level effects are presented in
   landscape structural diversity (labelled biodiversity in
                                        Table 3. Only those variables are included that are statis-
   Table 2). Surface and groundwater recharge (labelled
                                        tically signi®cant at the 0.1 level. The ®xed part of the
   water generation in Table 2) has the lowest value. As
                                        model represents the ®xed or mean effects of each variable,
   expected, larger sites result in higher WTP. An incon-
                                        as for an OLS regression model, while the random part
   sistency is found between the categories `small' and `very
                                        displays the variance and covariance parameters that
   small'.
                                        model heteroscedasticity.
   Use values associated with wetland functions are almost
                                        The estimates for the regression results are obtained
   twice as high as non-use values. However, a combination
                                        through Maximum Likelihood techniques (e.g. Maddala
   of the two is not equal to their sum, suggesting some non-
                                        1983). The outcome of the likelihood ratio test
   linear relationship between the two. Socio-psychological
                                        (v2 ˆ 96X51; P < 0.01) rejects the null hypothesis of zero
   and related factors underpinning so-called embedding ef-           14
   fects, where the sum of the valuations placed on the parts         effects for all explanatory variables. A pseudo R-squared
   of a commodity exceeds that for the whole (Bateman et al.         was calculated from the log likelihood (LL) function. The
   1997), may be one important reason.                    outcome corresponds with the goodness of ®t measures
   North Americans are willing to pay, on average, more than         usually found in CV studies. Since the pseudo R-squared
   Europeans. Since average income data for the survey            lacks the straightforward explained variance interpretation
   samples are missing in most of the studies reviewed, the          of R-squared in OLS regression (Hamilton 1993), it is used
   country in which the wetland sites are found and the CV          here as a rough indicator for the model's goodness-of-®t.


52   Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
                                                              Original article



  Table 2
Summary statistics (WTP willingness to pay; SDR Special Drawing Rights)

                                                           na   v2 (p<)b
                    Mean WTP (SDRs)    Standard error    Min (SDRs)    Max (SDRs)

Wetland type                                                         15.2 (0.05)
Saltwater                56.2         27.2         19       137        4
Marine                 22.7          3.7         19        26        2
Lagoonal                136.6          ±           ±        ±        1
Lake                  42.8          ±           ±        ±        1
Freshwater               58.9          6.1          1       267        97
Riverine                71.7         13.7          1       267        38
Lacustrine               36.8          9.4         12        88        9
Palustrine               36.9          4.3          9       117        31
Groundwater              125.7         24.3         99       174        3
Fresh- and saltwater          237.5         106.2         131       344        2
Wetland function                                                        7.8 (0.05)
Flood control              92.6         24.4         24       177        5
Water generation            21.5          6.8          3        59        9
Water quality              52.5          5.9          9       174        43
Biodiversity              76.1         12.8          1       344        46
Relative wetland size                                                     13.1 (0.01)
Very large               86.9         17.6         19       177        8
Large                  70.3         21.6         12       344        16
Medium                 67.0          8.9          3       267        58
Small                  29.5         13.2          1       137        13
Very small               53.4         13.8         24       105        6
Value type                                                           6.1 (0.05)
Use value                68.1          8.4          9       344        50
Non-use value              35.5          4.8         12        78        13
Use and non-use values         63.8         12.9          1       267        40
                                                               A3.0 (0.003)c
Country
USA and Canada             70.8          7.8          3       344        80
Europe                 32.8          8.4          1       177        23
Payment mode                                                         27.4 (0.001)
Income tax (1)             121.3         18.1          2       267        22
Entrance fee/private fund (2)      28.6          5.7          1       137        28
Product prices (3)           47.8          8.9          3       174        22
Combination of (1) and (3)       42.8          6.3          9       117        26
Trip expenditures           102.9          6.8         89       112        3
Not speci®ed              237.5         106.2         131       344        2
Elicitation format                                                      10.1 (0.01)
Open-ended               37.4          6.5          1       137        35
Dichotomous choice           91.2         17.1          3       344        29
Iterative bidding            78.5         14.9          9       244        20
Payment card              47.1          8.4         10       174        19
a
 Number of observations does not sum up to 103 in all cases as a result of missing values
b
 Outcome of the non-parametric Kruskal±Wallis test statistic which has approximately a Chi-squared distribution under the null
hypothesis of equal average WTP in all groups
c
 Outcome of the non-parametric Mann±Whitney test statistic for two independent samples which has approximately a standard
normal distribution under the same null hypothesis


                                 WTP is reduced, on average, by 41% (ceteris paribus)
The estimated models account for approximately 37% of
                                 when using studies in a value transfer exercise which are
the observed variability in the mean WTP values found in
                                 based on an open-ended WTP question.
individual studies.
                                 The basic model also indicates that study location has
For the ®xed part of the basic and extended model, the
                                 a signi®cant impact on average WTP. The dummy
estimated coef®cients in the semi-log function represent
                                 variable has a value of 1 if the research took place in North
the constant proportional rate of change in the dependent
                                 America and zero if in Europe. As shown before, average
variable per unit change in the independent variables
                                 WTP is substantially higher in North America than in
(Johnston 1984). Hence, the coef®cient estimated for the
                                 Europe. The parameter estimates for the four main wet-
dummy variable `Payment vehicle' in the basic model re-
                                 land functions are particularly interesting. These functions
¯ects, ceteris paribus, an almost twice as higher average
                                 are found to have a statistically signi®cant role in
WTP for an increase in income tax than for any other
                                 explaining variance in average WTP. The size of the
payment vehicle. Compared with other elicitation formats,


                                                                       53
                                 Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
     Original article



     Table 3
   Generalised least squares results for the basic and extended model. As a result of missing values for explanatory variables,
   number of observations is reduced from 103 to 92

   Parameter           Parameter de®nition              Basic model            Extended model

                                          Estimate    Standard error   Estimate    Standard error

   Fixed effects
   Constant            Intercept                  3.356***     0.100         3.311***  0.247
   Payment vehicle        Dummy: 1   =  income tax; 0 = other    1.880***     0.265         1.576***  0.362
   Elicitation format       Dummy: 1   =  open-ended; 0 = other   A0.411**     0.130        A0.376*   0.183
   Country            Dummy: 1   =  North America; 0 = other  1.861***     0.217         1.629***  0.363
   Response rate (1)       Dummy: 1   =  30±50%; 0 = other     A2.253***     0.326        A1.722***  0.451
   Response rate (2)       Dummy: 1   =  >50%; 0 = other      A1.904***     0.333        A1.461**   0.450
   Flood control         Dummy: 1   =  ¯ood control; 0 = other   1.477***     0.240         1.134*   0.456
   Water generation        Dummy: 1   =  water generation; 0 = other 0.691*      0.342         0.441    0.479
                                         0.545       0.282         0.659*   0.327
   Water quality         Dummy: 1   =  water quality; 0 = other
   Random effects
   Between studies
   r2               Variance                     ±      ±           0.160*   0.071
    constant
   Between average WTP
   r2               Variance                     0.059*    0.029        0.045    0.028
    constant
   rpayment vehicle, constant   Covariance                    0.020    0.043        0.001    0.036
   rcountry, constant       Covariance                    0.689**   0.222        0.351**   0.129
   rcountry, payment vehicle   Covariance                   A0.707**   0.226        A0.345**   0.134
   r¯ood control, constant    Covariance                   A0.013    0.050        0.027    0.060
                                                           A0.266 
   rwater generation, constant  Covariance                   A0.637**   0.227               0.153
                                          A0.424 
   rwater quality, constant    Covariance                          0.231        A0.188    0.135
   LLunconstrained                                A83.907              A81.874
   Pseudo R-squared                                0.365               0.380
   n                                       92                92
    
     Signi®cant at 0.10; * signi®cant at 0.05; ** signi®cant at 0.01; *** signi®cant at 0.001


                                          effects is as follows: the variance of the constant in the
   estimated parameters indicates that average WTP is, as
                                          basic and extended model is the variance associated with
   before, highest for ¯ood control, but this time, whilst
                                          the baseline case, i.e. where the value of all the explanatory
   controlling for other explanatory factors, followed by
                                          variables is zero. The variance of, for example, North
   water generation and water quality and lowest for the
                                          American studies in the basic model is
   wetland function biodiversity supply. The latter is used as
                                          (0.059 + 2 ´ 0.689) ˆ 1.437. Hence, North American
   the baseline category in the regression analysis in order to
                                          studies are more variable than European ones. In this way
   avoid multicolinearity. The positive parameter estimates
                                          heteroscedasticity can be modelled in the basic model.
   for the three other wetland functions indicate that these
                                          Another example is North American studies using income
   functions generate higher values than the baseline function
                                          taxation as a payment vehicle:
   biodiversity supply.
   This suggests the prominence of use over non-use moti-
                                          r2
   vations underpinning stated WTP amounts. The distinc-              countryY payment vehicle ˆ 0X059 ‡ 2  0X689
   tion between use and/or non-use values does not have a
                                                       ‡ 2  0X020 À 2  0X707 ˆ 0X023X
   signi®cant impact on average WTP, probably because the
   corresponding variance is already accounted for by the
   distinction between wetland functions. Also, relative wet-          The last term in this equation is the covariance between
   land size is statistically not signi®cant. Higher response          payment vehicle and country. So, wetland CV studies
   rates, a rough indicator of better overall study quality,           based on income taxation in the US appear to have a
   appear to result in signi®cantly lower average WTP than            particularly low variance.
   low response rates. A practical explanation may be that            Accounting for study-level effects in the basic model
   low response rates are sometimes biased towards includ-            signi®cantly reduces the sample variance or standard
   ing a relatively large number of sample respondents with a          deviation of average WTP in the extended model
                                          (v2 ˆ 4X06; P < 0.05). The extended, multilevel model ac-
   greater interest than average in environmental protection            1
   and corresponding WTP.                            counting for the random effects between studies hence
   The model's random effects can be used to (1) model              provides a signi®cant improvement over conventional
   heteroscedasticity and (2) investigate the suitability of           meta-analysis by allowing for the hierarchical structure of
   using speci®c CV results in a value transfer exercise. This          data implicit in clustering of multiple results from
   will be explained below. The interpretation of the random           single studies. As expected, having explained some of the


54     Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
                                                            Original article



                                with the resource. It is people's perception of a good's
variance in the model by study-level effects, the random
                                characteristics or functions that in¯uence their attitudes
effects between log(WTP) amounts have decreased, except
                                and behaviour, not necessarily the good's `objective'
for the covariance between ¯ood control and the intercept.
                                characteristics (e.g. see Adamowicz et al. 1997). Finally,
However, in the extended model the ®xed effects have
                                accounting for intra-study variability, the statistical anal-
slightly decreased, while the signi®cance level of half of the
                                ysis produces slightly different results regarding the sig-
explanatory variables is lower. Only the signi®cance of the
                                ni®cance and size of the effect of the main functions on the
wetland function water quality has increased. The wetland
                                WTP values. The low number of observations also has to
function water generation has become statistically insig-
                                be taken into consideration.
ni®cant and has switched place with the function water
                                On the other hand, the study progressed meta-analytical
quality supply in the order of size, suggesting that the
                                research in environmental economics by providing a
results must be viewed with some caution when properly
                                statistical multilevel model which accounts for the clus-
accounting for intra-study variability.
                                tering of results from the same studies, for example as a
If low variance is considered an estimate of quality in the
                                result of identical survey design or sample population,
sense that study results are better suited for comparison
                                and owing to the fact that results from some studies may
and hence can be more readily put together in a value
                                be more variable than others. In the GLS models used,
transfer exercise, then it can be concluded that, on the
                                the variance and covariance estimators not only enabled
whole, studies using income taxation as a payment vehicle
                                us to model heteroscedasticity, but also provided im-
are better suited than other payment vehicles, and that
                                portant background information for environmental value
studies valuing wetland biodiversity tend to be less
                                transfer.
variable than studies valuing wetlands in their capacity
                                Finally, although considerable effort has been put into
of generating water or maintaining water quality.
                                specifying the characteristics of the environmental func-
                                tions and correspondingly the environmental goods and
                                services involved, other important aspects that may have
      Discussion and conclusions             helped to explain differences in valuation outcomes re-
                                main unde®ned. This is a common problem in meta-an-
In this paper, estimates for socio-economic use and non- alyses in the ®eld of environmental valuation as a result of
use values attached to different hydro-ecological and bio- insuf®cient and inadequate information provided in pub-
                                lished valuation studies. Relevant information about the
geochemical wetland functions were compared and syn-
                                samples' socio-economic values is missing in many stud-
thesised in a meta-analysis of wetland CV studies. The
meta-analysis provides insights into the factors that have to ies, let alone respondents' socio-psychological and cultural
be considered when attempting to transfer environmental characteristics.
values on the basis of CV studies. A statistically signi®cant In meta-analysis, inferences are made on the basis of
breakdown of WTP values for four main wetland functions information on global statistics, such as the mean and
has been presented. Although single ecosystem character- standard deviations of parameter estimates. These may or
                                may not describe individual behaviour adequately. In or-
istics or functions are given meaning and value within
                                der to overcome this potential problem and to increase the
existing ecosystem structures, the distinction between
functions is essential for a valid transfer of the economic study's validity and reliability, a logical next step would be
                                to gather more information about sample population
values generated within an ecosystem's primary self-or-
                                characteristics by complementing the analysis with the
ganising capacity. It reduces the risk of double counting
                                underlying individual responses. This will provide an
when attempting to assess a natural resource's total eco-
                                important test of the appropriateness of meta-analysis as
nomic value on the basis of different valuation studies.
                                an instrument to synthesise CV outcomes for the purpose
From an anthropocentric point of view, the size of the
                                of value transfer.
estimated parameters in the estimated basic model is as
expected. Average WTP is highest for ¯ood control, be-
cause of the possible risks to life and livelihood as a result Acknowledgements The Centre for Social and Economic
of ¯ooding and the capacity of wetlands to reduce this risk, Research on the Global Environment (CSERGE) is a designated
                                research centre of the UK Economic and Social Research Council
followed by water supply and water quality and ®nally the (ESRC). This research paper is part of the European Commission
provision and maintenance of biodiversity. However, these Directorate General XII-funded project `Ecological±Economic
results have to be handled with care for a number of      Analysis of Wetlands' (ECOWET), contract no. ENV4-CT96±
reasons.                            0273, co-ordinated by CSERGE.
Ecosystem structures and processes provide a heteroge-
neous complex of highly interrelated socio-economic
functions. The analysis presents a simple and arbitrary
                                      References
breakdown of these functions into independent compo-
nents. The distinction between four main wetland func-     ADAMOWICZ W, SWAIT J, BOXALL P, LOUVIERE J, WILLIAMS M (1997)
                                 DAMOWICZ   WAIT   OXALL  OUVIERE   ILLIAMS
tions does also not necessarily correspond with people's     Perceptions versus objective measures of environmental quality
perception of the various functions wetlands perform,      in combined revealed and stated preference models of envi-
which depends upon their own knowledge and experience ronmental valuation. J Environ Econ Manage 32:65±84


                                                                      55
                                 Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
    Original article



                                        DESVOUSGES WH, NAUGHTON MC, PARSONS GR (1992) Bene®t
   ARROW K, SOLOW R, PORTNEY PR, LEAMER EE, RADNER R, SCHUMAN H
   RROW    OLOW    ORTNEY    EAMER    ADNER   CHUMAN      ESVOUSGES      AUGHTON    ARSONS
   (1993) Report of the NOAA Panel on Contingent Valuation.          transfer: conceptual problems in estimating water quality ben-
   Federal Register (15 January) 58(10):4601±4614               e®ts using existing studies. Water Resour Res 28(3):675±683
   BARBIER EB (1993) Sustainable use of wetlands. Valuing tropical      DUGAN PJ (ed) (1990) Wetland conservation. A review of current
   ARBIER                                   UGAN
   wetland bene®ts: economic methodologies and applications.          issues and required action. World Conservation Union, Gland,
   Geogr J 159(1):22±32                            Switzerland
   BATEMAN IJ, LANGFORD IH, TURNER RK, WILLIS KG, GARROD GD          FARBER S (1988) The value of coastal wetlands for recreation: an
   ATEMAN    ANGFORD    URNER     ILLIS    ARROD        ARBER
   (1995) Elicitation and truncation effects in contingent valuation      application of travel cost and contingent valuation methodolo-
   studies. Ecol Econ 12(2):161±179                      gies. J Environ Manage 26:299±312
   BATEMAN IJ, MUNRO A, RHODES B, STARMER C, SUGDEN R (1997) Does       GARROD GD, WILLIS KG (1996) Estimating the bene®ts of envi-
   ATEMAN    UNRO   HODES   TARMER    UGDEN            ARROD     ILLIS
   part±whole bias exist? An experimental investigation. Econ J        ronmental enhancement: a case study of the River Darent. J
   107(441):322±332                              Environ Plann Manage 39(2):189±203
   BERGH van den JCJM, BUTTON KJ, NIJKAMP P, PEPPING GC            GLASS GV, MCGAW B, SMITH ML (1981) Meta-analysis in social
   ERGH           UTTON     IJKAMP    EPPING        LASS     C AW    MITH
   (1997) Meta-analysis in environmental economics. Kluwer,          research. Sage Publications, Beverly Hills
   Dordrecht                                 GREEN CH, TUNSTALL SM (1991) The evaluation of river water
                                         REEN     UNSTALL
   BERGSTROM JC, STOLL JR, TITRE JP, WRIGHT VL (1990) Economic         quality improvements by the contingent valuation method. Appl
   ERGSTROM    TOLL   ITRE    RIGHT
   value of wetlands-based recreation. Ecol Econ 2:129±147           Econ 23:1135±1146
   BISHOP RC, BOYLE KJ (1985) The economic value of Illinois Beach      GREENLEY DA, WALSH RG, YOUNG RA (1981) Option value: empir-
   ISHOP    OYLE                              REENLEY     ALSH    OUNG
   State Nature Reserve. Consultancy report. HBRS, Madison,          ical evidence from a case study of recreation and water quality.
   Wisconsin                                  Q J Econ 96(4):657±673
                                               È
   BISHOP RC, BOYLE KJ, WELSH MP (1987) Toward total economic         GREN I-M, SoDERQVIST T (1994) Economic valuation of wetlands:
   ISHOP    OYLE    ELSH                        REN
   valuation of Great Lakes ®shery resources. Trans Am Fish Soc        a survey. Beijer Discussion Paper Series no. 54, Beijer Institute,
   116:339±345                                 Stockholm
   BLAMEY RK (1995) Citizens, consumers and contingent valuation:       HAMILTON LC (1993) Statistics with Stata 3. Duxbury Press,
   LAMEY                                    AMILTON
   an investigation into respondent behaviour. PhD Diss, Austra-        Belmont, California
   lian National University, Canberra                    IMF (1996) International ®nancial statistics yearbook. Interna-
   BROUWER R, LANGFORD IH, BATEMAN IJ, CROWARDS TC, TURNER RK          tional Monetary Fund, Washington
   ROUWER    ANGFORD    ATEMAN    ROWARDS     URNER
   (1997) A meta-analysis of wetland contingent valuation studies.      JOHNSTON J (1984) Econometric methods, 3rd edn. McGraw-Hill,
                                         OHNSTON
   Global Environmental Change Working Paper 97±20, Centre for         Maidenhead
   Social and Economic Research on the Global Environment          JORDAN JL, ELNAGHEEB AH (1993) Willingness to pay for improve-
                                         ORDAN    LNAGHEEB
   (CSERGE), University of East Anglia, Norwich, and University        ments in drinking water quality. Water Resour Res 29(2):237±
   College, London                               245
   BROUWER R, SLANGEN LHG (1998) Contingent valuation of the         KAORU Y (1993) Differentiating use and nonuse values for coastal
   ROUWER    LANGEN                             AORU
   public bene®ts of agricultural wildlife management: the case of       pond water quality improvements. Environ Resour Econ 3:487±
   Dutch peat meadow land. Eur Rev Agric Econ 25:53±72             494
   BROUWER R (1998) Future research priorities for valid and reliable     KOSZ M (1996) Valuing riverside wetlands: the case of the
   ROUWER                                   OSZ
   environmental value transfer. Global Environmental Change          ``Donau-Auen'' National Park. Ecol Econ 16:109±127
   Working Paper 98±28, Centre for Social and Economic Research       LANGFORD IH (1994) Using a generalized linear mixed model to
                                         ANGFORD
   on the Global Environment (CSERGE), University of East An-         analyze dichotomous choice contingent valuation data. Land
   glia, Norwich, and University College, London                Econ 70(4):507±514
   CARSON RT, MITCHELL RC (1993) The value of clean water: the        LANGFORD IH, BATEMAN IJ, JONES AP, LANGFORD HD, GEORGIOU S
   ARSON     ITCHELL                           ANGFORD     ATEMAN    ONES   ANGFORD    EORGIOU
   public's willingness to pay for boatable, ®shable, and swimm-        (1998) Improved estimation of willingness to pay in dichoto-
   able quality water. Water Resour Res 29(7):2445±2454            mous choice contingent valuation studies. Land Econ 74(1):65±
   CARSON RT, FLORES NE, MARTIN KM, WRIGHT JL (1996) Contingent         75
   ARSON    LORES    ARTIN     RIGHT
   valuation and revealed preference methodologies: comparing        LANT CL, ROBERTS RS (1990) Greenbelts in the cornbelt: riparian
                                         ANT    OBERTS
   the estimates for quasi-public goods. Land Econ 72:80±99          wetlands, intrinsic values, and market failure. Environ Plan A
   COOPER J, LOOMIS JB (1991) Economic value of wildlife resources in      22:1375±1388
   OOPER   OOMIS
   the San Joaquin Valley: hunting and viewing values. In: Dinar A,     LOOMIS JB (1987) Balancing public trust resources of Mono Lake
                                         OOMIS
   ZILBERMAN D (eds) The economics and management of water           and Los Angeles' water right: an economic approach. Water
    ILBERMAN
   drainage. Kluwer, Dordrecht                         Resour Res 23(8):1449±1456
   COSTANZA R, d'ARGE R, de GROOT R, FARBER S, GRASSO M, HANNON B,      LOOMIS JB, HANEMANN M, KANNINEN B, WEGGE T (1991) Willingness
   OSTANZA    ARGE     ROOT   ARBER    RASSO    ANNON     OOMIS     ANEMANN    ANNINEN   EGGE
   LIMBURG K, NAEEM S, O'NEILL RV, PARUELO J, RASKIN RG, SUTTON P,       to pay to protect wetlands and reduce wildlife contamination
    IMBURG   AEEM    EILL    ARUELO   ASKIN    UTTON
   van den BELT M (1997) The value of the world's ecosystem          from agricultural drainage. In: Dinar A, Zilberman D (eds) The
         ELT
   services and natural capital. Nature 387:253±260              economics and management of water drainage. Kluwer, Do-
   CROWARDS TC, TURNER RK (1996) FAEWE sub-project report:           rdrecht
   ROWARDS     URNER
   economic valuation of wetlands. Centre for Social and Economic      MADDALA GS (1983) Limited-dependent and qualitative variables
                                          ADDALA
   Research on the Global Environment (CSERGE), University of         in econometrics. Cambridge University Press, Cambridge
   East Anglia, Norwich, and University College, London           MCFADDEN D (1994) Contingent valuation and social choice. Am J
                                          C ADDEN
   CUMMINGS RG, GANDERTON PT, MCGUCKIN T (1994) Substitution          Agric Econ 76:689±708
   UMMINGS     ANDERTON     C UCKIN
   effects in cvm values. Am J Agric Econ 76:205±214             MITCHELL RC, CARSON RT (1989) Using surveys to value public
                                          ITCHELL    ARSON
   DESVOUSGES WH, SMITH VK, MCGIVNEY MP (1983) A comparison of         goods: the contingent valuation method. Resources for the Fu-
   ESVOUSGES     MITH     C IVNEY
   alternative approaches for estimation of recreational and related      ture, Washington, DC
   bene®ts of water quality improvement. Report to the US Envi-       OLSEN D, RICHARDS J, SCOTT RD (1991) Existence and sport values
                                         LSEN   ICHARDS   COTT
   ronmental Protection Agency, Washington, DC                 for doubling the size of Columbia River Basin salmon and
   DESVOUSGES WH, SMITH VK, FISHER A (1987) Option price estimates       steelhead runs. Rivers 2(1):44±56
   ESVOUSGES     MITH    ISHER
   for water quality improvements: a contingent valuation study       PEARCE DW, TURNER RK (1990) Economics of natural resources
                                         EARCE     URNER
   for the Monongahela River. J Environ Econ Manage 14:248±267         and the environment. Harvester Wheatsheaf, Hemel Hempstead



56   Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag
                                                               Original article



PHILLIPS WE, HANEY TJ, ADAMOWICZ WL (1993) An economic        SUTHERLAND RJ, WALSH RG (1985) Effect of distance on the
HILLIPS     ANEY   DAMOWICZ                  UTHERLAND    ALSH
 analysis of wildlife habitat preservation in Alberta. Can J Agric  preservation value of water quality. Land Econ 61(3):281±291
 Econ 41:411±418                           TOLBA MK, EL-KHOLY OA (1992) The world environment 1972±
                                    OLBA    L HOLY
RAMSAR (1975) Convention on wetlands of international impor-      1992. Chapman and Hall on behalf of the UN Environment
 AMSAR
 tance especially as waterfowl habitat. Ramsar Convention, Iraq    Programme, London
RASBASH J, WOODHOUSE G (1995) MLn command reference.         TURNER RK (1992) Policy failures in managing wetlands. In: OECD
 ASBASH   OODHOUSE                         URNER
 Institute of Education, University of London             market and government failures in environmental management.
SANDERS LD, WALSH RG, LOOMIS JB (1990) Toward empirical esti-     OECD, Paris
ANDERS     ALSH   OOMIS
 mation of the total value of protecting rivers. Water Resour Res  WHITEHEAD JC, BLOMQUIST GC (1991) Measuring contingent values
                                    HITEHEAD    LOMQUIST
 26(7):1345±1357                           for wetlands: effects of information about related environmental
SCHULTZ SD, LINDSAY BE (1990) The willingness to pay for        goods. Water Resour Res 27(10):2523±2531
CHULTZ    INDSAY
 groundwater protection. Water Resour Res 26(9):1869±1875      WILLIS KG (1990) Valuing non-market wildlife commodities: an
                                    ILLIS
SILVANDER U (1991) The willingness to pay for angling and ground    evaluation and comparison of bene®ts and costs. Appl Econ
ILVANDER
 water in Sweden. PhD Diss, Swedish University of Agricultural    22:13±30
 Sciences, Uppsala                          WILLIS KG, GARROD GD, SAUNDERS CM (1995) Bene®ts of environ-
                                    ILLIS   ARROD     AUNDERS
SPANINKS FA (1993) Een Schatting van de Sociale Baten van Be-     mentally sensitive area policy in England: a contingent valuation
PANINKS
 heersovereenkomsten met behulp van de Contingent Valuation      assessment. J Environ Manage 44:105±125
 Methode. MSc Diss, Department of Agricultural Economics and     WOLF FM (1986) Meta-analysis. Sage Publications, Beverly Hills,
                                    OLF
 Policy, Wageningen Agricultural University              California
SPANINKS FA, KUIK OJ, HOOGEVEEN JGM (1996) Willingness to pay of
PANINKS    UIK   OOGEVEEN
 Dutch households for a natural Wadden Sea. An application of
 the contingent valuation method. Report E-96/6, Institute for
 Environmental Studies, Amsterdam




                                                                         57
                                    Regional Environmental Change 1 (1) November 1999 á ã Springer-Verlag