Identifying the scope effect on a meta-analysis of biodiversity valuation studies

https://doi.org/10.1016/j.reseneeco.2011.03.002Get rights and content

Abstract

Insensitivity to scope is an issue of much debate in contingent valuation (CV) even today. Although the literature about insensitivity to scope is abundant, the reasons for it are not yet well known. Through a meta-analysis of CV studies of biodiversity, the treatment of the different possible measures of the quantity of the good is explored in relation to scope sensitivity. Overall, the findings suggest that CV results are sensitive to the amount of the good being valued, although the result depends on how the environmental change is measured. Results support the use of absolute sizes when conducting CV studies.

Highlights

► First study investigating the impact of the size variable format on scope effects. ► Sensitivity to scope in a meta-analysis varies with the size variable format. ► CV biodiversity studies pass the scope test with absolute changes in size. ► We explore novel explanatory variables in meta-analysis (between-subjects variable). ► Biodiversity nonuse values are more likely to derive in lower WTP.

Introduction

Contingent valuation (hereafter CV) is a popular approach to assessing environmental values in the absence of real markets. This method is especially relevant for its ability to estimate non-use values that could not otherwise be measured. However, since its appearance, it has received much criticism over biases and implementation problems. One of the major criticisms leveled at the method is its insensitivity to the size or amount of the good being valued (Arrow et al., 1993). This problem has been addressed by implementing a reliability test known as the scope test. Kahneman and Knetsch (1992) conducted one of the first studies claiming that CV led to inconsistent economic estimates. They based their claim on insensitivity to scope and explained their results by arguing that willingness to pay (hereafter WTP) estimates are not economic choices but rather choices based on moral satisfaction. Many other authors also argue that the method is inconsistent due to insensitivity to scope (Boyle et al., 1994b, Desvousges et al., 1993, Diamond and Hausman, 1994), while others argue that well conducted CV results pass the scope test, and there is no reason to invalidate the method as a whole (Carson and Mitchell, 1993, Carson, 1997, Loomis and White, 1996, Smith, 1992, among others). Therefore, the scope test criterion has become highly relevant in validating any given CV exercise.

Insensitivity to scope is nowadays looked at cautiously and several justifications of this finding have been presented in the relevant literature. The explanations put forward range from differences in the amount of information provided or in the respondent's perceptions (Carson and Mitchell, 1993), to the visible choice set (Bateman et al., 2005) and the ‘warm glow’ effect (Nunes and Schokkaert, 2003).2 Schulze et al. (1998) reject the view that increased market context can reduce scope insensitivity in CV studies, and they relate embedding effects to the nonuse nature of the goods. There are also authors (Banerjee and Murphy, 2005, Heberlein et al., 2005) who defend the idea that scope tests should not be a criterion of validity and that both results (scope sensitivity and insensitivity) may be expected depending upon the circumstances and characteristics of the valuation study and analysis. While Heberlein et al. (2005) find that careful design and analysis can still lead to scope insensitivity in some cases, Amiran and Hagen (2010) recently look at consumer utility theory to argue that a very low degree of sensitivity to scope can be justified when analyzing some classes of utility functions (named directionally bounded utility functions).

Besides being the focus of many CV studies, scope sensitivity has also been a key issue in meta-analyses conducted in stated preferences research. Meta-analyses are important analytical tools to better understand the findings of specific literature (Bateman and Jones, 2003). There are meta-analyses showing how WTP estimates are sensitive to the amount proposed for specific environmental goods. Cases in point include surface water quality improvements (Johnston et al., 2003, Van Houtven et al., 2007), increments in coral reef dive sites (Brander et al., 2007), visibility improvements in National Parks (Smith and Osborne, 1996), wildlife valuation studies (Loomis and White, 1996, Richardson and Loomis, 2009) and conservation of wetland ecosystems (Brouwer and Slangen, 1998, Brouwer et al., 1999). Other meta-analyses have shown diminishing marginal utility with size. Woodward and Wui (2001) find decreasing marginal values for the size of wetlands, where increasing their area had a negative effect on the payment per acre; along the same lines Johnston and Duke (2009) find decreasing values when the number of acres under farmland preservation programs is increased, as Ojea et al. (2010) for the case of forest ecosystems. In spite of this evidence, many studies have found insensitivity to scope. In forest valuation, Lindhjem (2007) reviews stated preference literature in Scandinavia for the last 20 years, finding that WTP estimates are insensitive to the size of the forest. More recently, Barrio and Loureiro (2010) obtain the same results regarding size insensitivity from a worldwide forest valuation study, while Zandersen and Tol (2009) employ European studies valuing forest recreation through travel cost and also find insensitivity to scope. As Table 1 shows, there is suggestive evidence pointing to the fact that the different units used to measure size changes in the environmental commodities may play a role in these results. Thus, a detailed analysis of the way in which researchers measure changes in environmental goods and how this affects scope sensitivity seems necessary in order to assess whether this relationship is a causal one.

The main contribution of this article is therefore specifically to test for scope sensitivity in a CV meta-analysis and compare the valuation of different environmental goods when sizes are presented in different formats. Based on an extensive review of the relevant literature, size measures were found not to be uniform across studies and at this point there has not yet been any discussion in the literature as to what size change measure might be adequate in terms of scope. To this end, sensitivity to scope when the measurement of size change varies is analyzed in this exercise, where we explore the sensitivity of results with respect to the various ways of size measurement in a meta-analysis. To address this issue, the next section of this article presents evidence of insensitivity to scope in previous CV literature and looks at how meta-analyses have addressed this problem. The data selection and data classification criteria are described next. Then the paper continues with a description of the model and the explanatory variables. Data analysis and main research hypotheses are then discussed, and the paper ends with the main findings and some concluding remarks.

Section snippets

Scope tests in CV literature

Scope tests are seen as consistency tests for CV-based estimates of WTP. These tests require WTP estimates to be responsive to the amount, or scope, of the environmental amenity being offered. A number of studies have dealt with the responsiveness of WTP to increments in the size of the environmental good. In ecosystem valuation, Loomis et al. (1993) show scope sensitivity between two levels of forest conservation, although no difference was found in terms of WTP estimates for an additional

Selection criteria and data classification

The data for the analysis that follows come from a review of the CV literature valuing ecosystems and biodiversity over the last twenty years. Published studies are used due to higher quality standards and availability. However, whenever possible additional unpublished references are incorporated into the dataset, following Horowitz and McConnell (2002).3

Modeling and specification

The dependent variable in the models is the estimated mean WTP reported by each original study.8 WTP estimates are first converted to dollars with the Purchasing Power Parity index (PPP) from the Penn World Table (Heston et al., 2006), and then converted to $2007 currency with the Consumer Price Index obtained from the IFM (2007). This is a common procedure used in meta-analyses (

Data analysis and results

With the dataset described, a meta-regression is estimated to further explore the sensitivity to scope of CV estimates, and in particular, sensitivity to different interpretations and codifications of the variable size. Following Wooldridge (2003) and Nelson and Kennedy (2009), we firstly conduct a Generalized Least Square (GLS) regression model, and secondly a random effects regression.9

Conclusions and implications

The present work summarizes the results from the CV literature valuing biodiversity at different levels, from the ecosystems to the species level, collecting together 355 observations from original studies. The main focus of this effort is to shed some light on the long term discussion on WTP sensitivity to scope, in the context of meta-analyses. Given the mixed results in scope test literature and in previous meta-analyses for environmental goods, as well as the heterogeneity that exists in

Acknowledgments

The authors wish to thank the editor and two anonymous reviewers for helpful comments. John Loomis, Ramon Ortiz, and Philip Wandschneider also provided great comments in earlier versions of this paper. Maria Loureiro acknowledges financial support from the BIODIVERSA-ERANET project number 2008-6504 “FIREMAN”.

References (153)

  • E.H. Bulte et al.

    The effect of varying the causes of environmental problems on stated WTP values: evidence from a field study

    Journal of Environmental Economics and Management

    (2005)
  • S.M. Chilton et al.

    A qualitative examination of how respondents in a contingent valuation study rationalize their WTP responses to an increase in the quantity of the environmental good

    Journal of Economic Psychology

    (2003)
  • M. Christie et al.

    Valuing the diversity of biodiversity

    Ecological Economics

    (2006)
  • S. Colombo et al.

    Analyzing the social benefits of soil conservation measures using stated preference methods

    Ecological Economics

    (2006)
  • K.L. Giraud et al.

    Internal and external scope in willingness-to-pay estimates for threatened and endangered wildlife

    Journal of Environmental Management

    (1999)
  • K. Giraud et al.

    Economic benefit of the protection program for the Steller sea lion

    Marine Policy

    (2002)
  • N. Hadker et al.

    Willingness-to-pay for Borivli National Park: evidence from a contingent valuation

    Ecological Economics

    (1997)
  • N. Hanley et al.

    Aggregating the benefits of environmental improvements: distance-decay functions for use and non-use values

    Journal of Environmental Management

    (2003)
  • T.A. Heberlein et al.

    Rethinking the scope test as a criterion for validity in contingent valuation

    Journal of Environmental Economics and Management

    (2005)
  • T.P. Holmes et al.

    Contingent valuation, net marginal benefits, and the scale of riparian ecosystem restoration

    Ecological Economics

    (2004)
  • J. Huang et al.

    Willingness to pay for quality improvements: should revealed and stated preference data be combined?

    Journal of Environmental Economics and Management

    (1997)
  • D. Kahneman et al.

    Valuing public goods: the purchase of moral satisfaction

    Journal of Environmental Economics and Management

    (1992)
  • J.E. Keith et al.

    Preservation or use: a contingent valuation study of wilderness designation in Utah

    Ecological Economics

    (1996)
  • M. Kniivila et al.

    Costs and benefits of forest conservation: regional and local comparisons in Eastern Finland

    Journal of Forest Economics

    (2002)
  • M. Kosz

    Valuing riverside wetlands: the case of the “Donau-Auen” National Park

    Ecological Economics

    (1996)
  • M.J. Kotchen et al.

    Environmental attitudes, motivations, and contingent valuation of non-use values: a case study involving endangered species

    Ecological Economics

    (2000)
  • E. Lehtonen et al.

    Non-market benefits of forest conservation in southern Finland

    Environmental Science and Policy

    (2003)
  • H. Lindhjem

    20 years of stated preference valuation of non-timber benefits from Fennoscandian forests: a meta-analysis

    Journal of Forest Economics

    (2007)
  • J.B. Loomis et al.

    Comparison of mail and telephone-mail contingent valuation surveys

    Journal of Environmental Economics and Management

    (1994)
  • J.B. Loomis et al.

    Economic benefits of rare and endangered species: summary and meta-analysis

    Ecological Economics

    (1996)
  • J.B. Loomis et al.

    Some empirical evidence on embedding effects in contingent valuation of forest protection

    Journal of Environmental Economics and Management

    (1993)
  • J.B. Loomis et al.

    Measuring the total economic value of restoring ecosystem services in an impaired river basin: results from a contingent valuation survey

    Ecological Economics

    (2000)
  • B. Martin-Lopez et al.

    The non-economic motives behind the willingness to pay for biodiversity conservation

    Biology Conservation

    (2007)
  • T.L. McDaniels et al.

    Multiattribute elicitation of wilderness preservation benefits: a constructive approach

    Ecological Economics

    (1998)
  • K. Arrow et al.

    Report of the NOAA panel on contingent valuation

    Federal Register

    (1993)
  • S. Banerjee et al.

    The scope test revisited

    Applied Economic Letters

    (2005)
  • H.S. Banzhaf et al.

    Valuation of natural resource improvements in the Adirondacks

    Land Economics

    (2006)
  • I.J. Bateman et al.

    Contrasting conventional with multi-level modelling approaches to meta-analysis: expectation consistency in U.K. woodland recreation values

    Land Economics

    (2003)
  • I.J. Bateman et al.

    Non-users’ willingness to pay for a National Park: an application and critique of the contingent valuation method

    Regional Studies

    (1997)
  • I.J. Bateman et al.

    Consistency between contingent valuation estimates: a comparison of two studies of UK National Parks

    Regional Studies

    (1994)
  • I.J. Bateman et al.

    Economic valuation of policies for managing acidity in remote mountain lakes: examining validity through scope sensitivity testing

    Aquatic Sciences

    (2005)
  • J. Bennett et al.

    Testing the validity of responses to the contingent valuation questioning

    Australian Journal of Agricultural and Resource Economics

    (1998)
  • J.M. Bowker et al.

    Use of dichotomous choice nonmarket methods to value the whooping crane resource

    Land Economics

    (1988)
  • K.J. Boyle et al.

    Valuing wildlife in benefit–cost analysis: a case study involving endangered species

    Water Resources Research

    (1987)
  • K.J. Boyle et al.

    What do we know about groundwater values? Preliminary implications from a meta-analysis of contingent-valuation studies

    American Journal of Agricultural Economics

    (1994)
  • Bright, P.W., Halliwell, E.C., 1999. Species recovery programme for the pine marten in England. English Nature Research...
  • D.S. Brookshire et al.

    Estimating option prices and existence values for wildlife resources

    Land Economics

    (1983)
  • R. Brouwer et al.

    Contingent valuation of the public benefits of agricultural wildlife management: the case of Dutch peat meadow land

    European Review of Agricultural Economics

    (1998)
  • R. Brouwer et al.

    A meta-analysis of wetland contingent valuation studies

    Regional Environmental Change

    (1999)
  • E.H. Bulte et al.

    Marginal valuation of charismatic species: implications for conservation

    Environmental and Resource Economics

    (1999)
  • Cited by (0)

    1

    Tel.: +34 981563100x14337.

    View full text