Economic valuation of species loss in the open sea

Ecological Economics 70 (2011) 729–739

Contents lists available at ScienceDirect

Ecological Economics j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e c o l e c o n

Analysis

Economic valuation of species loss in the open sea Adriana Ressurreição a,⁎, James Gibbons b, Tomaz Ponce Dentinho c, Michel Kaiser d, Ricardo S. Santos a, Gareth Edwards-Jones b a

Department of Oceanography and Fisheries, University of Azores, Cais de Santa Cruz 9901-862 Horta, Portugal School of the Environment, Natural Resources & Geography, Bangor University, Gwynedd LL57 2UW, United Kingdom Department of Agrarian Sciences, University of Azores, Terra Chã 9701-851 Angra do Heroismo, Portugal d School of Ocean Science, Bangor University, Menai Bridge, Anglesey LL59 5AB, United Kingdom b c

a r t i c l e

i n f o

Article history: Received 29 January 2010 Received in revised form 9 November 2010 Accepted 10 November 2010 Available online 16 December 2010 Keywords: Marine biodiversity Conservation WTP Azores

a b s t r a c t Although the oceans cover 70% of the surface of the planet few studies have considered the economic valuation of marine biodiversity, despite the importance of such information for marine management and conservation. This study uses a contingent valuation method to estimate the public's willingness to pay (WTP) to avoid loss in the number of marine species in the waters around the Azores archipelago. We estimated the marginal value associated with increased levels of species loss (10% and 25%) in five marine taxa (mammals, fish, algae, birds and invertebrates) and all marine species considered as a whole, via a face to face survey of residents and visitors to two Azorean islands. The results suggest small but statistically significant differences in the WTP to prevent losses in the different taxa (mammals= fishN birds = invertebrates= algae). The results also suggest a greater WTP to preserve all marine taxa as a whole, than for a series of individual marine taxa. The valuation of the ecosystem and taxa may be influenced by the maritime culture of the respondents, but despite this, the findings challenge the commonly held premise that charismatic taxa have a disproportionately strong influence on WTP, and they provide important insights into human preferences for biodiversity conservation. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Environmental valuation studies undertaken to date have tended to focus on terrestrial systems, whereas marine systems have received relatively little attention (Turner et al., 2003). Marine and coastal systems cover 70% of the surface of the planet and provide a wide range of direct, indirect and non-use values (Beaumont et al., 2007). A better understanding of the perceived value of marine systems would inform conservation and other management decisions that affect societal welfare. Not only are there relatively few studies on the value of marine biodiversity, but existing studies tend to be restricted with regard to the taxa and locations they consider. A taxonomic bias is illustrated by a recent meta-analysis of the economic values of species (Martín-López et al., 2008) which identified twelve studies on marine mammals (e.g. whales, otters and seals), six on fish, two on birds and one on reptiles (Table 1). A large body of literature has concluded a strong public preference for the allocation of conservation funds to likeable and charismatic individual species (White et al., 1997, 2001; Metrick and Weitzman, 1996; Loomis and White, 1996). However, the strength of this preference may be overestimated by the bias in valuation studies which are largely based on single and high profile species, and greater attention should be given to ⁎ Corresponding author. Tel.: +351 292 200 400; fax: +351 292 200 411. E-mail addresses: [email protected] (A. Ressurreição), [email protected] (J. Gibbons), [email protected] (T.P. Dentinho), [email protected] (M. Kaiser), [email protected] (R.S. Santos), [email protected] (G. Edwards-Jones). 0921-8009/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ecolecon.2010.11.009

ensure that wider assessments are conducted. The valuation of less charismatic species is important in order to provide relevant data on all elements of marine ecosystems to decision-makers. The current bias towards charismatic species in the literature may lead decision-makers to make inappropriate allocation of funds for conservation which could lead to a failure to meet conservation goals. The geographic bias in the literature on the value of marine biodiversity arises as much of the research in this field is U.S. focussed (Martín-López et al., 2008), with few studies being reported for Europe, and almost none in southern Europe. This is unfortunate as current European legislation requires that the value of biodiversity is included in conservation and management decisions (e.g. the Marine Strategy Framework Directive). The lack of valuation studies that consider all taxa and represent the cultural and ecological diversity across Europe presents serious challenges to decision makers seeking to establish mechanisms for marine conservation within the European Union. The present study addresses some of the issues introduced earlier using a contingent valuation method to estimate the public's WTP to avoid marine species loss in the waters around the Azores archipelago. Coverage of marine taxa is broad and the aim was to estimate the marginal value associated with increased levels of species loss (reduction in species richness) and also to estimate WTP to avoid the loss of species in different marine taxa. Finally we tested for differences in the WTP between residents and visitors to the Azores. To the best of our knowledge this is the first time these issues have been tested in the context of marine ecosystems.

730

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

Table 1 Summary of published economic values of marine species(a). Taxa

Commom name

Scientific name

Mean value(b) (US$2005)

Reference

Mammal Mammal Mammal Mammal Mammal Mammal Mammal Mammal Mammal Mammal Mammal Mammal Birds Birds Reptile Fish Fish Fish Fish Fish Fish

California Sea Otter European Otter Gray seals Hawaiian monk seal Mediterranean monk seal Northern elephant seal Steller sea lion Beluga whale Blue whale Bottlenose dolphin Gray whale Humpback whale Harlequin Duck Whooping Crane Loggerhead sea turtle Atlantic salmon Chinook salmon Steelhead Shortnose sturgeon Kelp bass White croaker

Enhydra lutris nereis Lutra lutra Halichoerus grypus Monachus schauinslandi M. monachus Mirounga angustirostris Eumetopias jubatus Delphinapterus leucas Balaenoptera musculus Tursiops truncatus Eschrichtius robustus Megaptera novaeangliae Histrionicus histrionicus Grus americana Caretta caretta Salmo salar Oncorhynchus tshawytscha O. mykiss Acipenser brevirostrum Paralabrax clathratus Genyonemus lineatus

36.76 24.4 13.81 12.83 93.87 17.54 31.53 73.83 14.2 44.57 23.17 128.34 11.15 53.42 16.98 9.45 126.66 64.47 30.86 43.35 43.35

Hageman, 1985, 1986 White et al., 1997 Bosetti and Pearce, 2003 Samples and Hollyer, 1990; Brown et al., 1994 Langford et al., 1998 Hageman, 1986 Giraud et al., 2002 Tkac, 1998 Hageman, 1985, 1986; Bulte and van Kooten, 1999 Hageman, 1986 Hageman, 1985, 1986; Loomis and Larson, 1994 Samples et al., 1986 Tkac, 1998 Bowker and Stoll, 1988 Whitehead, 1992; Wilson and Tisdell, 2003 Stevens et al., 1991; Bulte and van Kooten, 1999 Hanemann et al., 1991; Olsen et al., 1991 Olsen et al., 1991 Kotchen and Reiling, 1998 Carson et al., 1994 Carson et al., 1994

(a)

Adapted from Martín-López et al. (2008). Values refer to the mean of values from the studies cited in Reference column.

(b)

1.1. The Azores Case Study The Azores archipelago (36°–40°N, 24–32°W) is composed of nine volcanic islands and several small islets, scattered in three main groups along 600 km of the northern part of the Mid Atlantic Ridge. The coastline is approximately 790 km in length (Menezes, 2003), and the marine environment of the Azorean archipelago and its surrounding Economic Exclusive Zone covers roughly 1 million square kilometres, with an average depth of 3000 m. The Azores archipelago is of

considerable conservation and marine biological interest due to its isolated position in the middle of the north eastern Atlantic and its relatively young age (Santos et al., 1995). Owing to the impracticality of undertaking a valuation survey in all islands due to time and budget constraints, two islands of the central group — Pico and Faial — were chosen as the location for this study (Fig. 1). These islands are closely associated both physically and socially; they are divided by a strait 8.3 km long and have a strong interchange of residents and visitors. Communities on both islands have a tradition of engaging in activities

Fig. 1. Azores archipelago.

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

related to marine biodiversity such as whale watching, scuba diving, professional and recreational fisheries and scientific research. 2. Contingent Markets and Marine Biodiversity Conservation: Methodology Contingent valuation is one of the most accepted methods for assessing the total economic value (TEV) of species (Richardson and Loomis, 2009). Contingent valuation offers flexibility as it is (i) capable of capturing all components of TEV including non-use values; (ii) allows the valuation of environmental changes that have not yet occurred (iii) provides a full socio-demographic profile of the target population (iv) allows contingent scenarios to be designed to directly elicit the value of the change under scrutiny and (v) allows a better alignment of public expectations and political initiatives — consistent with the democratic perspective — as the valuation process is submitted to public discussion. Although widely used, contingent studies are the subject of strong criticism related to the validity and reliability of the results (Venkatachalam, 2004; White et al., 2001 page 163). Extensive investigation has been dedicated to the identification and quantification of potential biases in CVM (for review see Venkatachalam, 2004); such as hypothetical bias (potential divergence between the real and hypothetical payments (Cummings et al., 1986)) and embedding or scope effect (valuation being rather insensitive to the scale of the physical change (Nijkamp et al., 2008)). Contingent markets are typically defined according to three key elements: (i) the valuation scenarios associated with different levels of the goods or service to be valued; (ii) the identification of the target population that are asked to pay for the goods or service to be valued; and (iii) the design of the questionnaire that defines the manner in which the goods or service are provided. 2.1. The Definition of Valuation Scenarios The economic valuation of changes in biodiversity can be affected by critical choices regarding the level of biological diversity portrayed to respondents, the scale of the change in that biodiversity and the type/amount of information conveyed as part of the survey. As such it is a challenging task to devise survey instruments that enable the average citizen to understand and appreciate the consequent welfare implications of changes in marine biodiversity. Christie et al. (2006) provided an extensive review on how valuation of biodiversity has been operationalized in previous studies: single species, multiple species, genetic diversity, diversity within habitats, ecosystem functions and services are some of the options presented in this review. Many of these ecological concepts such as genetic diversity and ecosystem function are complex and it is unlikely that the general public has a complete understanding of them (Christie et al., 2006; Nunes and van den Bergh, 2001). Consequently, the value of changes in biodiversity has often been assessed through a simplified approach using individual species as proxies of biological diversity. Although the public may find it easier to understand biodiversity conservation in such terms, the science of nature conservation is moving away from single species approaches and towards community or ecosystem-based approaches. Economic valuation of biodiversity is also conditioned by the scale of the hypothesised change in that metric. Valuation is most meaningful when considering small or marginal changes (Turner et al., 2003). Hence, in the present study, in order to minimize the embedding effect, the scale of the change was considered to occur at a regional scale (Turpie, 2003). The valuation scenarios considered marine biodiversity in terms of species richness for five separate taxa: marine mammals, sea birds, fish, invertebrates and algae. We assessed the willingness to pay (WTP) expressed by members of the general public to prevent a decline in these five taxa from their current level of species richness (Table 2). Respondents were randomly assigned to one of two groups,

731

Table 2 The current level of species richness for 5 different marine taxa in the Azores. Marine Taxon

Current level

Fish Sea mammals Algae Sea birds Invertebrates

520 24 368 56 1700

Porteiro (2007) Gonçalves (2007) Neto et al. (2006) Clarke (2006) Gonçalves (2007)

Personal communicationa Personal communication

Personal communication

The species list for the Azores is not yet complete, and new species are being discovered in the region on a regular basis. a The personal communications were given by marine experts in the University of the Azores.

and were asked to express a value to avoid either a 10% or 25% reduction in species richness for each taxon. The split sample design reduced the cognitive burden of the respondents and decreased the chance of dependencies between the WTP responses for different levels of species loss. By using this approach, the responses are statistically independent and hence this increases greatly the strength of any predictive relationship between WTP and percentage of species loss. Initial valuation scenarios defined for use in the survey were discussed with an interdisciplinary research team sponsored by EU Network of Excellence: Marine Biodiversity and Ecosystem Function (MarBEF), during a training course on valuation methodologies for marine environments. Comments from these experts served to improve and refine the scenarios, the instrument survey and the sampling strategy before the pilot study was conducted. 2.2. Identification of Target Population Issues relating to the identification of the human population influenced by changes in biodiversity are contested in the literature. Nijkamp et al. (2008) suggested that loss of biodiversity can impact the welfare of distant populations; hence assessments should extend beyond the local and towards the global scale. In contrast, Whitehead et al. (1995) differentiated between three types of respondents; on-site users, off-site users and non-users. These definitions were based on their familiarity with the resource such that WTP is more reliable for on-site users, since non-users do not consider their income constraints when stating their willingness to pay. Paradiso and Trisorio (2001) also advocated that the validity of the CVM estimates improved with direct knowledge of the good and that this reduces the disparity between hypothetical and real WTP. Boyle et al. (1994) claimed that biases, such as embedding effects are more likely to occur among non-users. Given these insights, in this study we interviewed on-site users (residents and visitors) since both constitute the primary population affected by the loss of marine biodiversity around the Azorean islands. 2.3. Questionnaire Design The questionnaire was divided into 6 parts (A–F) and comprised twenty eight questions in total. The questions were designed to examine respondents' knowledge about marine biodiversity, details of their visit (if relevant), respondents' general behaviour and attitudes towards the conservation of marine biodiversity, the economic valuation exercise and socio-demographic information. Section A contained a multiple-choice question which included formal definitions of the ecological terms ‘ecosystem,’ ‘species richness’ and ‘marine biodiversity.’ Respondents were asked to choose the right definition of marine biodiversity amongst these alternatives (including “don't know”). This had the purpose of obtaining an initial indicator of the respondent's knowledge on biodiversity and to encourage further discussion if the respondent had any doubts or misconceptions on this subject. As respondents were preparing to answer the valuation

732

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

exercise, they were informed about the number of species currently present in each marine taxon in the Azores, as well the number of species that hypothetically would be lost from each taxon under a scenario of a 10% and 25% reduction in species. The questionnaire also included explicit warnings about budget constraints and substitutes that were designed to mitigate positive hypothetical bias in valuation. As recommended by Arrow et al. (1993) a series of follow up questions were asked to determine the motives behind the responses and to diagnose if those refusing to pay indicated a valid representation of their value (genuine zero bidders) or reflected a protest about some feature of the simulated market (protest responses) (Mitchell and Carson, 1989). The role of non-use and use values in supporting the conservation of marine species was examined by presenting the respondents with several statements that describe possible reasons for valuing marine biodiversity. These statements corresponded to direct, indirect, existence, option and bequest values.1 In addition respondents also had the opportunity to use their own words to justify their reasons for paying to prevent the loss of marine species. Given society's general unfamiliarity with hypothetical transactions involving marginal changes in marine biodiversity, a payment card was adopted as the elicitation format. Payment cards avoid the high rate of item non-responses more common in open-ended valuation questions (Cameron and Huppert, 1989; Veisten et al., 2004). Possible negative effects of using payment cards are that the range of values, anchoring effects and size of intervals displayed on the card may influence the WTP responses (Cameron and Huppert, 1989). In order to mitigate some of these problems the range of values displayed on the payment card were based on the results of the pilot survey (see Section 2.5). This resulted in a payment card with a range of choices from 0€ to 5000€, displayed randomly. Each respondent was asked five different valuation questions, one each for marine mammals, seabirds, fish, marine invertebrates and algae. The order in which the questions are asked can lead to question-order bias (Mitchell and Carson, 1989) or sequencing effect (Cummings et al., 1986). In the current study, the sequencing of the five taxa was randomised for each questionnaire. Assuming a sequencing effect and no interaction between position of presentation and taxa this approach reduces the bias and increases the variance of the WTP estimate for each taxon. We believe that this is a reasonable compromise i.e. it seems plausible that individuals may increase or decrease their bid under repeated questions but less plausible that the direction or magnitude of the change would be influenced by taxa. Following Mitchell and Carson (1989) two further procedures were also adopted to mitigate this effect (i) respondents were informed about the survey structure before asking the WTP questions and (ii) during the interview respondents were permitted to change their answers. The exact wording of the WTP question for one of the marine taxons (fish for instance) was as follows: “First let's consider fish, what would be the maximum amount you would be willing to pay, in a once only payment, to a conservation trust, in order to avoid a decline in the number of species of fish by 10%?”. Photomontages that depicted representative organisms of each marine taxon being valued were shown to respondents during the survey. These photomontages were used merely as information tools and were not intended to represent the different levels of biodiversity decline. Twenty photos of species in each taxon were presented and 1 The exact wording of the claims presented to the respondents to justify their support towards the conservation of marine species was: i) The marine biodiversity of this region provide an option for leisure/recreation, food provision and it is important for the local economy; ii) The marine biodiversity of this region is precious and it benefits the human well being (water quality, erosion control, coastal protection, etc.); iii) I enjoy knowing that marine biodiversity exists in this region even if I never see or use it: iv) The marine biodiversity of this region has the right to exist even if it does not appear important to human well being today; v) I enjoy knowing that the future generations will be able to enjoy marine biodiversity of this region; vi) Other (please specify).

each photo represented one individual per species. Each taxon was represented by a range of species, and care was taken to avoid only including the charismatic, valuable, beautiful or large species (for example see Fig. 2).

2.4. Study Design and Survey Implementation The original CVM questionnaire was piloted with a random sample of 64 residents and visitors from the Pico and Faial islands. This study enabled evaluation of the acceptability of the payment method and the plausibility of the valuation scenarios. The pilot results suggested that a once only payment to a neutral trust fund was the most favoured option presented.2 The pilot study also highlighted the desire of most respondents to value changes in the number of all marine species (i.e. to undertake an assessment of the whole ecosystem) rather than be confined to valuing only changes in the number of species of specific marine taxa. This resulted in the addition of an additional question that offered respondents the option to value all species in the marine ecosystem around Pico and Faial islands. The format of this additional valuation question was left open-ended (OE) as the potential range of bids had not been pre-tested. Different elicitation methods may, in theory, influence the WTP responses. Because of this theoretical possibility caution must be taken when making direct comparisons between WTP derived for taxa using the payment card and for the entire marine ecosystem, which was elicited with an open ended bid question. While this issue does complicate interpretation of the results, those respondents who participated in the pilot study explicitly expressed a desire to value the entire ecosystem. It was for this reason that the open-ended question about ecosystem values was included in the final questionnaire. Residents were randomly recruited from among pedestrians in market places, the airport, commercial ports, streets, marinas on Pico and Faial islands. However, in order to ensure that visitors were interviewed only after their visit, the majority of the interviews were carried out in the departure lounge of the airport on Faial. Interviews were carried out in Portuguese for national citizens and in English for foreign people. The majority of the interviews were performed when respondents were apparently relaxed, unoccupied or waiting for transportation. At the beginning of the survey respondents were asked to identify themselves as residents or visitors and the sampling procedure was defined to allow any adult who claimed to be a household head to be the spokesperson for the household. Respondents were randomly allocated to one of two groups relating to the scenarios that valued a 10% species loss and a 25% species loss. A total of 507 (255 visitors and 252 residents)3 interviews were undertaken by three trained interviewers over 20 days between August and September 2007. The overall response rate was greater than 90%.

2.5. Data Analysis Differences in demographics between respondents of the four subsamples were tested using a Mann–Whitney test. Of the 507 respondents 392 (77%) were willing to pay and 115 (23%) refused; 66 (13% of the total responses) of the refusals were considered protest responses and 49 (10% of the total responses) genuine zero bidders. As is common in most CVM studies protest responses were excluded from the analysis of WTP (Adams et al., 2008; Jones et al., 2008; 2 The frequency and payment vehicles pre-tested in the pilot survey included: (i) an increase in your income tax (2%); (ii) creation of an extra regional tax (2%); (iii) an annual donation to a regional NGO (33%); (iv) a donation, in a once only payment, to a regional NGO (10%); (v) a donation to a neutral trust fund (53%) and (vi) Other (0%). 3 The sample was split into 4 subsamples accordingly: residents 10% (n = 126); residents 25% (n = 126); visitors 10% (n = 126) and visitors 25% (n = 129).

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

733

Fig. 2. Photomontage presented for fish.

Kotchen and Reiling, 2000; Mitchell and Carson, 1989; Spash et al., 2009; Zhongmin et al., 2003). Data analysis followed the approach of Cameron and Huppert (1989) and estimated log(WTP) from socio-demographic and attitudinal variables. The use of the payment card meant that the respondents' true valuation (Yi) is unobserved but lies between the payment card limits tli and tui, hence log (Yi) lies between log (tli) and log (tui). Assuming that log (Yi) is a linear function of individual covariates (xi), then log (Yi) = xi′β + ui with ui normally distributed with mean 0 and standard deviation σ. The alternative approaches of fixing Yi at the payment card intervals or the midpoint between them introduces bias in the coefficient estimates Following Cameron and Huppert (1989) we avoided this bias by estimating the parameters that maximised the likelihood function: n

L= Π ϕ i=1




 
log ðtui Þ−xi′β = σ −ϕ log ðtli Þ−xi′β =σ

Where ϕ is the cumulative standard normal density function. Genuine zero bidders were assigned a value on the interval [0; 0.01]. If σ N 0 then estimating E[Yi] from exp(log(Yi)) is biased, being closer to the median. These “conditional” medians can be considered a valid measure of the central tendency of the unobserved dependent variable Y (Cameron and Huppert, 1989). To estimate Yi from estimates of log(Yi) we used the non-parametric smearing estimate proposed by Duan (1983). 3. Results 3.1. Empirical Findings 3.1.1. Socio-demographic Profile of the Sample The socio-economic profile of the sample was calculated using the entire sample of 507 questionnaires since the protest responses and/or null WTP do not interfere with the results (Adams et al., 2008). Non-parametric tests for differences in means did not suggest any significant differences between the profiles of respondents of the two

survey versions for the level of loss within each group (visitors: age = 0.680; household= 0.292; income = 0.340; education = 0.535; residents: age = 0.423; household = 0.424; income = 0.954; education = 0.468) nor even between the profiles of respondent groups for each survey version (age p = 0.405; household p = 0.202; income p = 0.545; education p = 0.487). This is expected as respondents were randomly allocated to one of the two questionnaire alternatives. As there were no demographic differences between these two sample groups, it was possible to evaluate the effect of the degree of loss on WTP. Respondents had an average income of 2345€ per month per household (visitors 2894€; residents 1789€: P b 0.001). The level of education profile revealed that 27% of the respondents were educated to the elementary level, 29% obtained a high school diploma, 37% were undergraduate and 7% had postgraduate level education. Visitors had a higher level of education than residents (visitors 2.6; residents 1.9: P b 0.001). In general, the size of the household varied between 1 and 4, with a mean of 2.9 per household (visitors 2.8; residents 3.1; P = 0.002). About 49% of those interviewed were male and the mean age of the sample was 39 years (visitors 39.5; residents 37.7: P = 0.139). Visitors had been to Pico and Faial islands a mean of three times prior to the survey and the standard duration for the visit was two weeks. The main motivation for visiting these islands was (i) local landscape and natural features (ii) peacefulness and (iii) aquatic activities (e.g. whale watching, scuba diving, snorkelling, recreational fishing, etc.). A common statement from many visitors was, that even though marine biodiversity had not been a major factor in deciding to visit the Azores, during the course of their visit they had realised that the diversity of marine life in the Azores was one of the most impressive features of the islands. Surprisingly, despite the complexity of the concept of biodiversity, approximately 73% of the respondents chose the correct definition of marine biodiversity among the available options. 3.1.2. Acceptance of the Valuation Scenario In total 84% of respondents declared that marine biodiversity conservation should be a priority for governments at a national and

734

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

Overall, the analysis of the stated WTP distribution suggested that some respondents expressed what seem to be implausibly high WTPs. While the presence of outliers may increase average WTP dramatically leading to an overestimation of the worth of a public good (Chaudhry et al., 2007), identifying the differences between a genuinely well thought-out high WTP and an erroneous high WTP is a subjective process. In an attempt to obtain conservative estimates of the WTP we removed respondents who had bid above a threshold that we felt was unrealistic, and valuation functions were re-estimated. The threshold for removal was when the WTP bid was greater than 1% of household monthly income (equivalent to approximately 0.083% of the annual salary) multiplied for a working life period of 40 years (in order to reflect the one-off nature of the bid process) (i.e. for a person on a monthly salary of US$2000 the threshold that would trigger rejection from the analysis would be a WTP bid greater than US$800). We acknowledge that the validity of this, and any other subjective threshold, is subject for debate. Applying the threshold for outliers resulted in 67 respondents being dropped from the database. The elimination of extreme observations was based on α-trimmed means proposed by Mitchell and Carson (1989) as robust estimators with OE CVM data in the presence of outliers. The main difference lies on the correction criterion, which is a probability level for Mitchell and Carson and a bid/income level in our study.

global level. Of the 392 respondents that expressed a positive willingness to pay, at least 71% declared both use and non-use values to support marine species preservation. 3.1.3. Reasons for rejection of the Valuation Scenario and Protest Responses One hundred and fifteen respondents refused to pay for species preservation in the Azores. Analysis of these refusals was based on the follow up questions that showed that among genuine zero bidders, 23 respondents considered that society has more important problems than protecting marine biodiversity, 20 could not afford to pay, while only six respondents declared that they would not receive sufficient benefit from any such payment. The examination of protest responses revealed that the majority of the refusals (46 responses) were based on the belief that environmental conservation is the responsibility of the government rather than the individual. The remainder of the protest responses reflected disagreements with particular elements of the valuation process such as: insufficient information (two responses), objection towards the valuation question (five responses), refusal to put a price on marine biodiversity (eight responses), while five respondents stated “other” reasons to protest against the valuation survey. These protest responses, although representing valid concerns, are usually not considered as valid representations of the individual WTP (Giraud et al., 2002).

3.3. Test of Theoretical Validity: The Bid Curve for Marine Species Conservation

3.2. Analysis of Stated Willingness to Pay Statistical regression models that express the functional relationships between willingness to pay responses and other variables that normally affect demand are commonly known as bid curves or valuation functions (Jones et al., 2008). Three categories of variables were included in the regression models based on theoretical expectations from classical economics (Samnaliev et al., 2006): (i) valuation scenario (marine taxa and level of decrease); (ii) attitudes and concerns towards marine biodiversity conservation and (iii) socio-demographic characteristics.

The WTP bids for specific taxa were similar in their distribution (Fig. 3). The distribution of the WTP data for individual taxa was truncated to zero and the maximum amount pledged for the preservation of a specific marine taxon was 5000€. Data on the value of all marine species, which was given in response to an open-ended question, was relatively more skewed to the right and the maximum value was 50,000€, with a mode of 500€. 70 60

Fish

Frequency

50

Mammals

40

Algae 30

Birds

20

Inverts

10 5.000 €

3.000 €

2.500 €

2.000 €

1.500 €

500 €

1.000 €

250 €

200 €

175 €

150 €

125 €

100 €

75 €

50 €

40 €

30 €

25 €

20 €

15 €

10 €

5€

2€

0€

1€

0

WTP bid amount (lower bounds) - PC data 60

All Azorean marine species

Frequency

50 40 30 20 10

0 2 3 5 10 15 20 25 40 50 60 70 75 80 100 120 125 150 175 200 250 255 275 300 350 400 500 600 625 650 700 750 1.000 1.200 1.250 1.500 2.000 2.500 2.900 3.000 3.200 5.000 7.500 10.000 15.000 25.000 30.000 50.000

0

WTP (€) bid amount - Open Ended data Fig. 3. The frequency distribution of respondents by bid amount for specific marine taxon groups (payment card data) and for all marine species preservation (open ended data).

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

All the demographic variables, with the exception of occupation, were included as continuous variables. Log income was included as a linear term and the other demographic continuous variables as linear and quadratic terms allowing a curvilinear relationship between these variables and the dependent variable. Hence we assume that log (WTP) is a smooth continuous function of the demographic data while, for model parsimony, restricting the parameter number. For the ordinal variables age data were included as mid-points and education levels from 1 (basic) to 4 (postgraduate). Following Cameron and Huppert (1989) a maximum likelihood interval regression approach was used. The models were fitted using the survival package in R (R Development Core Team, 2008). Two separate valuation functions were estimated and presented, one for the payment card (PC) data regarding the valuation of specific marine taxa and the other for the open ended (OE) value responses for the preservation of all marine species. Table 3 shows the parameter estimates for the bid function for the PC data. Interval regression parameter estimates can be interpreted as in OLS regression (Blaine et al., 2005). The results showed that significant differences occurred between WTP values for algae and fish and between algae and marine mammals (pb 0.05). No significant differences were found between bids for birds, marine invertebrates and Table 3 The factors influencing the WTP responses to avoid two levels of species loss of five marine taxa: interval regression results — payment card data. Dependent variable: WTP to prevent a decline in specific marine taxon X2 = 726.23 D.F: 33 p=0 Scale = 1.63

Number of observations: 1870 Log-likelihood model: −5149.4 Log-likelihood (intercept only): −5512.5 Log normal distribution

p

Sig.c

Variable labels

Parameter estimates

Standard z error

Intercept 25% or 10% level of species loss (1.0) Scenario: birdsa Scenario: fisha Scenario: invertsa Scenario: mammalsa Log monthly income (€) Male/female (1.0) Fisherman Public employee Private employee Self employed Student Retired Unemployed Homemaker Household — linear term Household — quadratic term Age — linear term Age — quadratic term Resident/visitor condition (1.0) Biodiversity not priorityb Biodiversity not importantb Don't knowb Education — linear term Education — quadratic term Read/TV env. conservation (1.0) Recycling of household goods (1.0) Subscribe a magazine (10) Products env. friendly (1.0) Donations for charities (1.0) Actively campaigned (1.0) Membership conserv. group (1.0) Log scale

−2.81 0.21

0.72 0.08

−3.89 −2.59

0.000 *** 0.010 ***

0.07 0.26 0.03 0.25 0.88 0.35 1.01 −0.03 −0.02 −0.18 −0.53 −0.15 0.65 0.55 −0.13 0.00 −0.06 0.00 0.24 −1.77 −2.77 −0.43 −0.03 0.07 0.21 −0.12 −0.33 0.42 0.03 0.55 0.27 0.49

0.12 0.12 0.12 0.12 0.07 0.08 0.31 0.24 0.24 0.25 0.29 0.33 0.45 0.33 0.11 0.02 0.02 0.00 0.09 0.13 0.45 0.33 0.24 0.05 0.12 0.10 0.13 0.09 0.13 0.22 0.18 0.02

0.59 2.14 0.23 2.09 1.22 4.21 3.22 −0.13 −0.08 −0.71 −1.81 −0.44 1.46 1.64 −1.16 0.20 −2.73 2.01 2.55 −1.37 −6.10 −1.32 −0.11 1.33 1.78 −1.16 −2.56 4.67 0.25 2.47 1.51 26.31

0.554 0.032 0.819 0.037 0.000 0.000 0.001 0.899 0.938 0.476 0.070 0.659 0.145 0.100 0.246 0.843 0.006 0.044 0.011 0.000 0.000 0.188 0.916 0.185 0.076 0.245 0.010 0.000 0.803 0.014 0.131 0.000

n.s. ** n.s. ** *** *** *** n.s. n.s. n.s. * n.s. n.s. n.s. n.s. n.s. *** ** ** *** *** n.s. n.s. n.s. * n.s. ** *** n.s. ** n.s. ***

a The coefficients for the specific marine taxon are relative to taxon Algae which implicitly has a coefficient of zero. b The coefficients are relative to respondents that consider that biodiversity conservation should be a priority for governments. c p b 0.01 (***); p b 0.05(**); p b 0.1 (*); n.s.: non significant.

735

algae. Respondents seemed to distinguish between different levels of species loss and to reflect it in the valuation function of specific marine taxa. Log Income had a significant positive effect on log WTP, consistent with the standard theory of budget constrained consumer choice. Men had a significantly higher WTP than women. Previous findings from CVM studies indicated that the impact of gender on WTP is mixed (Berrens et al., 1997; Bord and O'Connor, 1997; Brown and Taylor, 2000; in Birol et al., 2006). From the nine professional occupation categories (farmer was the baseline category) only two were significant predictors of WTP. Respondents that were fishermen had a significantly higher WTP compared with other professions (Table 3). This finding is not surprising given that a decrease in marine species richness may have a direct impact on fishermens' livelihoods. In contrast, the occupation “student” had a significant and negative impact on WTP. The results suggested that age has a negative effect on WTP as found elsewhere (Adams et al., 2008; Chaudhry et al., 2007; Jorgensen et al., 2001). The results also suggest that the number of the household members and level of education are not significant predictors of WTP. The regression results highlighted significant differences between the WTP of residents and visitors. Although there is evidence that visitors' incomes and education were higher than those of residents (Section 3.1.1), when comparing residents and visitors with the same socio-demographic profile, residents were more likely to attach higher values to prevent species loss in the sea around Pico and Faial islands. In this case, the degree of attachment to the place is probably the main driver in valuation. In order to preserve local marine species, residents are willing to forego a greater proportion of their income than visitors (Table 3). Respondents that engaged in positive environmental behaviours or attitudes tend to give higher values for environmental conservation than non-environmentalists (Carson et al., 2001; Kotchen and Reiling, 2000). Of the behaviours associated with the respondents' environmental awareness or commitment tested in our study, three had a positive and significant effect on WTP: (i) ‘read or watch TV about environmental education’; (ii) ‘selected one product over another because it's more environmentally friendly’ and (iii) ‘actively campaigned about an environmental issue.’ The behaviour ‘subscribed to a magazine concerned with environment conservation’ had a negative effect on WTP. There is no apparent explanation for this result. The questionnaire also explored peoples' views on marine biodiversity conservation. The possible alternative responses ranged between considering biodiversity conservation as (i) a priority for governments; (ii) important but not a priority and finally (iii) as not important. A ‘don't know’ category was also specified. As expected those respondents that did not regard biodiversity conservation as a priority were less likely to pay for marine species preservation. Overall the parameter estimates are consistent with economic theory which supports the theoretical validity of the WTP estimates. Table 4 includes the results of the open-ended portion of the survey to prevent a loss in the number of all marine species in the region. In general, similar patterns are found on the effect of the socio-demographic variables on the WTP: log income and gender have a positive and significant impact while age had a predominantly linear and negative effect. However, none of the professional occupations had a significant influence on WTP and there were no significant differences between residents and visitors which suggested that there were no intrinsic differences beyond the other demographic variables. Although the WTP of respondents to avoid 25% of loss of all marine species was greater than to avoid a 10% loss, the trend was not significant (Table 4), a result which merits further investigation. The consideration of all marine species may render the valuation exercise too complex for respondents to translate marginal changes into WTP values. This scope insensitivity may be affected by cognitive limitations, but also by warm glow (charitable behaviour) (Kahneman et al., 1993) or fixed expenses constraints (respondents identify an amount from their budget that they feel that they can afford to spend on the good in question which is invariant with the amount of the good offered) (Chilton and Hutchinson, 2003).

736

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

Table 4 The factors influencing the WTP responses to avoid two levels of species loss of all marine species: interval regression results — open ended data. Dependent variable: WTP to prevent a decline in all marine species X2 = 156.71 D.F: 29 p=0 scale = 2.02

Number of observations: 374 Log-likelihood model: −3834.8 Log-likelihood (intercept only): −3913.2 Log normal distribution Variable labels

Parameter estimates

Standard error

z

p

Sig.a

Intercept 25% or 10% level of species loss(1.0) Log monthly income (€) Male/female (1.0) Fisherman Public employee Private employee Self employed Student Retired Unemployed Homemaker Household — linear term Household — quadratic term Age — linear term Age — quadratic term Resident/visitor condition (1.0) Biodiversity not priorityb Biodiversity not importantb Don't knowb Education — linear term Education — quadratic term Read/TV env. conservation (1.0) Recycling of household goods (1.0) Subscribe a magazine (1.0) Products env. friendly (1.0) Donations for charities (1.0) Actively campaigned(1.0) Membership conserv. group (1.0) Log scale

−2.28 0.24 1.17 0.45 0.91 −0.41 −0.41 −0.47 −1.16 −0.90 0.50 0.39 −0.09 −0.01 −0.12 0.00 0.38 −2.34 −3.99 −0.64 0.08 0.06 0.47 −0.20 −0.22 0.47 −0.18 0.76 0.11 0.70

1.98 0.22 0.20 0.23 0.86 0.67 0.66 0.70 0.81 0.92 1.24 0.92 0.32 0.04 0.06 0.00 0.26 0.35 1.25 0.90 0.66 0.14 0.33 0.27 0.36 0.25 0.36 0.61 0.50 0.04

−1.16 −1.10 5.85 1.95 1.06 −0.60 −0.62 −0.68 −1.44 −0.99 0.41 0.42 −0.30 −0.30 −1.82 1.48 1.45 −6.61 −3.21 −0.71 0.13 0.45 1.41 −0.74 −0.62 1.89 −0.49 1.24 0.23 17.14

0.248 0.273 0.000 0.051 0.290 0.546 0.535 0.498 0.149 0.323 0.684 0.675 0.765 0.762 0.070 0.138 0.148 0.000 0.001 0.479 0.900 0.656 0.160 0.460 0.537 0.059 0.627 0.215 0.818 0.000

n.s. n.s. *** * n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. * n.s. n.s. *** ** n.s. n.s. n.s. n.s. n.s. n.s. * n.s. n.s. n.s. ***

a b

p b 0.01 (***); p b 0.05(**); p b 0.1 (*); n.s.: non significant. The coefficients are relative to respondents that consider that biodiversity conservation should be a priority for governments.

3.4. Individual WTP Estimates for Hypothetical Scenarios of Loss in Marine Species Applied welfare economics does not exclusively require the use of sample mean values (Cameron and Huppert, 1989). When σ N 0 mean WTP and median WTP will take different values. Some authors have suggested that the median, rather than the mean, should be used as an indicator of the central tendency of WTP, as the former is more robust to outlier observations (Hanemann, 1984; Harrison and Kriström, 1995). However, although the median does have this desirable statistical property, mean WTP is more appropriate for benefit–cost analysis (Loomis and White, 1996). For the perspective of the decision-makers that wish to make an option based on efficiency criteria, the mean is then the most appropriate measure (Brent, 1998). In this study both mean and median WTP were estimated as central tendency measures of WTP. As significant differences were found between the levels of loss and between visitors and residents, means and medians were calculated separately for each group within the sample (Table 5). Overall, the results of this contingent valuation study indicated that the visitors and residents of Pico and Faial islands attached positive and significant values to the conservation of marine species. Although different payment vehicles were used, which restricts the direct comparison of results, it is interesting to note that the willingness to pay to avoid a loss in the number of all marine species (ranging from 405€ to 605€/only payment) was higher than the sum of the individual taxa (for individual marine taxa the maximum payments range from 45€ to 100€). This suggests respondents valued the conservation of the ecosystem as a whole rather than simply as

the sum of its components. These values may be interpreted as lower bound estimates since respondents were faced with hypothetical losses in species richness without prior information of either the particular stresses affecting the marine environment of the region or Table 5 WTP point estimates for hypothetical scenarios of species loss. Valuation scenario

Level of loss

Sample group

Median

Mean

Algae Birds Fish Inverts Mammals All marine species Algae Birds Fish Inverts Mammals All marine species Algae Birds Fish Inverts Mammals All marine species Algae Birds Fish Inverts Mammals All marine species

10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% 25% 25% 25% 25% 25% 25% 25% 25% 25% 25% 25% 25%

Visitors Visitors Visitors Visitors Visitors Visitor Residents Residents Residents Residents Residents Resident Visitors Visitors Visitors Visitors Visitors Visitor Residents Residents Residents Residents Residents Resident

23 € 25 € 30 € 24 € 30 € 138 € 16 € 17 € 20 € 16 € 20 € 96 € 27 € 29 € 35 € 28 € 35 € 158 € 18 € 19 € 23 € 18 € 23 € 110 €

66 71 86 68 85 581 45 48 58 46 58 405 77 83 100 79 99 665 51 55 66 52 66 463

€ € € € € € € € € € € € € € € € € € € € € € € €

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

the degree of endangerment concerning species; aspects that usually affect the public's investment in biodiversity conservation (Bandara and Tisdell, 2005; Loomis and white, 1996). 4. Discussion and Conclusions This study was designed to estimate the public's willingness to pay (WTP) to avoid loss in the number of marine species in the waters around the Azores archipelago and to investigate people's knowledge, perceptions and economic preferences for marine conservation. This involved critical choices regarding the level of biological diversity, the scale of the change and the information conveyed to the respondents. The results suggested that, contrary to the statement of Christie et al. (2006) that the “general public has a low level of understanding of what biodiversity is and why it matters,” in this survey the vast majority (73%) of respondents chose the correct definition of marine biodiversity from among the available options. Although, it should be noted, that choosing the right definition among a defined set of options is a simplified exercise and respondents may tend to choose the most complete or complex one. Such a choice may not mean that they possess a real understanding of the concept of biodiversity, or even complete awareness of why conservation of biodiversity is important to human welfare; nevertheless it is an indicator that the society is aware that biodiversity is much more expansive concept than simple measures of species richness. Both use and non-use values were referred to by respondents when forming their willingness to pay for marine biodiversity conservation. This indicates that valuations of aspects of biodiversity should examine all components of total economic value rather than individual elements of TEV. In addition, the results demonstrated that the majority of respondents are willing to pay for marine biodiversity conservation, and that respondents generally acknowledged that marine biodiversity conservation should be a priority for governments. The present study provided important new insights into human preferences for aspects of biodiversity. Previous studies have shown that likeable, valuable and charismatic species dominate the WTP for species preservation (White et al., 1997, 2001; Metrick and Weitzman, 1996; Loomis and White, 1996). The present results demonstrate that although there were significant differences in the valuations of marine mammals and fish compared with birds, algae and invertebrates, these differences are not as large as expected. Moreover, despite the significant direct use of fish to humans generally “fishes are not considered charismatic mega fauna” (Richardson and Loomis, 2009) and usually attract lower WTP when compared with marine mammals and birds (Loomis and White, 1996; Martín-López et al., 2008). The difference in the values of the taxa reported here for the Azores and those reported elsewhere in the literature may be due to the intimate and frequent contact between respondents and marine mammals near the coast, making whale watching in the Azores an expanding and more significant activity compared to bird watching. Furthermore, both scuba diving and fisheries (recreational and professional) are activities with high economic and cultural relevance at the regional level where the integrity and diversity of fish populations plays an important role. This may explain why being a fisherman (signified by the variable name “occupation fishermen”) was a positive and significant predictor of WTP for the individual marine taxa model. Thus there is evidence of context dependent valuation of individual components of marine biodiversity. The present study also offers novel insights regarding the economic preferences of visitors and residents. Loomis and White (1996) claimed that visitors allocated higher values to species preservation than resident households, since visitors are likely to have a large recreational component to their total value and are likely to be more knowledgeable about the species. A higher allocation of funds for visitors was also found by Loomis and Larson (1994), MartínLópez et al. (2008) and Richardson and Loomis (2009). In our study,

737

although the measures of the central tendency of WTP (mean and median) are higher for visitors, econometric evidence suggested that when comparing residents and visitors with similar socio-demographic profiles, residents are more likely to attach higher values to prevent species loss in the sea around Pico and Faial islands than visitors. In this case, the degree of attachment to the study site is possibly a driver of valuation rather than level of income or education. Economically consistent measures of WTP are expected to adjust with the scale of the change (Smith and Osborne, 1996). In other words, respondents are assumed to be able to distinguish between different quantities/qualities of the good and reflect this variability in their valuation function (Chilton and Hutchinson, 2003). The results suggest that the level of loss of species richness only had a significant effect on WTP when considering an individual marine taxon, and not when all marine species were considered. In some ways this is surprising, as the results of the pilot survey clearly demonstrated a desire by respondents to value all species in the marine environment as a holistic whole (cf as an ecosystem). This may indicate that asking respondents to consider changes in the number of all marine species, even at a regional scale, may be “beyond the margin of analysis” (Turner et al., 2003) imposing difficulties on accurate assessments of such changes on human welfare. Poor understanding about the welfare implications of such biodiversity loss among members of the public is excusable given that there is limited science about the ecological consequences of marginal or severe biodiversity loss. From a scientific perspective, several aspects regarding the relationship between marine biodiversity and ecosystem functions still need further clarification; e.g. (i) ecological processes that need to be included in the assessment of the ecosystem services supported or provided by marine biodiversity (ii) ecological indicators to measure marginal changes (iii) critical thresholds, etc.. Ecological uncertainty may contribute to ambiguity on the valuation figures enhancing the probability for responses insensitive to the scope of the change, such as charitable behaviour and fixed (self imposed) budget constraints, rather than responses that are economically rational in the traditional sense. Overall, the conclusion from this study is that visitors and residents of Pico and Faial islands attach positive and significant values to the local conservation of marine species. Although different payment vehicles were used and, therefore, direct comparisons between the results should be taken cautiously, it is worth noting that the willingness to pay to avoid a decrease in the number of all marine species is higher than the sum of the individual taxa. Bearing in mind that biodiversity around Pico and Faial islands is much more extensive than the five taxa under valuation; this result suggests that respondents were actually assessing what they were being asked and not simply providing symbolic responses. Furthermore, it also substantiates the observation that the public is able to perceive biodiversity conservation in wider terms than the single species/ group approaches, and that greater benefits are attached to the conservation of all ecosystem rather than partial conservations plans. Finally, this study also provides some insight into methodological considerations about the conceptual framework used to assess the valuation of marine biodiversity changes, namely the level of biological diversity and the scale of the change. Using multi-species or multi-groups of species as proxies of biological diversity may be more sensible than the alternative approach of valuing individual species. Despite public preferences for high profile species, multi species/ groups valuations improve public perceptions and awareness about the ecological and functional inter-relationships among organisms highlighting that the conservation of charismatic mega fauna implies the conservation of base-line taxa. In order words, the entire ecosystem needs to be protected. This finding is relevant and may help to correct inherent bias associated with species preservation. Another aspect that mainly conditioned the economic valuation of biodiversity is the scale of the change under scrutiny. Small and well defined changes provide more

738

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739

reliable values. Judging what is and what is not a marginal change is far from straightforward (Turner et al., 1998). However we assumed that the embedding effect would be minimized by considering the changes at a regional scale, when considering losses in the number of all marine species people fail to distinguish between two levels of loss — 10% and 25%. Whether the size of the change or the split sample experimental design conditioned these findings remains unclear. It should be noted that tests for scope sensitivity have traditionally relied on split sample designs (Carson and Mitchell, 1993; Loomis et al., 1993). More recent studies have tested scope sensitivity within sample design (see Chilton and Hutchinson, 2003) suggesting that testing this same questionnaire within the same sample could provide further insights about human motivations for marine species conservation. It could be argued that the complexity of biodiversity and problems like embedding or hypothetical bias undermine the use of stated preferences methods to assess the value of biodiversity changes. On the other hand, it could pragmatically be alleged that in a world of scarce resources and conflicting demands some information on public preferences for marine biodiversity conservation is better than no information, especially if society wishes to make sensible and politically inclusive choice (Christie et al., 2006). We believe that stated preferences methods, such as contingent valuation, provide useful information about individual consumer preferences that complement biological information/research contributing for the design of projects and policy measures that seek sustainable use of marine resources. Further research and interdisciplinary work is still necessary both to provide some contributions towards the definition of feasible approaches to capture the value of complex environmental goods as marine biodiversity and also to shed some light on the uncertainty underlying biodiversity loss and human welfare implications. This will certainly contribute to more reliable and valid valuation figures. Acknowledgments This work was supported by the Fundação para a Ciência e Tecnologia (FCT), through PhD grant SFRH/BD/31286/2006 (Adriana Ressurreição). The authors would like to thank the 3 anonymous reviewers for their helpful suggestions on an early version of the paper; and all MarBEF Theme 3 members and students, especially Melanie Austen, Stephen Mangi, Humberta Silva and Adelaide Costa, for their input and support. The authors also acknowledge the support by the MarBEF Network of Excellence ‘Marine Biodiversity and Ecosystem Functioning’ which is funded by the Sustainable Development, Global Change and Ecosystems RTD Programme of the EU's Sixth Framework Programme (contract no.GOCE-CT-2003-505446). References Adams, C., Seroa da Motta, R., Ortiz, R.A., Reid, J., Ebersbach Aznar, C., de Almeida Sinisgalli, P.A., 2008. The use of contingent valuation for evaluating protected areas in the developing world: economic valuation of Morro do Diabo State Park, Atlantic Rainforest, São Paulo State (Brazil). Ecological Economics 66, 359–370. Arrow, K., Solow, R., Portney, P., Leamer, E., Radner, R., Schuman, H., 1993. Report of the NOAA panel on contingent valuation. Federal Register 58 (10), 4602–4614 Washington DC.. Bandara, R., Tisdell, C., 2005. Changing abundance of elephants and willingness to pay for their conservation. Journal of Environmental Management 76, 47–59. Beaumont, N.J., Austen, M.C., Atkins, J.P., Burdon, D., Degraer, S., Dentinho, T.P., Derous, S., Holm, P., Horton, T., van Ierland, E., Marboe, A.H., Starkey, D.J., Townsend, M., Zarzycki, T., 2007. Identification, definition and quantification of goods and services provided by marine biodiversity: implications for the ecosystem approach. Marine Pollution Bulletin 54, 253–265. Berrens, R.P., Bohara, A., Kerkvliet, J., 1997. A randomized response approach to dichotomous choice contingent valuation. American Journal of Agricultural Economics 79, 252–266. Birol, E., Karousakis, K., Koundouri, P., 2006. Using economic valuation techniques to inform water resources management: a survey and critical appraisal of available techniques and an application. Science of The Total Environment 365, 105–122. Blaine, T.W., Lichtkoppler, F.R., Jones, K.R., Zondag, R.H., 2005. An assessment of household willingness to pay for curbside recycling: a comparison of payment card and referendum approaches. Journal of Environmental Management 76, 15–22.

Bord, R.J., O'Connor, R.E., 1997. The gender gap in environmental attitudes: the case of perceived vulnerability to risk. Social Science Quarterly 78, 830–840. Bosetti, V., Pearce, D., 2003. A study of environmental conflict: the economic value of grey seals in Southwest England. Biodiversity and Conservation 12, 2361–2392. Bowker, J.M., Stoll, J.R., 1988. Use of dichotomous choice nonmarket methods to value the whooping crane resource. Journal of Agricultural Economics 70, 327–381. Boyle, K.J., Desvousges, W.H., Johnson, F.R., Dunford, R.W., Hudson, S.P., 1994. An investigation of part–whole biases in contingent-valuation studies. Journal of Environmental Economics and Management 27, 64–83. Brent, R., 1998. Cost–Benefit Analysis for Developing Countries. Edward Elgar Publishing Limited, Cheltenham, UK. Brown, K., Taylor, L., 2000. Do as you say, say as you do: evidence on gender differences in actual and stated contributions to public goods. Economic Behaviour and Organization 43, 127–139. Brown, G., Layton, D., Lazo, J., 1994. Valuing Habitat and Endangered Species. Institute for Economic Research, University of Washington. Bulte, E.H., van Kooten, G.C., 1999. Marginal valuation of charismatic species: implications for conservation. Environmental and Resource Economics 14, 119–130. Cameron, T.A., Huppert, D.D., 1989. OLS versus ML estimation of non-market resource values with payment card interval data. Journal of Environmental Economics and Management 17, 230–246. Carson, R.T., Mitchell, R.C., 1993. The value of clean water: the public's willingness to pay for boatable, fishable and swimmable quality water. Water Resources Research 29, 2445–2454. Carson, R.T., Wilks, L., Imber, D., 1994. Valuing the preservation of Australia's Kakadu conservation zone. Oxford Economic Papers 46, 727–749. Carson, R.T., Flores, N.E., Meade, N.F., 2001. Contingent valuation: controversies and evidence. Environmental and Resource Economics 19, 173–210. Chaudhry, P., Singh, B., Tewari, V.P., 2007. Non-market economic valuation in developing countries: role of participant observation method in CVM analysis. Journal of Forest Economics 13, 259–275. Chilton, S.M., Hutchinson, W.G., 2003. A qualitative examination of how respondents in a contingent valuation study rationalise their WTP responses to an increase in the quantity of the environmental good. Journal of Economic Psychology 24, 65–75. Christie, M., Hanley, N., Warren, J., Murphy, K., Wright, R., Hyde, T., 2006. Valuing the diversity of biodiversity. Ecological Economics 58, 304–317. Clarke, T., 2006. Birds of the Atlantic Islands. Canary Islands, Madeira, Azores, & Cape Verde. Christopher Helm, London. Cummings, R.G., Brookshire, D.S., Schulze, W.D., 1986. Valuing Environmental Goods: An Assessment of the Contingent Valuation Method. Rowan & Allanheld, Totowa. Development, Core, Team, 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing3-900051-07-0. URL: http:// www.R-project.org. Vienna, Austria. Duan, N., 1983. Smearing estimate: a nonparametric retransformation method. Journal of the American Statistical Association 78, 605–610. Giraud, K., Turcin, B., Loomis, J., Cooper, J., 2002. Economic benefit of the protection program for the Steller sea lion. Marine Policy 26, 451–458. Hageman, R., 1985. Valuing Marine Mammal Populations: Benefits Valuations in a Multi-species Ecosystem. Administrative Report LJ-85-22. Southwest Fisheries Center — National Marine Fisheries Service, La Jolla, California. Hageman, R., 1986. Economic Valuation of Marine Wildlife: Existence Value Exist? Marine Pollution Environmental Damage Assessment Report EE-0121. National Center for Environmental Economics U.S. Environmental Protection Agency, Washington D.C. Hanemann, M., 1984. Welfare evaluations in contingent valuation experiments with discrete responses. American Journal of Agricultural Economics 66, 332–341. Hanemann, M., Loomis, J., Kanninen, B., 1991. Statistical efficiency of double-bound dichotomous choice contingent valuation. American Journal of Agricultural Economics 73, 1255–1263. Harrison, G.W., Kriström, B., 1995. On the interpretation of responses in contingent valuation surveys. In: Johansson, P., Kriström, B., Maler, K.G. (Eds.), Current Issues in Environmental Economics. Manchester University Press, UK. Jones, N., Sophoulis, C.M., Malesios, C., 2008. Economic valuation of coastal water quality and protest responses: a case study in Mitilini, Greece. Journal of SocioEconomics 37, 2478–2491. Jorgensen, B.S., Wilson, M.A., Heberlein, T.A., 2001. Fairness in the contingent valuation of environmental public goods: attitude toward paying for environmental improvements at two levels of scope. Ecological Economics 36, 133–148. Kahneman, D., Ritov, L., Jacowitz, K.E., Grant, P., 1993. Stated willingness to pay for public goods. Psychological Science 4, 310–315. Kotchen, M.J., Reiling, S.D., 1998. Estimation and questioning economic values for endangered species: an application and discussion. Endangered Species Update 15, 77–83. Kotchen, M.J., Reiling, S.D., 2000. Environmental attitudes, motivations, and contingent valuation of nonuse values: a case study involving endangered species. Ecological Economics 32, 93–107. Langford, I.H., Kontogianni, A., Skourtos, M.S., Georgiou, S., Bateman, I.J., 1998. Multivariate mixed models for open-ended contingent valuation data: willingness to pay for conservation of monk seals. Environmental and Resource Economics 12, 443–456. Loomis, J.B., Larson, D.M., 1994. Total economic value of increasing gray whale populations: results from a contingent valuation survey of visitors and households. Marine Resource Economics 9, 275–286. Loomis, J.B., White, D.S., 1996. Economic benefits of rare and endangered species: summary and meta-analysis. Ecological Economics 18, 197–206. Loomis, J., Lockwood, M., DeLacy, T., 1993. Some empirical evidence on embedding effects in contingent valuation of forest protection. Journal of Environmental Economics and Management 25, 45–55.

A. Ressurreição et al. / Ecological Economics 70 (2011) 729–739 Martín-López, B., Montes, C., Benayas, J., 2008. Economic valuation of biodiversity conservation: the meaning of numbers. Conservation Biology 22, 624–635. Menezes, G., 2003. Demersal Fish Assemblages in the Atlantic Archipelagos of the Azores, Madeira and Cape Verde, Department of Oceanography and Fisheries. University of the Azores. Metrick, A., Weitzman, M.L., 1996. Patterns of behavior in endangered species preservation. Land Economics 72, 1–16. Mitchell, R.C., Carson, R.T., 1989. Using Surveys to Value Public Goods: The Contingent Valuation Method. Resource for the Future Washington, DC. Neto, A.I., Tittley, I., Raposeiro, P.M., 2006. Rocky Shore Marine Flora of the Azores. Regional Council of the Environment and the Sea. Nijkamp, P., Vindigni, G., Nunes, P.A.L.D., 2008. Economic valuation of biodiversity: a comparative study. Ecological Economics 67, 217–231. Nunes, P.A.L.D., van den Bergh, J.C.J.M., 2001. Economic valuation of biodiversity: sense or nonsense? Ecological Economics 39, 203–222. Olsen, D., Richards, J., Scott, D., 1991. Existence and sport values for doubling the size of Columbia river basin salmon and steelhead runs. Rivers 2, 44–56. Paradiso, M., Trisorio, A., 2001. The effect of knowledge on the disparity between hypothetical and real willingness to pay. Applied Economics 33, 1359–1364. Richardson, L., Loomis, J., 2009. The total economic value of threatened, endangered and rare species: an updated meta-analysis. Ecological Economics 68, 1535–1548. Samnaliev, M., Stevens, T.H., More, T., 2006. A comparison of alternative certainty calibration techniques in contingent valuation. Ecological Economics 57, 507–519. Samples, K., Hollyer, J., 1990. Contingent valuation of wildlife resources in the presence of substitutes and complements. In: Johnson, R., Johnson, G. (Eds.), Economic Valuation of Natural Resources: Issues, Theory and Application. Westview Press, Boulder, Colorado, pp. 177–192. Samples, K., Dixon, J., Gowenand, M., 1986. Information disclosure and endangered species valuation. Land Economics 62, 306–312. Santos, R.S., Hawkins, S., Monteiro, L.R., Alves, M., Isidro, E.J., 1995. Marine research, resources and conservation in the Azores. Aquatic Conservation: Marine and Freshwater Ecosystems 5, 311–354. Smith, V.K., Osborne, L.L., 1996. Do contingent valuation estimates pass a “scope” test? A meta-analysis. Journal of Environmental Economics and Management 31, 287–301. Spash, C.L., Urama, K., Burton, R., Kenyon, W., Shannon, P., Hill, G., 2009. Motives behind willingness to pay for improving biodiversity in a water ecosystem: economics, ethics and social psychology. Ecological Economics 68, 955–964.

739

Stevens, T., Echeverría, J., Glass, R., Hager, T., More, T., 1991. Measuring the existence value of wildlife: what do CVM estimates really show. Land Economics 67, 390–400. Tkac, J., 1998. The effects of information on willingness-to-pay values of endangered species. American Journal of Agricultural Economics 80, 1214–1220. Turner, R.K., Adger, W.N., Brouwer, R., 1998. Ecosystem services value, research needs, and policy relevance: a commentary. Ecological Economics 25, 61–65. Turner, R.K., Paavola, J., Cooper, P., Farber, S., Jessamy, V., Georgiou, S., 2003. Valuing nature: lessons learned and future research directions. Ecological Economics 46, 493–510. Turpie, J.K., 2003. The existence value of biodiversity in South Africa: how interest, experience, knowledge, income and perceived level of threat influence local willingness to pay. Ecological Economics 46, 199–216. Veisten, K., Fredrik Hoen, H., Navrud, S., Strand, J., 2004. Scope insensitivity in contingent valuation of complex environmental amenities. Journal of Environmental Management 73, 317–331. Venkatachalam, L., 2004. The contingent valuation method: a review. Environmental Impact Assessment Review 24, 89–124. White, P.C.L., Gregory, K.W., Lindley, P.J., Richards, G., 1997. Economic values of threatened mammals in Britain: a case study of the otter Lutra lutra and the water vole Arvicola terrestris. Biological Conservation 82, 345–354. White, P.C.L., Bennett, A.C., Hayes, E.J.V., 2001. The use of willingness-to-pay approaches in mammal conservation. Mammal Review 31, 151–167. Whitehead, J.C., 1992. Ex ante willingness to pay with supply and demand uncertainty: implications for valuing a sea turtle protection programme. Applied Economics 24, 981–988. Whitehead, J.C., Blomquist, G.C., Hoban, T.J., Clifford, W.B., 1995. Assessing the validity and reliability of contingent values: a comparison of on-site users, off-site users, and non-users. Journal of Environmental Economics and Management 29, 238–251. Wilson, C., Tisdell, C., 2003. Conservation and economic benefits of wildlife-based marine tourism: sea turtles and whales as case studies. Human Dimensions of Wildlife 8, 49–58. Zhongmin, X., Guodong, C., Zhiqiang, Z., Zhiyong, S., Loomis, J., 2003. Applying contingent valuation in China to measure the total economic value of restoring ecosystem services in Ejina region. Ecological Economics 44, 345–358.