The role of fairness norms the household-based natural forest conservation The case of Wolong, China

EcologicalEconomicsxxx(2010)xxx–xxx

ContentslistsavailableatScienceDirect

EcologicalEconomics

journalhomepage:www.elsevier.com/locate/ecolecon

Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China

ZhiyuanSong1,ZhiyunOuyang⁎,WeihuaXu

StateKeyLabofUrbanandRegionalEcology,ResearchCenterforEco-EnvironmentalSciences,ChineseAcademyofSciences,Beijing100085,China

articleinfoabstract

Theshadowof“thetragedyofthecommons”withitspopularizedassumptionofselfishindividualbehaviorconcernspolicy-makers,andconsequentlythecooperativecapabilityoflocalcommunitiesincommonresourcemanagementisusuallyunderrated.Nevertheless,hereweproposeahypothesis,basedonagametheorymodel,thatthesocialnormoffairnessratherthanself-interestmightmotivatecooperationinnaturalresourceconservationbydiscountingtheutilitygainfromillegalloggingatthecostofanotherhousehold'ssubsidyloss.InWolongNatureReserveofChina,arecenthousehold-basednaturalforestconservationprogramhasledtoremarkableprogressinprotectingthehabitatofgiantpandas(Ailuropodamelanoleuca)withlowmonitoringpressure.Weundertookexperimentsbasedonaone-shotUltimatumGametotesttheprevalenceofthenormoffairnessinthelocalcommunities.Mostproposersintheexperimentsshowedstrongpreferencesforfairoffersinspiteoftheresponders'reluctancetorejectlowoffers,andtheresultcontradictedtheexpectationofpureself-interest.Takingintoaccountthenormoffairness,thepredictionofthemodelisconsistentwiththelocalperformanceofthehousehold-basedconservationpolicy.Ithighlightsthepotentialoflocalsocialnormstofacilitateparticipationandcooperationbythelocalcommunityincommonnaturalresourceconservation.

©2010PublishedbyElsevierB.V.

Articlehistory:

Received9September2008

Receivedinrevisedform4June2010Accepted7June2010Availableonlinexxxx

Keywords:

Socialnormoffairness

Household-basedforestconservationGametheoryUltimatumGameWolong

1.Introduction

ThemodelofHomoeconomicus,whichhasguidedthemainstreamneoclassicaleconomicsforalongtime,assumesthatpeoplearenarrowlyself-regardingegoists(Gintis,2000).Inaccordancewiththisview,thetragedyofthecommonsseemsunavoidableaseveryonestrugglesforhisorherowninterestsandlosesthegameasawhole(Hardin,1968).However,empiricalcasestudies(Ostrom,1998;Ostrometal.,2002;Dietzetal.,2003)haverepeatedlyrejectedtheassumptionofuniversalselfishnessbysuggestingthatpeoplehavethecapabilitytopro-sociallycooperatewitheachother,thusgivingthecommongoodsgameahappyending.Inthesecases,membersofagroupusuallyconformtosomesocialnorms,i.e.,informalpublicrulesthatshapethesocialbehaviorinacertainregularandobservablepattern(Bicchieri,2005).Forinstance,empiricalexperimentshaveshownthatthesocialnormoffairnessiswidelyheldinhumansocieties(Sanfeyetal.,2003;FehrandFischbacher,2004;Bicchieri,2005;Henrichetal.,2005;Henrichetal.,2006;Dawesetal.,2007).Whenpro-socialnormsarestrongenoughtooverrideself-interestedmotives,cooperationincommonresourcemanagementcouldbeself-enforced.Asaresult,socialcapitalprovidesapossiblerouteforsustainablemanagementand

⁎Correspondingauthor.Tel.:+861062849191.E-mailaddress:zyouyang@rcees.ac.cn(Z.Ouyang).1Currentaddress:DepartmentofBiology,StanfordUniversity,Stanford,CA94305,UnitedStates.

0921-8009/$–seefrontmatter©2010PublishedbyElsevierB.V.doi:10.1016/j.ecolecon.2010.06.006

governanceofcommonnaturalresources(Pretty,2003),andboundedrationalitymodelsmightbeofgreateruseinenvironmentalresearchandpolicymaking(Venkatachalam,2008).

InChina,thecentralgovernmenthasrecentlylaunchedanationalNaturalForestConservationProgram(NFCP)asacriticalprojectforpreservingecosystemservices(Wangetal.,2007).Ahousehold-basedmanagementpolicyhasbeenimplementedinsomeprotectedareas,whereeachhouseholdorgroupofhouseholdsisassignedapatchofnaturalforesttomonitorandreceivesacertainannualsubsidyaccordingtothecontractifnoillegalloggingoccurswithinthepatch.Localgovernmentsoftenintuitivelyassumethatthematerialsubsidymotivatesthelocalhouseholdstoenforcetheforestmonitoring,butthesubsidybudgetforconservationisusuallyinsufficienttocoverthecostofintensivemonitoring(Liuetal.,2003).Nevertheless,somecasesshedlightonthedeadlock.Forinstance,thehousehold-basedNFCPinWolongNatureReserve(Fig.1),oneofthelargestgiantpanda(Ailuropodamelanoleuca)reserves,hassofarworkedeffectivelybyimprovingconservationoftheforestsandpandahabitatwhilekeepingthemonitoringintensitylow.Theevidencesuggeststhattherecouldbesomethingelseatworkotherthanthematerialincentivestoenforcesocialcooperationinforestconservation(Ostrom,2005).Oneofthepotentialfactorsthatneedtobestudiedisthesocialnormoffairness.

Theobjectiveofthisstudyistounveilthepossiblerolethesocialnormoffairnesscouldplayincommonnaturalresourceconservation.WefirstmodeledthehouseholdbehaviorintheNFCPtoanalyzethepotentialimpactofthefairnessnormontheequilibriaanddynamicsof

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

2Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx

Fig.1.ThemapofWolongNatureReserve.Thegraycolormarkshousehold-monitorednaturalforestpatches.Blackdotsmarkthedistributionofresidences,andexperimentswereundertakenin:V1—Wolongguan,V2—Huahongshu,V3—Yusidong,andV4—Gengda.

thesystem.Second,weundertookexperimentsbasedontheone-shotUltimatumGame(UG)inWolongtotesthowthelocalresidentsactuallyconformtothesocialnormoffairnessinageneralcontext.Basedonthemodelandtheexperimentdata,wemadeapredictionaboutthepossibleimpactofthefairnessnormoncooperationinthehousehold-basedNFCPandcompareditwiththedatafromafieldsurvey.Moreover,wediscussedtheevolutionandvariationofthesocial

norminthechangingenvironment,andthepossibleconsequencesfortheconservationofcommonnaturalresources.2.TheHousehold-basedNFCPModel

Inthemodelofthehousehold-basedNFCP,eachhouseholdorgroupofhouseholdsassignedapatchofnaturalforestactsasaplayer

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx3

thatisabletomakedecisionsindependently,andeachplayerplaystworolessimultaneously.Forthefirstrole,eachoneplayseitherasacooperator(C,whodoesn'tloginothers'patches)orasadefector(D,whologsinothers'patches).Forthesecondrole,eachonechoosestomonitortheforest(M)ornottomonitor(N).Weassumethereisnocorrelationbetweenthetworolesofaspecificplayer.

Thisisapopulationgamewithinfiniterounds.Ineachround,bdenotesthenetmaterialbenefitfromillegallogging,Sisthesubsidyperround,cisthemateriallossforfailingtomonitorone'sownforestaccordingtothecontract,eisthemonitoringeffort,andpisthepunishmentagainstthedefectorifheorsheisprosecuted.Weassumetheinequalitybbcholdssothataplayerwon'tbecomebetteroffbylogginginhisorherownpatch.

Thusweobtainthepayoffmatrixofastagegame,UbbeingtheutilitygainofadefectorfromloggingandUctheutilitylosssufferedbythosewhodon'tmonitorduetologging.

M

CD

0,S−e−p,S−e

N0,S

Ub,S−UcHerewetakeintoaccountnotonlyself-interestedincentivesbutalsotheregardforfairnessinthelinearutilityfunctions,asfollows(FehrandSchmidt,1999):Uc=c+αðb+cÞUb=b−βðb+cÞ:

ð1Þ

Theparametersαandβrespectivelyrepresentthesensitivitytothenegativefeelingsbornebyadefector(e.g.,guiltoranxiety)andbyonewhodoesn'tmonitor(e.g.,envyoranger)whentheyinteractwitheachotherwiththenormoffairnessviolated.Apurelyself-interestedplayer,forwhomα=β=0,doesn'tcareaboutfairness.Forsimplicity,wedonotdistinguishthemotiveofinequityaversionfromotherrelatedmotiveslikeguiltaversion.

Wefirstassumeapopulationthathashomogeneousconformitytothenormoffairnessandiswellmixedwithequalprobabilityofplayersrandomlymeetingeachother.Eachplayerupdateshisorherstrategyusingaproportionalimitationrulesuchthatthosewhoarerelativelyworseoffinagivenroundtendtotrytheotherstrategyinthenextround,withtheprobabilityofswitchinginproportiontothecurrentpayoffdisadvantage(HofbauerandSigmund,1998;Schlag,1998).Letx1,x2,y1andy2denotetheproportionsofcooperators,defectors,non-monitorsandmonitorsinthepopulation,respectively,sox1+x2=y1+y2=1.Definingfiandgjasthepayoffsofaggressivestrategyianddefensivestrategyj,fandgthecorrespondingaveragepayoffs,andvandwtherespectiveintensitiesofselection,weobtainthereplicatorequations:˙i=vxiðfi−fÞx

˙i=wyiðgi−gÞ:y

Lettingx=x1,y=y1,wethenobtainfromEqs.(1)and(2)thereplicatorequations

󰀄󰀁󰀃󰀅

Ub

˙=vxð1−xÞ1−1+yx

p

󰀁󰀃e

˙=wyð1−yÞ−1+x:y

Uc

ð2Þ

ð3Þ

Fig.2.ThepopulationdynamicsofstrategiesintheNFCP.Arrowsindicatedirections.(a)UbN0,UcNe.Thetrajectoryisconstantcirclesaroundtheunstableequilibrium(1−e/Uc,p/(p+Ub)).(b)UbN0,eNUc.Theequilibrium(0,1)isgloballystableandthefinalpopulationiscomposedofpuredefectors.(c)Ubb0.Theequilibrium(1,1)isgloballystableandthefinalpopulationiscomposedofpurecooperators.

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

4Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx

FromthereplicatorEq.(3),itisobviousthat(0,0),(1,0),(0,1),(1,1)and(1−e/Uc,p/(p+Ub))areallequilibria,butthedynamicsdependonthevaluesoftheparameters.Giventhedefaultparametersettings,UbN0,UcNe,nopurestrategyisdominant,andthetrajectoryappearsasaseriesofconstantcyclesaroundtheunstableinteriorequilibrium(1−e/Uc,p/(p+Ub))(Fig.2a).Iftheneededmonitoringeffortinput,e,exceedsUc,whileUbremainspositive,nottomonitorwillbeadominantdefensivestrategy,andallthetrajectorieswillconvergetothestablepoint(0,1)withoutcooperators(Fig.2b).

AnoptimalstableequilibriumexistsonlyontheconditionthatUbb0,i.e.,βNβ0,wherethethresholdβ0=b/(b+c)≥b/(b+S).Inotherwords,whenthematerialbenefitfromlogginginothers'patchescouldnotoffsetthepsychologicalsufferingafterviolatingthefairnessnorm,cooperationisadominantstrategy,andtheequilibriumisasymptoticallystable(Fig.2c).

Furthermore,ifconformitytothenormamongthepopulationisheterogeneousandtheproportionoftheplayersisqforwhomβNβ0holds,therearealwaysatleast,butnotlimitedto,afractionofqcooperators.Thisindicatesthat,giventheeconomicvariablesb,c,e,andS(especiallywhenthematerialsubsidySislimitedandcomparabletothemonitoringefforte)conformitytothefairnessnorm—particularlyanegativefeelinguponviolatingthesocialnorm—couldplayabigroleinmotivatingcooperationforconservation.3.TheCaseStudyandExperimentalDesign3.1.TheCaseStudyBackground

TheWolongNatureReservewasdesignatedin1963forprotectingtheendangeredgiantpanda,andwasexpandedtoapproximately2000km2in1975.Naturalforestsandmeadowstogethercoverabout98%oftheareainthereserve,providinghabitatsforover4000plantspecies,1700invertebratespeciesand450vertebratespecies(WolongNatureReserve,2005).

Apopulationofabout5000residents,over80%ofwhomareTibetans,nowliveinsixvillagesinthereserve.Thetraditionalpastoralistshavechangedintosedentaryfarmersaftertheintroductionoflowlandcropssuchasmaize,potatoandcabbage,andstall-fedlivestocklikepigs.Inrecentyears,growingnumbersofsmallhouseholdsandshrinkingfarmlands,togetherwithintensifiedinteractionswithexternalmarkets,haveforcedthelocalpeopletoseeknewemploymentforcashincome.However,educationandinformationdeficitsdiscouragethemfromcompetingforurbanjobs.Temporarylaborworkinlocalinfrastructureconstructionwasanimportantbutunstableformofemployment,andthedailywagewasaround50RMB(about7USdollars)in2007.Asmallfractionofthelocalhouseholdsdealinbusinessrelatedtotourism.

Foreverydaylivingneeds(cookingandheating),especiallyinthehalf-year-longcoldwinter,fuelwoodcollectionwasidentifiedasoneofthemostsevereanthropogenicthreatstopandahabitat(Anetal.,2001;Liuetal.,2001).Theconsequentdegradationofforestscontinuedevenafterthereservewasestablishedanddeforestationwasprohibited.Thereserveadministrationforcedconservationbyofficialmonitoringandsanctioning,butthepolicynotonlyprovedtobeinefficientbutalsorepressedanycooperativeincentivesofthelocalresidents.Therecentavailabilityofelectricityasanalternativeenergypartlyalleviatedtheconflict,butitcouldnotreplacefuelwoodastheprimaryenergysourcebecauseoftheresidents'limitedabilitytoaffordtheelectricbill(Anetal.,2002).

Itwasaftertheimplementationofthehousehold-basedNFCP(Wangetal.,2007)inWolongthatforestcoverbegantoberestoredinthereserve.Theforestsonbothsidesofthevalleyinwhichtheresidentsliveweredividedintopatchesandassignedtolocalhouseholdstomonitor.Eachhouseholdreceivedanannualsubsidyofaround900RMBperhousehold,onaverage,ifnoillegalloggingoccurredwithintheirspecificpatch.Theresidentshavenopropertyrighttoconsumeorallocatetheresourceintheforest.Anothersmallerproject,theGrainforGreenProgram(Xuetal.,2006),waslaunchedalmostsimultaneouslyto

reforesttheslopingfarmlandtocontrolerosion,butithasarelativelylimiteddirectimpactonthenaturalforestconservation.Itwasreportedin2003thatthehousehold-levelfuelwoodconsumptionhaddecreasedbyatleast40%sinceNFCPwasimplementedinthereserve,andafurtherdecreaseofwoodloggingpercapitafrom1.4m3to0.3m3wasrecorded(WolongNatureReserve,2005;Viñaetal.,2007).Accordingtoafieldsurveyin2005,only5sitesoutof387werefoundwithillegalloggingtrails(Liuetal.,unpublisheddata,2005).Butsurprisingly,theresidentsactuallypatrolataverylowfrequency,commonlyoncepermonthorso,suggestingthatitisinappropriatetoattributetheenhancedcommunitycooperationinforestconservationtothemonitoringintensity.

3.2.ExperimentalDesign

WeundertooktheUltimatumGame(UG)basedexperiments(Camerer,2003)infourvillagesinthereserve(Fig.1),includingthreeinWolongTownship,namelyWolongguan,HuahongshuandYusidong(V1,V2,V3forshort),andYicun(V4)inGengdaTownship.Subjectswerevoluntarilyrecruitedfrom31householdsonaverageineachvillage,54%ofwhomwerefemaleand89.4%wereTibetans(Table1).

Theparsimoniouscontext-freedesignofourexperimentreferredtoacross-culturalexperimentprogram(Henrichetal.,2006).WedidnotmentionanythingaboutNFCPintheexperimenttominimizethepotentialexperimenterdemandeffects(Zizzo,2008).Atthebeginningofeachgameweexplainedtothesubjecttherulesofthegame,andthentestedwhetherheorshefullyunderstood.Onlythosewhounderstoodtheruleswereeligiblefortheexperiment.Werandomlypairedthesubjectsinagameandkeptthemanonymoustoeachother.

Thefirstplayer,the‘proposer’,wasprovisionallyallottedasumofcashworth50RMB(about7USdollars).Wethenaskedtheproposertomakeadecisiontoshareanyportionofthemoney(intheunitof5RMB)withasecondperson,the‘responder’.Beforebeinginformedoftheactualoffer,theresponderwasaskedtodecidewhethertoaccepteachpossibleofferfrom0to50RMB.Iftheactualamountofferedwasaccepted,thentheresponderreceivedtheamountofferedandtheproposerreceivedtheremainder.Iftheactualofferwasrejected,thenbothsubjectsreceivednothing.Insuchananonymousone-shotgame,ifthesubjectsarerationalandonlyself-interested,theexpectedequilibriumisthattheproposershouldmakethelowestoffer(5RMB)whiletherespondershouldacceptanypositiveofferbecausesomethingisbetterthannothing.However,ifthereareother-regardingsubjectswhocareaboutfairness,deviationsfromtheequilibriumareexpectedinthegame.

InV1,V2andV3,afterallottingaproposerthestake,weaskedhimorheronemorequestiontoimitateaDictatorGame(DG)beforetheUGwasintroduced,leavingallthesubsequentstepsunchanged:Howmuchwouldyouliketoofferiftheresponderisnotgivenanoptiontorejecttheoffer?Theexpectedansweris0forapurelyself-interestedproposerwhetherheorsheisrisk-aversiveornot,becausethereisnoriskofbeingrejectedinaDG.

WegatheredsubjectstogetherinV1andV4,andundertookanothertwosessionsdoor-to-doorinV2andV3,sothatitwouldbepossibletoexaminetheunknownimpactfrompotentialcommunicationsandcollusionbetweensubjects.Beforeeachexperiment,wetookhouseholdsurveystocollectdemographicinformationonthehouseholds,andwealsohadpost-gameone-to-oneinterviewswithsomesubjectstolearntheirpersonalperceptionsoftheexperimentandthemotivationsbehindtheirchoices.

Table1

Thedemographicdataofthesubjectsintheexperiments.Age(year)43.2±1.2

aMale45.50%

Tibetana89.4%

Education(year)5.1±0.3

Familysize5.3±0.2

Headoffamily46.30%

OtherethnicgroupsareHanandQiang.

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx5

4.Results

4.1.TheExperimentalResults

TheresultsoftheexperimentsarepresentedinTable2.ThemeanofferoftheUltimatumGameis50.6%(±0.9%,s.e.m.),showingastrongdeviationfromtheexpectationofself-interestmaximization(Wilcoxonranksumtest,pb0.0001).Theuniquemodalofferofthegamesinthefourvillagesisthefairoffer,50%,whichtakesaportionof85.9%inthetotalnumberoftheoffers.Sub-fairoffersarerare,takingonly2outof64,whilealltherest11%arehyper-fairoffers(Fig.3a).

Theresponders'minimumacceptableoffers(MAO)showabimodaldistribution,with0astheprimarymodalMAOand50%asthesub-modalMAO(Fig.3b).TheMAOis17.0%±2.6%,whichisstronglylowerthanthemeanoffer(bootstraptest,pb0.0001).Therewasnosub-fairofferactuallyrejectedinthegames.GiventheactualdistributionofMAO,wecalculatedtheexpectedincomeoftheproposerswithdifferentoffers,andtheresultshowsthattheminimumpositiveoffer,10%(5RMB),istheoptimaloptionthatwouldleadtothehighestexpectedincome(Fig.4),and20%(10RMB)follows.Thispredictionisstronglydifferentfromtheactualofferdistribution(Wilcoxonranksumtest,pb0.0001),soincome-maximizingincentivescannotexplainthebehavioroftheproposers.

Risk-aversionisanotherpossibleexplanationtobetested,andwecouldnotexcludethepossibilitybasedonthelimitedinformationsolelyfromtheUGresult.However,theanswerstotheDGquestionshowaconsistentpattern(51%±0.8%)withtheUGoffers(Fig.5),withnodifferenceapparentinthepairwisetest(Wilcoxonranksumtest,p=0.8086).Asaresult,itindicatesthatrisk-aversionisnotasoundexplanationforinterpretingthepreferenceforfairoffers,either.

Anotherresultworthnotingisthestronglylowermaximumacceptableoffer(MXAO)comparedwiththerationalexpectation(Mean=70.4%,Schlagmeantest(Schlag,2008),pb0.001).Thedistributionisbimodalwiththetwopeaksat50%(therelativefrequencyis43.9%)and100%(therelativefrequencyis29.8%),whichindicatesthestrikingfactthatalargefractionofsubjectsdidnotwanttoacceptthehyper-fairoffer.Furthermore,atestcomparingthedeviationsfromfairofferbetweenMAOandMXAOshowsthattherespondersshowedanunusualstrongersensitivityagainstthepositivedeviations(Wilcoxonranksumtest,p=0.0001).

Furthermore,thedifferencesintheexperimentresultsacrossvillagesarenotsignificant(Table3),indicatingthatpatternsofconformitytothesocialnormoffairnesswereconsistentamongthelocalcommunitiesandthatthedifferentexperimentalmethodshadnosignificantimpactontheresults.

Table2

ResultsoftheUltimatumGameexperiments.Village

No.pairsStake(RMB)OfferMean%

Model%SD%

MAOMean%

Model%SD%

MXAOMean%

Model%SD%

abcdV1a175048.85012.612.4019.274.810021.8

V2b155050.6502.623.41023.272.650/10023.8

V3c165051.8504.014.6019.460.85019.0

V4d165051.2503.417.9022.671.45021.4

Total645050.6507.017.0021.070.45021.8

Wolongguan.Huahongshu.Yusidong.Gengda.

Fig.3.TheresultsoftheUltimatumGames.(a)Offers(inpercentage);(b)minimumacceptableoffers(MAO,inpercentage);and(c)maximumacceptableoffers(MXAO,inpercentage).

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

6Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx

5.Discussion

5.1.LinkingtheExperimenttoNFCP

TheUGbasedexperimentsweredesignedtoreflecthowthesubjectsconformedtothenormoffairnessinaneutralsocialcontext,andwedidnotassumethatthesocialnormmustremainstrictlyconstantacrossdifferentcontexts(Bicchieri,1999).Weusedcontext-freegameinstructionsratherthananexplicitdescriptionoftheNFCPinordertominimizepotentialexperimenterdemandeffects.Althoughaspecificcontextmayhelpsubjects'understandinginrelationtorealworldexperience,likeinsomeotherfieldexperiments(e.g.,CastilloandSaysel,2005),atthesametime,however,itmaydistractsubjectsandallowthemtocarryoverunrealisticexpectationstothetaskandinduceexperimenterdemandeffects(Zizzo,2008).Inthisspecificcase,anygivencuesaboutNFCPwouldlikelyraisethesubjects'concernaboutthesuspectedrelationshipbetweentheexperimentandthegovernment,andmightconsequentlyleadtoabiasedresponsebecausetheyknewwellthegovernment'sexpectation.

Ontheotherhand,thesenseofjusticehasdeepbiologicalandculturalunderpinnings,whichprobablyevolvedinthelonghistory,andhencethesocialnormsinthesamecommunitymustkeepsomeconsistencyacrossdifferentcontexts(Trivers,2006).SinceNFCPisonlyaminorpartofthelivelihoodofthelocalresidentsandithasahistoryoflessthanadecade,distinctsocialnormsareunlikelytohaveevolvedtherein,andtheresidentsplausiblyadoptedthegeneraltraditionalnormfordecision-makingintheunfamiliartask.Thus,itisreasonabletoexpectconsistencytosomenontrivialextentfromthesocialnorminthegeneralcontexttothatinNFCP.Inaddition,theanonymousone-shotexperimentenvironmentprovidedweakersocialbondsthandoesdailylifeortheNFCP,sotheexperimentresultmayslightlyunderestimateratherthanoverestimatetheimpactofthesocialnorminreallife.

Thepost-gameinterviewconfirmedthenormoffairness.Whenaskedaboutthereasonsfortheirchoicesintheexperiment,amajorityoftheintervieweesexplicitlyattributedittothepreferenceforfairness,andexpressedtheirantipathytoselfishnessandgreed(describedas“heavyheart”inthelocaldialect).Incontrast,theydidnotrefertotheirconsiderationofthegamepartners'expectation.Asaresult,itsuggeststhatthesubjectsmadetheirdecisionsintheexperimentsheuristicallybasedonsomesimplenormabouttheoutcomeitselfratherthanbasedonsomesecond-orderbeliefaboutthepartners'expectationswhichwouldneedcommunicationandmorecomplicatedcalculation(CharnessandDufwenberg,2006).Butwedon'tprecludethepossibleimpactsofmorecomplicatedsocialnormsinthedecision-makingindifferentcontexts,andthedifferentsocialnorms,ifany,shouldnotbeassumedtobemutuallyexclusive.

Notwithstandingthefactthatmostrespondersshowedstrongreluctancetorejectlowoffersintheexperiments,manyrevealedthattheywouldnotcooperatewitha“heavy-hearted”personanymore.Thisshowsthatwithholdinghelpbyshunningorostracismisamorepopularwaytoavoidconflictsandretaliationthancostlypunishmentinalong-runrepeatedgame,andthatthesecond-orderfreeriderproblemofpunishmentcouldthusbeavoided(Milinskietal.,2002;PanchanathanandBoyd,2004).Thiscouldalsoexplaintheresponders'strongersensitivityagainstpositivedeviationsfromthefairoffer,whichappar-entlycontradictsthegeneralpredictionthatnegativedeviationscountmore(Loewensteinetal.,1989;FehrandSchmidt,1999).5.2.TheNormofFairnessEvolvedwithReputation

Thealmostubiquityofthefairnessnormacrosssocietieswithdistinctcultures(Henrichetal.,2005)suggeststhattheremightbesomebiologicalunderpinningsforother-regardingnormsinhumannature.Butthenotablevariationindicatesthatecologicalandculturalfactorsprobablyshapethenormsinonewayoranother.Forinstance,comparedwiththeresultsoftheprecedingcross-culturalexperiments

Fig.4.Theexpectedincome(inRMB)ofaproposerwithrespecttoher/hisoffer(inRMB),giventheprobabilitydistributionofMAO.

4.2.TheEstimatedImpactoftheFairnessNorminNFCP

WeestimatedgeneralconformitytothenormoffairnessbyderivingtheparameterβaccordingtoEq.(1).IntheUG,theproposer'ssurplusbenefitequaledthelossofthepairedresponder,thatis,b=c.WethenrewrotethefirstequationofEq.(1)intoUb=b(1−2β).Obviously,Ubismaximizedwhenb=0ifandonlyifthemarginalutilitydUb/db≤0.Therefore,β≥0.5holdsforthesubjectwhoproposedafairorhyper-fairoffer,andtherelevantproportionintheexperimentis96.7%.

IfeveryoneconformsconsistentlytothefairnessnormintheNFCPasintheexperiment,acomparableproportion(96.7%)ofthehouseholdsareexpectedtochoosethestrategyofcooperationinforestconservationevenwithoutanymonitoringpressure,inaccordancewiththemodelinwhichbbc.Theactualproportionofcooperatorsmaydeviatefromtheexpectationwhenconformitytothesocialnormvariesindifferentcontexts.Ingeneral,thispredictioncoincideswithfieldsurveydataindicatingthatonly5sitesoutof387werefoundwithillegalloggingtrailswhenthemonitoringintensitywaslow(Liuetal.,unpublisheddata,2005).

Fig.5.Offers(infraction)comparisonbetweenDictatorGameandUltimatumGame.Nosignificantdifferencewasdeclared.p=0.8086.

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx

Table3

ComparisonofUltimatumGameresultsacrossvillages.Village

Test

V2Offer

V1V2V3

ab7

V3

MAO0.4960.109––––

MXAO0.8140.761––––

Offer0.2560.2050.8990.600––

MAO0.7400.7620.2780.228––

MXAO0.2450.0570.3870.136––

V4Offer0.3710.3500.9921.0000.9971.000

MAO0.5990.5520.7240.3450.7940.762

MXAO0.9890.1310.7700.9670.2380.686

ESTWTbESTWTESTWT

a0.5140.631––––

EST:thepvaluesofEpps–Singletontest.

WT:thepvaluesofWilcoxonranksumtest(exacttestadjustedforties).

(Henrichetal.,2006),thesubjectsofWolongshowedrelativelystrongerpreferenceforfairnessinourexperiment(Fig.3).ThisraisesthequestionofhowfairnesslocallyevolvedorremainedasanimportantsocialnorminthesmallcommunitiesofWolong.

ThenormoffairnessisunlikelytoevolveindependentlyfromtheexperienceofNFCPasdiscussedearlier.Instead,wespeculatethatthenormmightevolvefromrepeateddailysocialinteractionsbasedonreciprocity(FehrandGächter,2000)andreputation(Nowaketal.,2000).Mostofthelocalresidentshavelivedtogetherinasmallcommunityforgenerations,andmaycontinuetointeractformuchoftheirlives,soitisnotsurprisingthatcooperationplaysanimportantpartintheirdailylivelihood.Evennowadays,thefarmworkremainslaborintensive(e.g.herding,seedingandharvesting)andmosthouseholdsaretoopoortoaffordtherentofapastureorsomemechanicalfacilities.Theyalsocooperatevoluntarilytobuildtemplesandroads.Therefore,theyneedtobuildandmaintaintrustworthinessbetweeneachother.Theindividualdecisionwhetherornottocooperatewithanothermaydependontheco-player'sreputation,andfairorhyper-fairofferscanhelptosignalone'sgenerosityandbuildupone'sgoodreputation(Gintisetal.,2001).Inadenselypopulatedcommunitywithhighviscosity,gossipscouldamplifytheinfluenceofone'sreputation(Sommerfeldetal.,2007).

However,forthespecificcaseofWolong,itistooearlytoconcludewhetherthenormoffairnessevolvedlocallyorwasinheritedfromthecultureandbiologicalnatureoftheancestorslongbeforethelocalcommunitywasformed.Theanswermightbeboth.

Asaresult,notonlytheeconomicinfluencebutalsothesocialimpactofgrowingmarketinteractionsmustbetakenintoconsideration.Thesub-optimalscenariocannotberuledout;i.e.thatthelimited-sizetourismandotherbusinesscouldnotbringinsufficientprofitstoeradicatepovertyfromthelocalcommunity,butthesocialcapitalwouldirreversiblydeclinethereafterandthuscooperationforconservationwouldbreakdown.Forthesakeofthesustainabilityofsocial–ecologicalsystems,thefunctionandevolutionofsocialcapitalcallsformoreattention.6.Conclusion

TheconservationofthegiantpandahabitatandtheecosystemservicesinWolongNatureReservehasremarkablyimprovedsincetheimplementationofthehousehold-basedNFCP,eventhoughthelimitedmaterialsubsidycouldnotenforcethecooperativecommitmentofthelocalresidentsiftheywereonlyself-regarding.ThisstudyindicatesthatthelocalsocialnormoffairnessmaybeplayingapositiveroleinfacilitatingcommunitycooperationinNFCPastheresidentsarereluctanttobenefitfromillegalloggingatthecostoftheirneighbors'loss.Itthereforecallsattentiontotheneedforpolicydesigntotakeintoaccounttheactivelocalsocialnorms.Moreover,inadynamicview,thereismuchworktodotounveilhowsocialnormswillevolveinchangingsocial–economiccontexts,andhowthiswillimpactthemanagementofacommonnaturalresource.Acknowledgements

5.3.SustainableDevelopmentandSocialCapital

Thehousehold-basedNFCPhassofarworkedeffectivelyinWolongNatureReserve.However,consideringtheuncertaintyofthefuture,especiallytheimpactofongoingeconomicdevelopment,thesustain-abilityofnaturalresourcemanagementinthelongrunstillprovidesachallenge.Ontheonehand,thegrowingexposuretomarketsmaycreatemoreopportunitiesforemploymentandincreasedincomesothattherelianceonfuelwoodandothernaturalresourcescouldbediminished.Ontheotherhand,thereisstilluncertaintyanddisagreementoverthesocialimpactoftheexperiencewithmarkets.Someevidencehintsthattheintegrityofthemarketmayenhancethenormoffairness(Gurven,2004),whereastheoppositepossibilityalsoexists,thatthebondingsocialcapitalinthelocalcommunitymightdeclineduetolesstimeandincentivetocooperatewitheachother(Putnam,2000).

InWolong,thesubjectswithcommercialexperiencewithtourismshowedasignificanthighertendencytobewillingtoacceptloweroffers(n=12,p=0.0269),inspiteofthesmallsample.Giventhegrowingeconomicindependenceandthepotentiallydecliningsocialcapital,thenormoffairnessmightloseground,followingasimilartrendasrecognizedbyGhimire:“theabandonmentofthecommunesystemandinfluenceofmodernizationhavealsomeantasteadydeclineinthepreviousreciprocitysystemofneighborshelpingeachotherwithhousebuilding”(Ghimire,1997).

Wethankthevillagersfortheirparticipation,JianYang,JinquanZhouandWeiLiufortheirassistanceintheexperimentandthehouseholdsurvey,JinCao,JoeHenrichandKarlSchlagfortechnicaladvices,MarcusFeldman,ArneTraulsen,HuaZheng,DeborahRogersandthreeanonymousreviewersforcommentsonthedraft.WealsoacknowledgethesupportfromNationalNaturalScienceFoundationofChina(No.40621061)andtheMajorStateBasicResearchDevelopmentProgramofChina(973Program,2009CB421105).References

An,L.,Liu,J.,Ouyang,Z.,Linderman,M.,Zhou,S.,Zhang,H.,2001.Simulating

demographicandsocioeconomicprocessesonhouseholdlevelandimplicationsforgiantpandahabitats.EcologicalModelling140,31–49.

An,L.,Lupi,F.,Liu,J.,Linderman,M.,Huang,J.,2002.Modelingthechoicetoswitchfrom

fuelwoodtoelectricity:implicationsforgiantpandahabitatconservation.EcologicalEconomics42,445–457.

Bicchieri,C.,1999.Localfairness.PhilosophyandPhenomenologicalResearchLIX,

229–236.

Bicchieri,C.,2005.Grammarofsociety.TheNatureandDynamicsofSocialNorms.

CambridgeUniversityPress.

Camerer,C.F.,2003.Behavioralgametheory.ExperimentsinStrategicInteraction.

PrincetonUniversityPress.

Castillo,D.,Saysel,A.K.,2005.Simulationofcommonpoolresourcefieldexperiments:a

behavioralmodelofcollectiveaction.EcologicalEconomics55,420–436.

Charness,G.,Dufwenberg,M.,2006.Promisesandpartnership.Econometrica74,

1579–1601.

Dawes,C.T.,Fowler,J.H.,Johnson,T.,McElreath,R.,Smirnov,O.,2007.Egalitarian

motivesinhumans.Nature446,794–796.

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006

8Z.Songetal./EcologicalEconomicsxxx(2010)xxx–xxx

Nowak,M.A.,Page,K.M.,Sigmund,K.,2000.FairnessversusreasonintheUltimatum

Game.Science289,1773–1775.

Ostrom,E.,1998.Abehavioralapproachtotherationalchoicetheoryofcollective

action.AmericanPoliticalScienceReview92,1–22.

Ostrom,E.,2005.Policiesthatcrowdoutreciprocityandcollectiveaction.In:Gintis,

H.,Bowles,S.,Boyd,R.,Fehr,E.(Eds.),MoralSentimentsandMaterialInterests:TheFoundationsofCooperationinEconomicLife.MITPress,Cambridge,Mass,pp.253–276.

Ostrom,E.,Dietz,T.,Dolsak,N.,Stern,P.C.,Stonich,S.,Weber,E.U.(Eds.),2002.The

DramaofTheCommons.NationalAcademyPress,Washington,DC.

Panchanathan,K.,Boyd,R.,2004.Indirectreciprocitycanstabilizecooperationwithout

thesecond-orderfreeriderproblem.Nature432,499–502.

Pretty,J.,2003.Socialcapitalandthecollectivemanagementofresources.Science302,

1912–1914.

Putnam,R.D.,2000.BowlingAlone:TheCollapseandRevivalofAmericanCommunity.

SimonandSchuster,NewYork.

Sanfey,A.G.,Rilling,J.K.,Aronson,J.A.,Nystrom,L.E.,Cohen,J.D.,2003.Theneuralbasis

ofeconomicdecision-makingintheUltimatumGame.Science300,1755–1758.Schlag,K.H.,1998.Whyimitate,andifso,how?Aboundedlyrationalapproachto

multi-armedbandits.JournalofEconomicTheory78,130–156.

Schlag,K.H.,2008.Exacthypothesistestingwithoutassumptions.Mimeo.

Sommerfeld,R.D.,Krambeck,H.-J.,Semmann,D.,Milinski,M.,2007.Gossipasan

alternativefordirectobservationingamesofindirectreciprocity.ProceedingsoftheNationalAcademyofSciences104,17435–17440.

Trivers,R.L.,2006.Reciprocalaltruism:30yearslater.In:Kappeler,P.M.,vanSchaik,C.P.

(Eds.),CooperationinPrimatesandHumans:MechanismsandEvolution.Springer,Berlin,Heidelberg,pp.67–84.

Venkatachalam,L.,2008.Behavioraleconomicsforenvironmentalpolicy.Ecological

Economics67,640–645.

Viña,A.,Bearer,S.,Chen,X.,He,G.,Linderman,M.,An,L.,Zhang,H.,Ouyang,Z.,Liu,J.,

2007.TemporalchangesingiantpandahabitatconnectivityacrossboundariesofWolongNatureReserve,China.EcologicalApplications17,1019–1030.

Wang,G.,Innes,J.L.,Lei,J.,Dai,S.,Wu,S.W.,2007.ECOLOGY:China'sforestryreforms.

Science318,1556–1557.

WolongNatureReserve,2005.HistoryoftheDevelopmentofWolongNatureReserve.

SichuanSciencePublisher,Chengdu,China.

Xu,J.,Chen,L.,Lu,Y.,Fu,B.,2006.Localpeople'sperceptionsasdecisionsupportfor

protectedareamanagementinWolongBiosphereReserve,China.JournalofEnvironmentalManagement78,362–372.

Zizzo,D.J.,2008.ExperimenterDemandEffectsinEconomicExperiments.SSRN.

Dietz,T.,Ostrom,E.,Stern,P.C.,2003.Thestruggletogovernthecommons.Science302,

1907–1912.

Fehr,E.,Fischbacher,U.,2004.Socialnormsandhumancooperation.TRENDSin

CognitiveSciences8,185–190.

Fehr,E.,Gächter,S.,2000.Fairnessandretaliation:theeconomicsofreciprocity.Journal

ofEconomicPerspectives14,159–181.

Fehr,E.,Schmidt,K.M.,1999.Atheoryoffairness,competitionandcooperation.The

QuarterlyJournalofEconomics114,817–868.

Ghimire,K.B.,1997.Conservationandsocialdevelopment:anassessmentofWolong

andotherpandareservesinChina.In:Ghimire,K.B.,Pimbert,M.P.(Eds.),SocialChangeandConservation:EnvironmentalPoliticsandImpactsofNationalParksandProtectedAreas.EarthscanPublications,London.

Gintis,H.,2000.BeyondHomoeconomicus:evidencefromexperimentaleconomics.

EcologicalEconomics35,311–322.

Gintis,H.,Smith,E.A.,Bowles,S.,2001.Costlysignalingandcooperation.Journalof

TheoreticalBiology213,103–119.

Gurven,M.,2004.Marketexposureandeconomicgamebehavior.In:Henrich,J.,Boyd,

R.,Bowles,S.,Camerer,C.F.,Fehr,E.,Gintis,H.(Eds.),FoundationsofHumanSociality:EconomicExperimentsandEthnographicEvidencefromFifteenSmall-scaleSocieties.OxfordUniversityPress,NewYork,pp.194–231.Hardin,G.,1968.Thetragedyofthecommons.Science162,1243–1248.

Henrich,J.,Boyd,R.,Bowles,S.,Camerer,C.,Fehr,E.,Gintis,H.,McElreath,R.,Alvard,

M.,Barr,A.,Ensminger,J.,Henrich,N.S.,Hill,K.,Gil-White,F.,Gurven,M.,Marlowe,F.W.,Patton,J.Q.,Tracer,D.,2005.‘EconomicMan’incross-culturalperspective:behavioralexperimentsin15small-scalesocieties.BehavioralandBrainSciences28,795–855.

Henrich,J.,McElreath,R.,Barr,A.,Ensminger,J.,Barrett,C.,Bolyanatz,A.,Cardenas,J.C.,

Gurven,M.,Gwako,E.,Henrich,N.,Lesorogol,C.,Marlowe,F.,Tracer,D.,Ziker,J.,2006.Costlypunishmentacrosshumansocieties.Science312,1767–1770.

Hofbauer,J.,Sigmund,K.,1998.EvolutionaryGamesandPopulationDynamics.

CambridgeUniversityPress,Cambridge.

Liu,J.,Linderman,M.,Ouyang,Z.,An,L.,Yang,J.,Zhang,H.,2001.Ecologicaldegradation

inprotectedareas:thecaseofWolongNatureReserveforgiantpandas.Science292,98–101.

Liu,J.,Ouyang,Z.,Pimm,S.L.,Raven,P.H.,Wang,X.,Miao,H.,Han,N.,2003.Protecting

China'sbiodiversity.Science300,1240–1241.

Loewenstein,G.F.,Thompson,L.,Bazerman,M.H.,1989.Socialutilityanddecisionmaking

ininterpersonalcontexts.JournalofPersonalityandSocialPsychology57,426–441.Milinski,M.,Semmann,D.,Krambeck,H.-J.,2002.Reputationhelpssolvethe‘tragedyof

thecommons’.Nature415,424–426.

Pleasecitethisarticleas:Song,Z.,etal.,Theroleoffairnessnormsthehousehold-basednaturalforestconservation:ThecaseofWolong,China,Ecol.Econ.(2010),doi:10.1016/j.ecolecon.2010.06.006