Economic values of species management options in human–wildlife conflicts: Hen Harriers in Scotland

Economic values of species management options in human–wildlife conflicts: Hen Harriers in Scotland

Please download to get full document.

View again

of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 4 | Pages: 7

Extension: PDF | Download: 0

  Analysis Economic values of species management options in human – wildlife con fl icts:Hen Harriers in Scotland Nick Hanley a, ⁎ , Mikolaj Czajkowski b , Rose Hanley-Nickolls c , Steve Redpath d a Economics Division, University of Stirling, UK  b Warsaw Ecological Economics Center, University of Warsaw, Poland c Scottish Agricultural College, Edinburgh, UK  d School of Biological Sciences, University of Aberdeen, UK  a b s t r a c ta r t i c l e i n f o  Article history: Received 26 March 2010Received in revised form 13 August 2010Accepted 14 August 2010Available online 15 September 2010 Keywords: Human wildlife con fl ictsChoice experimentsHeather moorlandsWildlife managementHen HarriersRaptors In this paper, we use the choice experiment method to investigate public preferences over alternativemanagement regimes for a top-level predator in UK moorlands, the Hen Harrier. These birds are at the centreof a con fl ict between moorland managers and conservation organisations. Illegal killing of Hen Harriers onmoorland managed for Red Grouse is considered to be one of the main factors limiting harrier populationgrowth in the UK. Incentives for persecution arise due to the impacts of Hen Harriers on populations of RedGrouse, which are managed for commercial shooting. Numerous alternatives have been proposed to managethis system. We considered three which have emerged from stakeholder debates and scienti fi c enquiry:tougher law enforcement, moving  “ excess ”  birds from grouse moors, and feeding of harriers. Results showedthat respondents, sampled from the Scottish general public, were willing to pay both for avoiding reductionsin harrier populations and for increases, but that these values were lower than those associated withequivalent changes for another raptor sharing the same moorland habitat, the Golden Eagle. Respondentsvalued a move away from current management, but were largely indifferent to which management optionwas taken up, suggesting that management options should be selected in terms of relative costs, and on whobears these costs. Differences within our sample of respondents in preferences across management optionsemerge when a latent class model is estimated.© 2010 Elsevier B.V. All rights reserved. 1. Introduction Single objective land management has historically resulted in thesuppression of many plant and animal species. This is most obviouslysointhecaseofagriculture,wherethedesiretomaximisepro fi tsfromcultivationofparticularcropshasresultedintheuseofherbicidesandpesticides to reduce competition for resources and to protect yields. If thesechangesimpactonthegoalsofotherstakeholders,thencon fl ictsover resources often ensue. For example, changing patterns of agricultural land use in parts of Africa have led to increased con fl ictbetween farmers and conservationists over crop raiding elephants(Thirgood et al., 2005). Understanding such con fl icts requires anappreciation of the economic costs (e.g. crop losses to farmers) andbene fi ts (e.g. non-market bene fi ts of elephant conservation to peopleliving outside elephant habitat) of land management, both in terms of the magnitude of these bene fi ts and costs, and their distributionacross stakeholders.InpartsofEurope,afocusonintensivegamemanagementforsporthasledtochangesintheabundanceofmanyspecies.Thisisillustratedby the historical impact of managing land for hunting on predators inthe UK, through poisoning, shooting or trapping (Lovegrove, 2007). The management of Red Grouse ( Lagopus lagopus scotticus ) in the UKuplands provides a case in point. Management of moorlands for RedGrouse shooting since the mid 19th Century has led to declines inmanyspeciesofpredators,andtheextirpationof  fi vespeciesofraptor(Newton, 1998). The aim of grouse management is to maximise numbers of birds available for shooting in the autumn. Thismanagement involves a mixture of vegetation management (e.g.heather burning) and predator control (Hudson and Newborn, 1995).There is some evidence to suggest that grouse shooting provideseconomic bene fi ts to rural communities (McGilvary, 1995; PACEC,2006), as well as conservation bene fi ts because it retains heathermoorland (Robertson et al., 2001) and associated wading birds wherewell-practised (Thompson et al., 1995; Tharme et al., 2001). One particular con fl ict which has arisen in this context concernsthe management of Hen Harriers ( Circus cyaneus ) on sporting estates.Hen Harriersare a medium-sized birdof preywhichbreed on heathermoorlands in the uplands. They are Red Listed due to populationdeclines in the last 200 years (Baillie et al., 2009). Most recent data Ecological Economics 70 (2010) 107 – 113 ⁎  Corresponding author. Economics Division, University of Stirling, Stirling FK9 4LA,Scotland, UK. Tel.: +44 1786 466410; fax: +44 1786 467469. E-mail address: (N. Hanley).0921-8009/$  –  see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.ecolecon.2010.08.009 Contents lists available at ScienceDirect Ecological Economics  journal homepage:  reveals spatial variation in population trends, with a 41% increase inthe UK and Isle of Man during 1998 – 2004, but with decreases onmoorland managed for grouse shooting in the Southern Uplands, eastHighlands and northern England (Sim et al., 2007). There are roughly633 pairs in Scotland at present. Hen Harriers have been protected bylaw since 1954, but illegal killing has occurred due to the economiccosts of Hen Harriers to grouse moor managers.The economic costs to grouse moor owners arise because harriersprey on grouse (Thirgood et al., 2000). Arguments between the conservation lobby and the sporting estate community have becomepolarised over time (Redpath et al., 2004; Thirgood and Redpath,2008). Evidence shows that (i) Hen Harrier densities can increase tothe extent that they make management for grouse shootingeconomically unviable;(ii) illegalkillinghas resultedin asuppressionof harrier populations in both England and Scotland (Etheridge et al.,1997; Sim et al., 2007); and (iii) that enforcement of current lawsprohibiting lethalcontrolhasbeenineffective(ThirgoodandRedpath,2008; Redpath et al., 2010). It is clear that sustainable managementoptions should be considered more broadly (Thompson et al., 2009). In this paper, we employ the choice experiment method toinvestigate public preferences for alternative management strategiesfor Hen Harriers on grouse moors. We view public preferences asimportant to the success of con fl ict management, since solutions areonly likely to be effective in the long term if they command majoritysupport. Non-hunting members of the public may well feel that theyalso have a stake in the future management of heather moorlandsmanaged for grouse, since a legal right of recreational access to allsuchareasexistsinScotland,andsincenon-usevaluesforwildlifeandlandscape associated with heather moorlands are also likely to beimportant. As noted by Redpath et al. (2010), a number of alternative management options have been proposed to help manage thisstakeholder con fl ict. Three of these alternatives are dealt with here:the establishment of feeding stations, providing harriers withalternative food sources to grouse; the establishment of quotas forHen Harrier densities on sporting estates maintained by physicallymoving eggs or chicks away from grouse moors to alternativelocations with lower or zero harrier populations; and increasing theprobability of detection of illegal persecution, by increasing policesurveillance on grouse moors. These are the alternatives most underdiscussion by stakeholders and scientists at the present time.Estimating people's willingness to pay for environmental goodshas been in focus of mainstream ecological economics for over30 years now, as in some cases it remains the only way of providingeconomic value of many environmental goods. Estimating thesevalues is necessary for conducting cost-bene fi t analyses, which mayprovide economic arguments for conservation. In particular, therehave been numerous valuation studies of improvements in thecondition of endangered species. A recent meta-analysis of 43 suchstudies is provided by Richardson and Loomis (2009a). The authorsreview67willingness-to-payestimatesforthreatened,endangeredorrare species and identify major drivers of these values. The reportedWTP for avoiding loses or ensuring survival of bird species are withinthe range or 11.38 to 130.19 $US with a mean close to 40 $US(Richardson and Loomis, 2009b).Similarly, Martín-López et al. (2008) and Jacobsen and Hanley(2009) review a large number of studies on economic valuation of biodiversity and perform metaanalyses of these studies to determinewhat factors affect willingness to pay. Among other things, they  fi ndthat anthropomorphic and anthropocentric characteristics, resultingfrom the public's attitude toward species, remain highly signi fi canteven when respondents are suf  fi ciently informed about  ‘ scienti fi cfactors ’  that need to be taken into consideration.A similar consideration for valuing animal species is provided by Jacobsen et al. (2008). The authors aim at valuing environmentalimprovements (preserving natural habitats) using number of speciesas an indicator of biodiversity. In one of the two treatments of theirexperiment they used quantitative description of changes in thenumber of species, while in the other they named a few species andobserved substantially higher WTP values. Since economic values areby de fi nition anthropogenic it is understandable that so called ‘ fl agship species ’  or  ‘ charismatic species ’  (usually top predators orspecies associated with national identity or culture) will receive moreattention and higher WTP values (May, 1995; Noss, 1990). The aims of this work were to estimate public preferences andwillingness to pay for the three management alternatives of Scottishmoorlands outlined above. To accomplish this, a Choice Experimentwasimplemented usingmembersofthe Scottishgeneralpublicasthesampling frame, allowing us to investigate preference heterogeneityof different groups of the public  —  in particular hunters and peopleliving in rural vs. urban communities. In our choice experiment, inaddition to the description of management alternatives we haveincluded descriptions of possible population changes of Hen Harriers.Toavoidafocussingbias(Kahneman,2010)onHenHarriers,asecondraptorpopulation — GoldenEagles(  Aquila chrysaetos ) — wasincludedin the experimental design. Golden Eagles are often found in HenHarrier habitat, and are also top predators which have been subject toillegal persecution, particularly in managed grouse moors (Watsonet al., 1989; Whit fi eld et al., 2007; RSPB, 2006). However, note thatfeeding stations established for Hen Harriers are not likely to bene fi tGolden Eagles due to differences in feeding behaviour.In what follows, Section 2 describes the questionnaire design andsamplingmethods; Section3 providessomeresults,whilstadiscussionand conclusion follow in Section 4. 2. Questionnaire Design and Sampling Methods In order to elicit WTP values we constructed a postal surveyquestionnaire containing choice experiments. The survey instrumentcan be obtained from the authors on request. The survey begins byintroducing respondents to the location of   “ the uplands ”  in the UK,and asking whether they have visited this area and for what reasons.The questionnaire then describes how some uplands areas aremanaged as grouse moors: “ About half of the heather moorlands found in the uplands aremanaged for grouse shooting. As well as using heather burning tokeep the heather in good condition, many landowners employgame keepers to look after the grouse, for example by protectingthem from predators. Red grouse shooting provides manyeconomic, social and ecological bene fi ts to an area. If there areenough grouse, shooting is the primary source of income in manyupland estates. In Scotland, grouse shooting contributes £5million to the rural economy a year and supports about 1,240 jobs.  ” Respondents were also told about the contribution of grousemanagement to maintaining heather moorlands.  “ In areas wherethere are no longer enough grouse to maintain a shoot, land ownersare using moorland for forestry or sheep grazing. This is thought tohavecontributedinparttothedeclineinheathermoorlandintheU.K.Since1950over30%of grousemoorlandhasbecomeunmanaged,andbetween 1940 and 1970, 20% of heather moorland was lost. ” The text then introduces the Hen Harrier, describing theirconservation status and threats from illegal persecution. Respondentsare asked whether they think any action should be taken with regardto the current situation. Next, the survey introduces Golden Eagles,and again describes their conservation status and current threats tothe species. The three alternatives for moorland management aimedat Hen Harriers are then described: setting up feeding stations,moving eggs and chicks to ensure local populations stay withinprescribed quotas, and stricter law enforcement. Respondents weretold that each of these alternatives would impose costs on society, for 108  N. Hanley et al. / Ecological Economics 70 (2010) 107  – 113  example in terms of extra policing, or labour costs for movement of birds, and that these costs would need to be paid for out of increasedtaxes.The choice sets were then presented to individuals. The choiceexperiment design consisted of four attributes. These were: –  Changes in the population of Hen Harriers on heather moorlandsin Scotland. The levels here were a 20% decline (used as the statusquo), maintaining current populations, and a 20% increase in thecurrent population. –  Changes in the population of Golden Eagles on heather moorlandsin Scotland. The levels here were a 20% decline (used as the statusquo), maintaining current populations, and a 20% increase in thecurrent population. –  Management options. These included the current situation,moving Hen Harriers ( “ MOVE ” ), diversionary feeding ( “ FEED ” )andtougherlawenforcement( “ LAW ” ).Theselevelswereincludedas labelled choices. That is, in each choice card, 4 options wereavailable. One represented the status quo, and then 3 choicecolumns showed variations in other attribute levels given aparticular, labelled management strategy. –  Cost of the policy. We told respondents that  “ the cost levelindicated is the amount of extra tax which a household like yoursmighthavetopayifthegovernmentwentaheadwiththatoption. ” The levels used were £0 (the status quo), £10, £20, £25, and £50.These costs are not linearly associated with any managementoption,sinceactualcostsareunknown,andsincethiswouldcreatedif  fi culties for estimating the choice model. Cost levels werechosen based on the results of a pilot survey.Fig. 1 gives an example of a choice card. Respondents were askedto carefully consider their budgets and current expenditures inmaking their choices, and that they should not worry if they did notfeel that they had expert knowledge on the issues, but that theiropinionwas importantto government policy making.Sixchoicecardswere given to each respondent. Those respondents who chose thestatus quo, zero cost option in each choice card were asked why thiswas, in order to separate out protest bidders from people who did notvalue Hen Harrier or Golden Eagle conservation in moorlands. Havingcompleted their choices, respondents were asked to read backcarefully through these to make sure they were happy with howthey had completed these tasks. Finally, a series of socio-economicand behavioural questions were asked, for example includinghousehold income, and whether the respondent was a hunter orhad ever been hunting.Thechoice experimentwas designed to minimizethe determinantof the AVC matrix of the parameters ( D-error  ) given the priors on theparameters of a representative respondent's utility function (Scarpaand Rose, 2008). We have accounted for uncertainty with respect toparameters' priors by allowing these priors to be random variablesfollowing a probability distribution ( Bayesian ef   fi cient design ; Sándorand Wedel, 2001). The parameters of this distribution were derivedfrom a preliminary model estimated on data available from a pilotstudy. Pilot surveys were undertaken using in-person surveys of arandom sample of Edinburgh households. 1 The  fi nal design consistedof 8 questionnaire versions, each with 6 choice cards per respondent.This  fi nal version of the questionnaire was mailed to a randomsample of 1000 addresses in Scotland during 2009. Addresses wereextracted from the database, which is based on the electoralregister and telephone records. Rural areas were deliberately over-sampled, and the sample was strati fi ed by region to ensure adequateregional coverage. Two weeks after mailing the  fi rst copy of thequestionnaire to a respondent, a reminder letter was sent to thosewho had not responded. Two weeks after this, a second reminder andcopy of the questionnaire was sent out to those who still had notresponded.We have received 223 surveys which were usable for futureanalysis, i.e. the respondents answeredthe mostvital questions of thesurvey (in particular marked their choices in the choice experimentsection). Comparing the representativeness of our sample with thecharacteristics of the general adult population of Scotland we mustnote that our sample suffered from some over-representation of respondents aged 41 – 65 and under-representation of younger adults(aged 15 – 25 and 26 – 40), and over-represented higher incomegroups. 2 The ratio of male and female respondents was very close tothe national average. In addition, 24% (54) of the respondents in oursampleturnedouttobehuntersorhaveeverbeenhunting,70%(149)were currently living in the countryside, while 51% (109) stated thatthey grew up in the rural communities. Those living in rural areaswere therefore more likely to return their questionnaires than thoseliving in urban areas.Protesting respondents were identi fi ed as those who at the sametime: (1) stated that hen harriers should be protected, (2) picked the ‘ do nothing ’  option in every choice set and (3) as a reason for picking ‘ do nothing ’  option stated that either they did not believe otheroptions would work, that the money would be wasted and not usedappropriatelyorthatothersshouldpayforconservation.Intotaltherewereonly3respondents(1%)whowere thusclassi fi edas  ‘ protesters ’ .To analyse the choice data, the Hen Harrier and Golden Eagleattributes were dummy coded as levels, to allow for non-constantmarginal utility of the improvements. Price was entered in poundssterling. The management alternatives FEED, LAW and MOVE enter aslabelled Alternative Speci fi c Constants. We ran a variety of RandomParameter Logit (RPL) 3 models on the data (Hensher and Greene,2003; Hensher et al., 2005). For the RPL, we ran models with various distributionsof parameters(normal,lognormal,triangular), includingheterogeneity in means and variances of these distributions withrespect to individual-speci fi c socio-economic data. In addition wetested models with error components included (with and withoutheteroscedascity in random error components). Finally, we estimatedmodels both with correlated and un-correlated parameters. Sinceeach respondent faced six choice-sets, a panel data speci fi cation of errors was used throughout. 3. Results In the best performing model all the attributes were speci fi ed asrandom parameters of a normal distribution. We allowed forcorrelations between random parameters (which proved to be highlysigni fi cant) and accounted for panel structure of our dataset (eachrespondent faced 6 choice-sets) by introducing a random effects typeof treatment  —  an additional random term for all observations fromthe same individual. In addition, we have introduced individualheterogeneity in all random parameters ’  distributions means usingindividual-speci fi c explanatory variables. This allowed us to observehowpreferencessystematicallydifferbetweengroupsofrespondents.In our case, the two most interesting variables  a priori  were (1) beinga hunter vs. not being a hunter, and (2) living in rural vs. urbancommunity. Our resultsshowed,however,that living(orgrowingup)in rural vs. urban areas did not allow to account for any unobservedheterogeneity. This led to the  fi rst conclusion that there were nosystematic differences in preferences between respondents living inrural vs. urban communities. Therefore, in the  fi nal model we have 1 The design for the pilot study was also generated for D-ef  fi ciency, using expert judgment priors. 2 17.5% of the respondents refused to provide their income what might furthercontribute to the high income bias. 3 The Random Parameters Logit model allows for a very wide range of modelspeci fi cations in terms of underlying choice processes and preference structures(McFadden and Train, 2000).109 N. Hanley et al. / Ecological Economics 70 (2010) 107  – 113  included only being a hunter as an explanatory variable of randomparameters means.Table 1 shows the results for our Random Parameters Logit model.Takingtheresultsin columnA,wecanseethatallof theattributesaresigni fi cant determinants of choice: people prefer either no decline oran increase in Hen Harriers relative to a decline, and the same holdsfor Golden Eagles. For the management alternatives, LAW, FEED andMOVE, their signs suggest that they are all preferred to the currentsituation. Price is a negative and signi fi cant determinant of choice.Column B provides strong evidence of unobserved preferenceheterogeneity in terms of signi fi cant standard deviation coef  fi cientsfor all attributes. This is especially apparent in the case of alternativespeci fi c constants  —  standard deviations of their distributions suggestthat at least for some respondents this coef  fi cient might be close to 0or negative (and hence they might prefer the status quo).Finally, in column C of the table we present the estimates of covariates of means or random parameters  —  i.e. how being a hunterin fl uencesthemeanofeachrandomparameter.Thecoef  fi cientssuggestthat hunters have signi fi cantly lower willingness-to-pay for anyimprovement in Hen Harrier populations. Similar pattern emerges forimprovements for golden eagles, however this result is not statisticallysigni fi cant. Finally, the means of ASCs for implementing each of themanagement options are substantially higher for hunters. 4 Table 2 shows implicit prices (willingness to pay values) for thewhole sample, and also implicit prices derived for hunters and non-hunters, along with associated standard deviations. 5 These weregenerated using parametric bootstrapping following Krinsky andRobb (1986). Since our price parameter was also random we havefollowed the simulation method proposed by Hu et al. (2005): inorder to avoid  “ exploding ”  implicit prices when random priceparameter was very close to zero we averaged over 10 4 draws of each parameter, for each round of Krinsky and Robb draws fromparameter distributions. Our results show that for our sample themean willingness to pay for maintaining current populations of theHen Harriers and Golden Eagles are around £36 and £52 perhousehold per year respectively. A 20% increase in the population of each species would be worth £44 and £61 respectively. These resultsare in line with other valuation studies focusing on birds of preyspecies (Richardson and Loomis, 2009a,b). WTPsfor improvements in the population of Golden Eagles exceeds WTPs for improvements inpopulation of Hen Harriers. This  fi nding con fi rms earlier results thatrespondents are generally willing to pay more for species with highercharisma  —  in this case larger and more culturally-dominant eaglesthan smaller harriers (e.g. Martín-López et al., 2008; Jacobsen andHanley, 2009). Comparing willingness to pay for stabilising thepopulation relative to a decline with willingness to pay for a 20%population increase suggests decreasing marginal utility from theseimprovements over the status quo.WTPs for all management options relative to the status quo is inthe order of £100/household/year, a surprisingly large amount. TheWTPsformanagementoptionsLAW,FEEDandMOVEareverysimilar,and not signi fi cantly different from each other at the 95% level. Thegeneral message seems to be that people are willing to pay for achange in the current management situation, but are ratherindifferent as to  which  policy option is implemented or  how  increasesin populations of the birds are achieved.Finally, we turn to analyzing systematic differences in preferencesof hunters vs. non-hunters. Our results indicate that hunters arewilling to pay substantially less for the proposed population increasesof Hen Harriers and Golden Eagles than non-hunters. In the case of Hen Harriers the hunters' WTPs are around £11 and £16, while fornon-hunters the implicit prices are £44 and £53 respectively. This isunderstandable, as increased populations of birds of prey might beassociatedwith lower populations of red grouse, and hence lower baglimits.ThesameholdsforGoldenEagles — £28and£49forhuntersvs.£49 and £65 for non-hunters. Despite this, hunters are willing to paysubstantially more for implementing one of the proposed manage-ment programs  —  their willingness to pay for feeding and relocatingprogram is roughly £50 – £57 higher, and for tougher law enforcementit is almost £75 higher than non-hunters' WTP. This may re fl ect awillingness to suppress illegal activity which brings a bad publicreputation to law-abiding estate owners.Overall WTPs for all management programs (controlling for HenHarriersandGoldenEaglespopulationchanges)aresurprisingly high.We  fi nd this result interesting and provide the following interpreta-tion. The respondents expected the proposed management options tosolve management-conservation con fl ict in managed moorlands inScotland. Reconciling the interests of hunters, conservationists andmanagers seems to be worth a lot, even if it does not causepopulations of the birds of prey to increase. In other words, ourrespondents might have associated implementing the programs withother improvements (not accounted for by our attributes). Forinstance they could have considered feeding program to eliminatethe pressure on red grouse populations, and be willing to pay for iteven if it would not increase Hen Harrier and Golden Eaglepopulations. If this was the case, it is not surprising that the hunterswere willing to pay even more for such a improvement, as it couldprovide economic bene fi ts not only to adjacent communities, but alsoto hunters directly, as there would be more game for shooting. 4 Should the reader be interested in inspecting correlation between the parameters,we report the estimates of the elements of lower triangular of Cholesky matrix (i.e.products of Cholesky decomposition of the variance-covariance matrix of coef  fi cients)in Annex 1. 5 Implicit prices were calculated for a model without the covariate in the mean of the distribution of the random price coef  fi cient which was found insigni fi cant (seeTable 1). DO NOTHING Maintain current management.   LAW Stricter law enforcement. FEED Feeding stations away from grouse. MOVE Move eggs and chicks to new sites. HEN HARRIER 20% population decline. Maintain current population. Maintain current population. Maintain current population. GOLDEN EAGLE 20% population decline. 20% populationincrease. Maintain current population. 20% population decline. COST £0£50 £50£10YOUR CHOICE (please tick one only) Fig. 1.  Example choice card.110  N. Hanley et al. / Ecological Economics 70 (2010) 107  – 113  4. Discussion and Conclusions In this paper, we have used choice modelling to investigate publicpreferences for the management of the Hen Harrier. Public prefer-ences matter if governments wish to partly base policy decisions overspecies protection and landmanagement onanassessment of relativecosts andbene fi ts(Hanleyand Barbier,2009).Here, we showthatthepublic is willing to pay to prevent further declines in harrierpopulations in Scotland, and to increase these populations. To avoida potential bias from asking respondents to focus on a singleenvironmental  “ good ” , the experimental design used here allowedto trade-offs between the protection of Hen Harriers and anothercharismatic raptor, the Golden Eagle. This showed that Golden Eagleconservation was valued more highly than Hen Harrier conservation.Due to sample selection effects, we are unable to aggregate oursample willingness to pay measures to the general population: wecannot also draw inferences about the preferences of those who self-selected into not responding to the survey.For the sample as a whole, the Random Parameters Logit modelsshowed that respondents were willing to pay to achieve more raptorconservation on moorlands, but were largely indifferent to how agiven population was achieved. That is, there were no statisticallysigni fi cant differences in the utility attached to either of the threelabelledmanagementalternatives — stricterlawenforcement,feedingstations and moving eggs and chicks (this implies that replacing thethree labelled Alternative Speci fi c Constants with one ASC for thestatus quo choice would yield qualitatively similar results). Noevidence emerged that any of the options was viewed as beingdifferent on ethical or conservation effectiveness grounds byrespondents. This indifference to  how  conservation objectives areachieved implies that bene fi ts are roughly equal across managementalternatives if the same outcome is achieved. In turn, this implies thateconomic ef  fi ciency would require the management action with thelowest costs to be chosen.Our results, in terms of WTP for increasing population levels are inthe range of £35 – £60 per year per household, depending on whetherHenHarrierorGoldenEaglepopulationwillincreaseandwhethertheincrease will allow to maintain current population or increase thepopulation a further 20%. These results seem to be similar to thoseobservedintheotherstudies(see RichardsonandLoomis,2009a,bfora comprehensive review). In addition, we observe that WTP for alarger, more charismatic species (Golden Eagle) is higher than for aless charismatic species. This is in line with earlier  fi ndings of otherstudies (Martín-López et al., 2008; Jacobsen and Hanley, 2009; Jacobsen et al., 2008; May, 1995; Noss, 1990). Furthermore, we observe a surprisingly high WTPs for implementing any of themanagement programs, irrespectively of the changes in populationsof birds of prey they incorporate.We ran a number of RPL models to allow for preferenceheterogeneity, and included socio-demographic explanatory variablesfor random parameter distributions means. Our results show thatalthough there was large preference heterogeneity, many individualcharacteristics (such as whether people lived in a rural or urban area;whether they were born in an urban area; or visit the uplands forrecreation) did not account for signi fi cant systematic differences inpreferences.However,beingahunterturnedouttosigni fi cantlyexplainrespondents'preferences.Overall, hunterswerewillingtopaysubstan-tiallylessforincreasesinthepopulationsofbirdsofprey,however,theywere willing to pay substantially more for implementing managementprograms independently of the impact on raptor populations.No data exists on the relative costs of the management optionspresented here to allow for a comparison, with the exception of onestudy on diversionary feeding (Redpath et al., 2001). An important distinction from a policy perspective is to think about on whom thesecosts would fall. For each management option, costs might be dividedinto private costs to estate owners and workers, and costs to thepublic purse. Costs to estate owners and workers would be highestwhen populations of Hen Harrier reach levels on a particular moor  Table 1 The results of the RPL models.Means of normally distributedparametersStandarddeviationsofnormallydistributed parametersCovariate of means of randomparameters (being a hunter)Coef  fi cient S.E. Coef  fi cient S.E. Coef  fi cient S.E.HH1  –  maintaining current populations 3.1112*** 0.5005 2.3347*** 0.4811  − 1.9225** 0.9016HH2  –  20% increase in populations 3.6114*** 0.4248 1.9138*** 0.2817  − 2.2369*** 0.6007GE1  –  maintaining current populations 4.5658*** 0.6574 3.6520*** 0.5027  − 1.3374 0.9761GE2  –  20% increase in populations 4.8116*** 0.6019 2.9145*** 0.4104  − 0.5768 0.7979LAW  –  alternative speci fi c constant 4.2940*** 1.1878 10.2201*** 0.6687 4.7938*** 1.1683FEED  –  alternative speci fi c constant 4.0083*** 1.1951 11.2235*** 0.5751 4.7247*** 1.1948MOVE  –  alternative speci fi c constant 3.5152*** 1.2053 11.4516*** 0.5570 4.5486*** 1.1676COST  –  monetary attribute  − 0.0860*** 0.0146 0.1811*** 0.0082 0.0370 0.0254Log likelihood function  − 1012.6131Pseudo  R 2 0.4441AIC 1.6204BIC 1.8254***, **, * Signi fi cance at 1%, 5%, 10% level.  Table 2 The implicit prices based on the RPL models [£/household/year].All sample Hunters OthersImplicit price Standard error Implicit price Standard error Implicit price Standard errorHH1  –  maintaining current populations 35.98*** 5.1355 10.70 9.2351 44.05*** 6.4356HH2  –  20% increase in populations 44.40*** 6.2125 16.47** 6.8377 53.31*** 7.9123GE1  –  maintaining current populations 51.80*** 6.5345 27.96*** 9.4326 59.42*** 8.1766GE2  –  20% increase in populations 61.24*** 7.7825 48.54*** 9.7525 65.30*** 8.9929LAW  –  alternative speci fi c constant 103.40*** 29.9435 159.96*** 42.1290 85.34*** 27.9055FEED  –  alternative speci fi c constant 101.22*** 29.9852 144.25*** 39.1485 87.48*** 28.5799MOVE  –  alternative speci fi c constant 97.98*** 29.3582 136.10*** 38.2710 85.81*** 27.9631***, **, * Signi fi cance at 1%, 5%, 10% level.111 N. Hanley et al. / Ecological Economics 70 (2010) 107  – 113
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks