’WPC+q JU> %Z1` 0C‘U<Ō# 0D1c D+”Ęæ 0C… 0EČ AY  0KfU>± B*ļU< fU aW fk bm aq f… C‡ a” fµ B•· 0[Ō a/ C  Š A©K ō U <ś &6 –!¦!¶!Ę!Ö!ę!ö!p"n­\ 0( !v'!p†*!ö/!3!6w@1 !&9!6<Ęq Ę7 Ęż ĘĆ!F?!VB!fE!vH!†K!–N!p¦Q!W!&Z!6]!F` )4Vc žŠc" (ee‰'ō\  `&Times RomanomanX 'Ż ƒ/“)ŻŻ  ŻÓ  Ó(’’$‘‘ņņŚ  ŚŚ  Śóóō\  `$Times NewRomanHP LaserJet 5M (PCL),ß'3 ŠpetŠ ^-®(4 HP5MPCL.PRSš,\,\šš’ĄHĄ !ėL- 0(3/“$©©Ż ƒ-!ŻŻ W Ż 'Ż ƒ/“)ŻŻ  ŻÓ  Ó'ČČČČdxd… Level 1 Level 2 Level 3 Level 4 Level 5(2€•$££Ż ƒ-!ŻŻ W Ż(’’$––ņ ņ(Ś  ŚŚ  Ś)ó ó Mü’ << deUU( ±ž$’’ņ ņFigure€Ś  Ś1Ś  Śó ó'ō\  `&Times Romanoman&’’0 d d%ō\  `$Times NewRomanTable_A łTable_A ’’’dTable_C d(3ę $„„Ż ƒ-!ŻŻ W ŻŚ  Ś0Ś  Ś.Table_F6  ›&OLE 2.0 Box <=8’’’’’’’’’’C’’ HKKKK½“Mtw%ō\  `$Times NewRoman ŲŲŲŲ«'ō\  `&Times RomanomanXoō\  PŽĀCXP($”” GEHIEJKELMENOEPLevel 1Level 2Level 3Level 4Level 5‚Level 1Level 2Level 3Level 4Level 5ƒLevel 1Level 2Level 3Level 4Level 5„Level 1Level 2Level 3Level 4Level 55Ō’y3|@b"‚u«Ü,Ł’’’’’’’’’’’’’’’’’’^:DPdd؜DDDdp4D48dddddddddd88pppXø„„|pDL|°p„p|¼|D8D\dDXdXdXDdd88d8œddddDL8ddddX`(`lńDl\DDD4DDDDDDdDdXXXXX°‚„X|X|X|X|XD8D8D8D8ddddddddddXdbdddpdXXX„X„X„X„X~|X|X|X|XddddddddD8D8D8D8ŒpLd|8|P|8|D|8dvdddd°„D„D„DpLpLpLpL|T|8|8dddddd¼d|X|X|Xd|8d„DpL|8dd~dd|XD88d44dd$/ŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒNHxxH\dDXddddĄd8@˜˜d<@˜˜d˜<DDXXdČDDxddÄHxxHppČDXØČ<«˜˜øČČd˜ČxtldpÄxx„„„„1plll l(llppČČddČdÄÄÄÄ4ŽPtDp,T`d"‚u«Ü,Ł’’’’’’’’’’’’’’’’’’^:DpddČØDDDdp4D48ddddddddddDDpppd¼„„|œœLdœ„¼œ|œp„Č„D8DtdDdpXpXDdp8Dp8ØpdppXLDpdddXP,PhńDhtDDD4DDDDDDdDpdddddȐX„X„X„X„XJ8J8J8J8pœdœdœdœdppppddpšdœdddzpdddXXXX”„X„X„X„XœdœdœdœdœdœdœpœpL8L8L8L8¢pdœp„8„Z„8„P„8pppœdœdʐXXXpLpLpLpL„f„D„DppppppȐd„X„X„Xp„8pXpL„Ddd’œdp„XL88œd44dd6/ŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒŒNHxxHjdDdddddĄd><q*«WWxxxWWkkxxx<xxxxåkkxxxk<((xxxxxWHxkWWWWWWWWWW(x<.<<<</>/>/>/x]‚S‚S‚S‚Sx]x]x]x]xSxSx]€S‚SxSxSf]xSxSxSxIxIxIxIx{nInInInI‚S‚S‚S‚S‚S‚S‚]‚]?/?/?/?/‡]S‚]n/nKn/nCn/x]xxx]x]‚S‚S„xxIxIxI]?]?]?]?nUn9n9x]x]x]x]x]x]§xxSnInInIx]n/x]xI]?n9xSxSxz‚Sx]nI?//‚S4+SS-/uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuN8HH"&H>XHH8HB8>HH^HH>"".2",2,2,"222N2222"&22H22,006ń"6."""""""""2"2H,H,H,H,H,XAB,>,>,>,>,""""H2H2H2H2H2H2H2H2H2H2H,H2H1H2H2H282H,H,H,B,B,B,B,H?>,>,>,>,H2H2H2H2H2H2H2H2""""F8&H2>>(>>">H2H;H2H2H2H2XHB"B"B"8&8&8&8&>*>>H2H2H2H2H2H2^HH2>,>,>,H2>H2B"8&>H2H2H?H2H2>,"H2422/FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFN$<<$.2",2222`2 LL2 LL2L"",,2d""<22b$<<$88d",Td«LL\dd2Ld<:628b<<BBBB186666668HHH8dd22d2bbbbG(:"8*02H"‚u«Ü,Ł’’’’’’’’’’’’’’’’’’^"*22TN"""2D"2222222222""DDD2\>>BH>>HH",B8TBH>H>28H>T>88&&*2"22,2,22,H2222&&2,B,,&((6ń"6*"""""""""2"2>2>2>2>2>2XBB,>,>,>,>,""""B2H2H2H2H2H2H2H2H28,>2H2H2H28,H2>2>2>2>2B,B,B,B,H<>,>,>,>,H2H2H2H2H2H2H2H2""""L2,B,88$88 8B2B:B2B2H2H2^B>&>&>&2&2&2&2&8$88H2H2H2H2H2H2TB8,8&8&8&H28B2>&2&88,8,H,"H2422/FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFN$<<$42&22222b2 LL2 LL2L""882X""<22b$<<$22X"8Td«LL\dd2Ld<:62Db<<BBBB1D666666DHHHDdd22d2bbbb>(:"D**2H"‚u«Ü,Ł’’’’’’’’’’’’’’’’’’^&822TN"""28"2222222222""8882TBBBHBBHN&2B>XHH>HB8>HBXB>>""82"22,2,"282N8222&&8,B2,&""8ń"88"""""""""2"2B2B2B2B2B2^HB,B,B,B,B,&&&&H8H2H2H2H2H8H8H8H8>,B2H2H2H2>,H2>2B2B2B2B,B,B,B,HJB,B,B,B,H2H2H2H2H2H2N8N8&&&&R82B2>>4>>&>H8HFH8H8H2H2^HB&B&B&8&8&8&8&>6>>H8H8H8H8H8H8XB>,>&>&>&H2>H8B&8&>>,>,HJH2H8B,&H2422/FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFN$<<$22&22222`2LL2"JJ2L""222d""<22d$<<$88d"2Td«LL\dd2LdA.SSxSSJJSJ§ńSSSSSS8SSSSSSSSSxJxJxJxJxJ”orJiJiJiJiJ8.8.8.8.{SxSxSxSxS{S{S{S{SxSxJ{SxSxSxS{S`SxIxJxIqIqIrJrJ{dgIiJiJgIxSxSxSxSxSxS{S{S8.8.8.8.z]AuSg/gZ?Z?Z?YAgYAi.xSxS{dxS{SiJ8..xS488SS88/§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§§NxxxSSS8JDDSS’SSS””S;SSSS”;88VVS”++§§SSdd§S§SxSc‰‰]]”8VSÉ;«xx§§§Sx§§Wxx””§§§S§§§§å””§§§”S88§§§§§xd§”xxxxxxxxxx8§SxS]S”§x§oS8SxJS`x‚l§§xxxxxxxxxx§§§§§M§xxxx§§§§§§§§§§§§§§§§§§§xx§§§§§§‰od‚§x§§§Gcxxxx§§§§§§ÅxxxSxx§xx§§§xxxJxxxxJxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx8xxx8xxx8xxx8xxxxxxxxxxxxxŲxdi]d]oJiAlJ{SxJ8.uJo]—]{JoSxJxd`SdSSiJxJodx]‚dddxdiA{S8..xJxJJddJdddd888SSSSSSSddddddddddddddddddddddAAAAAAASSSSSSSSSSSSSSSSSSSSSS............JJJJJJJJJJJJJJJJJJJdddddddddddddddddddddd`SJ8§"‚u«Ü,Ł’’’’’’’’’’’’’’’’’’^#.C<J"J.J"VCVTVCVCVń)>F)))))))))<)CVACVVVCxx<I\\š>>>\g0>03\\\\\\\\\\33gggQ©„yy„rg„„>F„r”„„g„ygr„„¬„„r>3>T\>Q\Q\Q>\\33\3\\\\>F3\\„\\QX%Xcń>cT>>>0>>>>>>\>\„Q„Q„Q„Q„Q”wyQrQrQrQrQ>3>3>3>3„\„\„\„\„\„\„\„\„\„\„Q„\„Z„\„\„\g\„Q„Q„QyQyQyQyQ„trQrQrQrQ„\„\„\„\„\„\„\„\>3>3>3>3€gF„\r3rIr3r>r3„\„l„\„\„\„\”„y>y>y>gFgFgFgFrMr3r3„\„\„\„\„\„\¬„„\rQrQrQ„\r3„\y>gFr3„\„\„t„\„\rQ>33„\40\\!/€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€NBnnBT\>Q\\\\°\3;‹‹\7;‹‹\‹7>>QQ\·??n\\“BnnBgg·>Qš·7«‹‹©··\‹·njc\g“nnyyyy1gcccc%ccg„„„g··\\·\““““0‚Ij>g(MX\„"‚u«Ü,Ł’’’’’’’’’’’’’’’’’’^5>g\\·š>>>\g0>03\\\\\\\\\\>>ggg\¬„y„„yrF\y¬„r„gy„„·„„y>3>j\>\gQgQ>\g3>g3šg\ggQF>g\„\\QI(I_ń>_j>>>0>>>>>>\>g„\„\„\„\„\·„„QyQyQyQyQD3D3D3D3„g\\\\„g„g„g„g„\„\„g\\„\„\pg„\„\„\„Q„Q„Q„Q„ˆyQyQyQyQ\\\\\\ggF3F3F3F3•g\gy3ySy3yIy3„g„„„g„g\\¶„„Q„Q„QgFgFgFgFy^y>y>„g„g„g„g„g„g·„„\yQyQyQ„gy3„g„QgFy>„\„\„†\„gyQF33\40\\2/€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€NBnnBa\>\\\\\°\7>‹‹\7>ˆˆ\‹7>>\\\·??n\\·BnnBgg·>\š·7«‹‹©··\‹·nlc\g“nnyyyy1gcccc(_cg„„„g··\\·\““““0„Ij>g2gF\%&'()*+./01245:;<=>?@ABCD)ī\  p`&Times New Roman’,ō4  p.Times New RomanBold€)ī*f9 p`2Times New RomanItalic)Ͼ¦ p`CG Times,ō4  p.Times New RomanBoldŠ*Ą¶č9 p<Times New RomanBold ItalicTA‘7RCR,,XRž’ž’ž’ž’ž’öž’’’g’’’’’’{’’’’ģ’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’āž’ž’ž’ž’’’”‘7WCW,,?XW_ž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’’’’’’’q’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’T?W!1(1,,h1ž’ž’ž’ž’ž’ž’ž’’’I’’’’’’’ ’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’Ųž’ž’ž’ž’’’”7t,F5F,,?ąF_ž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’’’’’’’{’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’TNt,B5B,,ąBž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’'“5RCR,,†XRöŻž’ž’ž’ž’ž’ž’ž’’’’’’’’’’’Ī’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’TJ8RCR,,ŃXRŁž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’ž’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’Tņ‘7RCR,,XRž’ž’ž’Äž’öö’’g’’’’’’S’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’āž’ž’ž’ž’’’TdV"1(1,,Ńh1Łž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’ž’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’”F‘7WCW,,XW_ž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’Ō=’6TCT,,IXT꓾’ž’ž’ž’ž’ž’ž’’’’’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’AX 1(1,,†h1öŻž’ž’ž’ž’ž’ž’ž’’’’’’’’’’’ ’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’TPt,B5B,,ąBž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’ž’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’T^y.D8D,,ōDž’ž’ž’ž’ž’+ž’’’5’’’’’’ŗ’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’œ ž’ž’ž’ž’’’”;y.I8I,,ōI_ž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’a{,D8D,,†ōDöŻž’ž’ž’ž’ž’ž’ž’’’’’’’’’’’°’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’ŌUz-F8F,,IōF꓾’ž’ž’ž’ž’ž’ž’’’’’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’TXH)!),,,)ž’ž’ž’ž’ž’ž’ž’’’¦’’’’’’ž’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’GI)!),,†,)öŻž’ž’ž’ž’ž’ž’ž’’’’’’’’’’’! ’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’ŌYI*!*,,I,*꓾’ž’ž’ž’ž’ž’ž’’’’’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’Tqx/D8D,,ŃōDŁž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’ž’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’Ō~W!2(2,,Ih2꓾’ž’ž’ž’ž’ž’ž’’’’’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’”uW!4(4,,h4_ž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’TŠ…2K=K,,&Kž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’ˆ ’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’ž’ž’ž’’’”…2P=P,,&P_ž’ž’ž’ž’ž’ž’ž’’’ž’’’’’’’’’’’’’ž’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’ž’ž’’’’’’’’’Ż ƒ-!ŻŻ W Ż›Ń7€XXdé7ŃŃ  ŃÓ  ÓŃ€Ńń`ńÖ€’’ńCa:;Öń`ńń`ńÖ€’’ńCa:;Öń`ńĢą@ģOģąˆą  ąrevised€Ō ƒIŃŌAugust€27,€1996Ō!  ŌŠ ™é Šą@ģOģąˆą  ąCopyright€1996Š ‚Ņ ŠĢĢĢÓ  Óą@@**&ģąņ ņDO€CONTINGENT€VALUATION€ESTIMATES€PASS€A€"SCOPE"€TEST?ó óŠ & v Š€ņ ņą@@GG:ģąA€META€œANALYSISó óņņ1›óóŠ  _ ŠĢĢĢĢą@@!!1ģąV.€Kerry€Smith€and€Laura€OsborneĢĢĢ€€Arts€and€Sciences€Professor€of€Environmental€Economics,€Duke€University€and€ResourcesĻfor€the€Future€University€Fellow€and€Assistant€Professor,€Department€of€Economics,€GeorgiaĻState€University,€respectively.€€ĢĢÓ  ÓÓÓĢņ ņņņÓ  ÓĢĢĢÓ  ÓScope€and€Contingent€ValuationŠ S£ ŠPlease€address€all€correspondence€to:Š _ Šóóó óÓÓProfessor€V.€Kerry€SmithŠ Ė МDepartment€of€Economics›ĢDuke€UniversityĢBox€œ90097›ĢDurham,€NC€27708Ģą @` ąPhone:€€919„613„8052ĢFax:€ą @ ąą @` ą919„684„8741ĢĢÓÓĢĢ1ņ ņņņŠ  ·*&+ ŠŌ‡XXXXŌóóDo€Contingent€Valuation€Estimates€Pass€a€š šScopešš€Test?Š ° ŠA€Meta€Analysisó óŠ ‚Ņ ŠĢV.€Kerry€Smith€and€Laura€OsborneĢĢAbstractĢÓ  Óą  ąThis€paper€considers€œthe€scope€test›€proposed€to€judge€the€internal€consistency€ofĻcontingent€valuation€estimates.€€œThe€test€is€shown€to€be›€quite€sensitive€to€the€maintainedĻhypotheses€required€to€derive€fairly€precise€expectations€for€the€properties€of€WTP€œfunctions,›Ļœand,€therefore,›€a€different€approach€may€be€needed€in€gauging€the€reliability€of€CV.€€ThisĻpaper€describes€œan›€approach€that€relies€on€a€weight€of€the€evidence€criterion€and€uses€metaĻanalysis€to€develop€a€systematic€appraisal€of€what€are€the€economic€values€of€changes€inĻamenity€resources.€€The€approach€is€illustrated€for€the€case€of€estimating€peopleššs€willingnessĻto€pay€for€improving€(or€maintaining)€visibility€at€the€national€parks.ĢŌ#†XXXXź#Ōņ ņņņŠ  ,| ŠI.ą  ąIntroductionó óóóŠ ° Šą  ąFew€aspects€of€current€economic€research€created€as€much€controversy€as€the€use€ofĻcontingent€valuation€to€elicit€people's€values€for€environmental€resources.€€While€contingent€Ļvaluation€(CV)€surveys€have€been€in€the€literature€over€twenty€years,€until€recently€theyĻattracted€little€attention€from€most€economists.ņņ2óó€€The€Exxon„Valdez€oil€spill€changed€this€stateŠ ā 2 Šof€"tolerant"€indifference,€and€marked€the€beginning€of€a€continuing€debate€over€the€reliabilityĻof€CV.€€A€central€premise€of€those€arguing€that€"...CV€is€a€deeply€flawed€methodology€thatĻdoes€not€estimate€what€its€proponents€claim€to€be€estimating"€(Diamond€and€Hausman€[12]€p.Ļ62)€is€that€there€have€been€no€internal€tests€of€the€validity€of€CV€results.ņņ3óó€€For€Diamond€andŠ *z ŠHausman€[12]€(DH),€one€way€to€evaluate€the€CV€method€is€to€compare€willingness€to€pay€Ļ(WTP)€functions€estimated€with€CV€surveys€with€the€specific,€observable€properties€thatĻeconomic€theory€implies€WTP€should€follow.€€DH€suggest€this€type€of€internal€validity€test€isĻespecially€important€in€situations€where€nonuse€values€dominate€the€responses€used€toĻconstruct€monetary€measures€of€economic€value.Ģą  ąThe€purpose€of€this€paper€is€to€consider€one€of€the€proposed€internal€consistency€testsĻfor€CV„based€estimates€of€WTP.€€This€test€requires€CV€estimates€of€willingness€to€pay€to€beĻresponsive€to€the€amount,€or€ņņscopeóó,€of€the€environmental€amenity€being€offered.€€To€testŠ ŗ!   Šwhether€CV€estimates€of€WTP€are€responsive€to€scope€we€conducted€a€meta€analysis€thatĻpools€WTP€estimates€for€visibility€improvements€at€U.S.€national€parks€from€five€differentĻcontingent€valuation€studies.€€Our€findings€support€a€positive,€statistically€significant,€(andĻrobust)€relationship€between€the€WTP€estimates€and€the€percentage€improvement€in€visibleĻrange.Š Ų*(&* Šą  ąBefore€the€results€of€our€meta€analysis€may€be€considered€evidence€that€€WTPĻestimates,€based€on€contingent€valuation€are€consistent€with€an€internal€validity€test,€theĻrequirements€of€a€scope€test€must€be€defined.€€The€scope€test€was€initially€proposed€by€aĻpanel€of€social€scientists,€composed€by€the€General€Counsel€of€€the€National€Oceanic€and€ĻAtmospheric€Administration€(NOAA).€€They€were€asked€to€evaluate€the€reliability€of€CV€forĻmeasuring€the€passive„use€values€of€environmental€resources€involved€in€natural€resourceĻdamage€assessments.ņņ4óó€€The€panelššs€report€(Arrow€et.al.€[2])€calls€for€WTP€estimates€that€areŠ œģ  Šš šadequatelyšš€responsive€to€the€scope€of€the€environmental€insult.€€€Four€of€the€original€sixĻNOAA€panelists€submitted€subsequent€comments€to€clarify€the€meaning€of€š šadequateĻresponsiveness.šš€€They€noted€a€clear€distinction€between€observing€a€statistically€significantĻdifference€in€the€WTP€estimates€for€two€different€sized€changes€in€an€environmental€good€andĻš šplausibly€responsivešš€values,€observing€that:ĢҰŅҰŅÓÓą ` ąš šA€survey€instrument€is€judged€unreliable€if€it€yields€estimates€whichĻare€implausibly€unresponsive€to€the€scope€of€the€insult.€€This,€of€course,€is€aĻjudgment€call,€and€cannot€be€tested€in€a€context„free€manner...šš€(Arrow,ĻLeamer,€Schuman€and€Solow€[1]).ĢҰŅҰŅĢÓÓTo€understand€the€theoretical€expectations€for€€š šplausibly€responsivešš€estimates€of€WTP,Ļsection€II€outlines€the€properties€WTP€responses€might€have€under€alternative€descriptions€ofĻhow€environmental€resources€enter€individual€preferences.€€We€use€our€meta€analysis€toĻexamine€whether€the€existing€empirical€evidence€indicates€plausibly€responsive€estimates€ofĻWTP€have€been€elicited€in€CV€surveys€for€a€one€environmental€commodity.€€Section€IIIĻdescribes€the€studies€included€in€out€meta€analysis€and€presents€the€results.€€To€includeĻestimates€where€both€use€and€nonuse€considerations€are€likely€to€be€factors€in€respondentsššĻchoices€we€used€CV€studies€that€had€both€on„site€surveys€and€surveys€conducted€at€locationsŠ Ó+#'+ Šthat€were€substantial€distances€away€from€the€national€parks.€€Variables€that€control€forĻfeatures€of€each€study€are€also€included€in€the€WTP€œmodels.€€The€last€section€discusses€theĻimplications€of€this€application€for€broader€uses€of€meta€analysis.›ĢĢņ ņņņII.ą  ąTesting€Economic€Consistency€of€WTPó óóóŠ ų H Šą  ąMeasures€of€the€economic€value€of€any€object€of€choice€are€constructed€from€peoples'Ļdecisions.€€Revealed€preference€(or€indirect)€methods€used€to€estimate€the€value€for€non„¼ń`ń¼ń`ńmarketed€goods€rely€on€theoretical€restrictions€to€supplement€observed€decisions€involvingĻmarketed€goods.€€These€restrictions€are€maintained€assumptions€that€are€required€to€isolateĻestimates€of€WTP€for€nonmarketed€environmental€resources.€€As€a€result,€the€analyst€cannotĻdiscern€whether€inconsistencies€in€any€estimated€WTP€value€indicate€an€unreliable€method€orĻthe€failure€of€one€or€more€of€the€maintained€assumptions.ņņ5óó€€Although€CV€surveys€can€provideŠ  š Šdirect€estimates€of€WTP,€they€are€not€free€of€important€maintained€hypotheses.€€Indeed,€theĻprocess€of€defining€tests€for€hypotheses€about€the€economic€consistency€of€contingentĻvaluationššs€WTP€estimates€requires€additional€maintained€hypotheses.€€Ģą  ąA€recent€example€illustrating€the€assumptions€needed€to€define€a€scope€test€for€WTPĻestimates€is€Diamondššs€[10]€derivation€of€the€properties€a€WTP€function.€Diamondššs€analysisĻdemonstrates€that€if€preferences€can€be€described€with€a€quasi„linear,€indirect€utility€function,Ļand€if€the€object€of€choice€involves€saving€an€amount€of€the€environmental€resource€fromĻbeing€lost€(e.g.,€avoiding€a€specified€number€of€birds€being€killed€from€a€larger€population),Ļthen€a€scope€test€would€require€the€ratio€of€two€WTP€estimates€for€two€different€numbers€ofĻbirds€saved€to€be€greater€than€the€ratio€of€the€respective€numbers€of€birds€saved.€€ForŠ Ō*$&* Šexample,€if€one€were€to€compare€saving€1,000€versus€100,000€birds,€this€framework€wouldĻimply€the€WTP€to€save€100,000€birds€must€be€at€least€100€times€greater€than€the€WTP€to€saveĻ1,000€birds.€€Unfortunately,€this€clear„cut€expectation€for€the€scope€properties€of€the€WTPĻfunction€does€not€hold€generally.€€If€we€assume€the€amount€of€a€resource€saved€by€theĻproposal€is€not€a€perfect€substitute€for€original€environmental€resource,€as€might€be€the€case€ifĻpeople€do€not€believe€the€proposal€would€be€completely€effective€or€if€the€resource€is€notĻperceived€by€individuals€on€a€simple€metric€such€as€a€count€of€birds€lost,€then€it€isĻstraightforward€to€show€that€the€only€bound€on€the€scope€test€we€can€provide€is€that€of€theĻNOAA€panel€„„€a€larger€amount€of€an€object€of€choice€should€have€a€larger€WTP€than€aĻsmaller€amount.ņņ6óóŠ b Šą  ąTo€understand€fully€how€an€individualššs€WTP€responds€to€all€types€of€changes€in€theĻobject€of€choice€(or,€equivalently,€the€economic€commodity)€would€require€a€rich€CV€designĻin€which€the€issues€influencing€the€perceived€substitutes€for€the€resource€and€other€aspects€ofĻa€proposed€policy€could€be€varied€as€part€of€the€survey€instrument.€€By€controlling€for€theseĻother€influences€it€would€be€possible€to€evaluate€the€partial€effect€of€scope.€€A€more€modestĻalternative€is€available€if€we€are€willing€to€use€past€CV€research.€€That€is,€by€using€€existingĻCV€estimates€for€specific€environmental€resources€it€is€possible€to€partially€mimic€the€surveyĻdesign€required€to€evaluate€these€issues.€€Evaluating€WTP€estimates€in€this€way€is€oneĻparticular€use€of€meta€analysis€for€testing€the€theoretical€properties€of€an€underlyingĻbehavioral€relationship.€€In€what€follows,€we€use€measures€of€the€economic€value€of€visibilityĻimprovements€at€parks€in€the€U.S.€to€illustrate€our€proposed€€approach€for€evaluating€CVĻestimates€of€WTP.Š ź*:&* Šą  ąņ ņĢą0  ąą ° ąIII.ą  ą€ą0` (#(#ąą  ąMeta€Analysis€Estimates€of€WTP€Functions€for€Visibility€Changes€at€NationalŠ ‚Ņ ŠParks€ó óŠT¤` (#` (# Šą  ąA€scope€test€meeting€the€recommendations€of€the€NOAA€Panel€must€demonstrate€thatĻWTP€is€responsive€to€the€amount€of€the€economic€commodity€offered.€€As€Arrow€et€al.€[1]Ļnote,€this€demonstration€is€not€simply€isolating€statistically€significant€differences,€it€alsoĻrequires€that€the€differences€be€consistent€with€economic€intuition.€€Unfortunately,€as€theyĻacknowledge,€these€prior€expectations€will€be€context€dependent.€€Ideally,€we€would€haveĻsome€prior€knowledge€about€how€people€conceptualize€what€is€offered€to€them€in€a€CVĻchoice.€€While€we€know€people€can€accurately€relate€differences€in€photos€of€differentĻvisibility€conditions€to€changes€in€the€visible€range,€this€does€not€mean€these€differences€fullyĻcharacterize€how€each€individual€interprets€a€CV€question€offering€a€policy€or€plan€intendedĻto€provide€some€visibility€change.ņņ7óó€€To€address€this€issue,€we€consider€both€the€sensitivity€ofŠ ˆŲ Šour€results€to€the€models€used€and€to€the€sample€œcomposition.›Ģą  ąThe€data€for€our€analysis€of€households'€valuation€of€improvements€in€visibility€wereĻassembled€from€a€review€of€thirteen€contingent€valuation€studies€presenting€air€pollutionĻchanges€and€their€influences€on€visibility€and€other€local€conditions.€€We€selected€five€ofĻthese€studies€for€the€meta€analysis€because€they€used€comparable€methods.€€These€studiesĻfocused€on€air€quality€as€a€key€element€in€visibility€and€associated€vistas€that€are€present€atĻspecific€national€parks.ņņ8óó€€In€addition,€these€studies€present€the€change€in€air€quality€in€a€wayŠ F'–"& Šthat€permits€computation€of€the€change€in€visible€range.€€Table€1€identifies€the€five€studiesĻand€indicates€the€number€of€observations€drawn€from€each€study;€€the€author(s)€of€each€study;Š ź*:&* Šthe€park€at€which€visibility€is€valued;€€a€few€summary€measures€of€the€survey€design;€WTPĻestimates;€€and€the€visibility€changes€considered€in€each€study.€€€Each€observation€in€our€metaĻanalysis€includes,€as€the€dependent€variable,€the€estimate€of€mean€willingness€to€pay€(MWTP)Ļfor€a€visibility€change€as€originally€reported€in€each€study.€€This€mean€is€typically€computedĻover€some€subset€of€the€total€study€sample€(i.e.,€subsamples€based€on€specific€changes€inĻvisibility€or€survey€design).€€Each€MWTP€estimate€has€been€elicited€as€(or€converted€to)€aĻmonthly€payment€expressed€in€1990€dollars€using€the€Consumer€Price€Index.Ģą  ąThe€proportionate€change€in€distance€of€visible€range€is€the€representation€for€theĻamenity€being€valued.€€Each€CV€study€used€photos€depicting€the€change€in€visibility€offeredĻat€each€national€park€to€describe€the€commodity€to€survey€respondents.€€Because€there€isĻsubstantial€evidence€supporting€people's€ability€to€connect€the€differences€in€visible€rangeĻpresented€with€photos€displaying€different€visibility€conditions€to€the€proportionate€change€inĻthe€distance,€this€proportionate€change€is€used€in€our€statistical€summaries€of€peoplesššĻresponses€to€the€CV€questions€associated€with€these€proposed€changes.€€As€might€be€expected,Ļthe€photos€used€in€each€study€depicted€different€scenes€(even€when€they€were€studying€theĻsame€parks).€€Moreover,€the€CV€questions€were€intended€to€ask€respondents€to€abstract€fromĻthe€specific€location€of€the€pictures€and€to€use€each€photo€as€a€means€of€portraying€conditionsĻat€the€site.€€€Thus,€it€is€important€to€acknowledge€that€the€importance€of€visibility€as€aĻcomplementary€dimension€of€a€vista€depends€on€what€is€œactually€presented›€as€part€of€the€CVĻquestion€in€relation€to€the€intended€object€of€choice€„„€the€visibility€conditions€at€a€specifiedĻpark.Ģą  ąThe€text€of€the€CV€questions€requires€a€description€of€the€frequency€(or€probability)Š ź*:&* Šthat€visibility€conditions€would€resemble€those€in€the€photographs.€€For€example,€the€surveyĻused€in€the€Chestnut„Rowe€[5]€study,€described€a€range€of€visibility€conditions€and€identifiedĻthe€average€conditions€as€part€of€the€description.€€In€their€CV€questions,€respondents€wereĻasked€about€changes€in€those€average€conditions.€€By€contrast,€the€Balson€et€al.€[3]€studyĻpresented€and€asked€valuation€questions€about€the€changes€for€a€specific€number€of€daysĻduring€a€season.€€Although€there€is€reasonably€wide€variation€in€the€proportionate€changes€inĻvisibility€conditions€across€all€five€studies,€each€study€does€not€treat€other€characteristics€thatĻmight€also€define€the€commodity,€such€as€the€number€of€days€when€the€changes€take€place,€inĻa€systematic€way.€€Therefore,€in€addition€to€the€absolute€value€of€the€proportionate€change€inĻthe€visible€range,€we€include€variables€describing€the€commodity,€the€complementaryĻresources,€or€any€descriptive€information€that€we€hypothesized€would€influence€how€theĻcommodity€is€interpreted.Ģą  ąThe€first€commodity„related€variable€included€in€the€data€is€a€qualitative€variable€thatĻdistinguishes€surveys€for€eastern€versus€western€parks.€€We€separated€the€visibility€effects€forĻthese€two€locations€because€the€vistas€affected€by€changes€in€visible€range€are€quite€differentĻin€western€parks€than€in€eastern€sites.ņņ9óó€€A€second€issue€related€to€the€commodity€arisesŠ žN Šbecause€in€some€cases€the€survey€questions€ask€respondents€about€their€WTP€to€avoid€aĻdecline€in€visibility.€€This€framing€corresponds€to€an€equivalent€surplus€definition€rather€thanĻa€compensating€surplus€format€of€the€WTP€questions.€€Because€we€have€treated€all€visibilityĻchanges€as€positive€values€(i.e.€the€proportionate€change€in€visible€range€that€is€provided€orĻavoided),€a€qualitative€variable€indicating€whether€or€not€the€WTP€is€for€an€increase€orĻdecrease€in€visible€range€is€also€included.€ą Ą ąThe€last€commodity„related€variable€includedŠ ź*:&* Šconcerns€the€extent€of€the€change€„„€whether€visibility€changes€are€to€occur€at€a€single€park€orĻfor€the€southwest€region€as€a€whole.€€In€the€abstract,€we€might€regard€this€variable€asĻexpanding€the€"amount"€of€improvement€in€visibility.€€However,€it€is€not€clear€howĻrespondents€would€interpret€questions€that€focus€on€local€verses€regional€changes.€€BecauseĻvisibility€changes€are€generally€explained€to€respondents€as€€arising€from€changes€in€airĻpollution,€the€physical€nature€of€the€process€that€"delivers"€them€to€a€single€park€affects,€toĻsome€degree,€all€parks€in€the€same€region.€€To€the€extent€respondents€understand€thisĻmechanism,€they€may€well€include€some€perception€of€effects€throughout€the€area€in€theirĻresponses€to€a€question€about€visibility€at€a€specific€park,€even€when€they€are€not€explicitlyĻidentified.Ģą  ąOur€sample€is€composed€of€a€limited€number€of€studies€and€includes€studies€usingĻpractices€that€are€not€part€of€current€CV€methodology.€€To€attempt€to€account€for€thisĻlimitation,€variables€describing€several€aspects€of€the€survey€design€are€also€included.€€TheseĻqualitative€variables€identify:€(1)€the€elicitation€format€„„€whether€the€WTP€was€elicited€withĻiterative€bidding,€a€direct€question,€or€a€format€that€allowed€respondents€to€select€their€desiredĻresponse€from€a€list€of€values€presented€to€them€(payment€card€or€check„list);€(2)€whether€orĻnot€the€interview€took€place€at€the€park€in€question€(or€with€households€at€other€locations);€Ļand€(3)€whether€or€not€households€live€in€the€state€where€the€park€is€located.€€Some€surveysĻasked€respondents€their€WTP€to€achieve€the€specified€visibility€change€at€the€respondentsššĻnext€visit€to€the€park.€€These€scenarios€are€identified€with€a€qualitative€variable,€because€theyĻintroduce€some€uncertainty€(i.e.€whether€the€respondent€would€have€a€future€use)€and€becauseĻthey€link€the€change€in€visible€range€to€future€use.ņņ10óóŠ ź*:&* Šą  ąBefore€discussing€our€results,€several€reasons€for€caution€in€interpreting€them€need€toĻbe€highlighted.€€In€some€cases,€the€qualitative€variables€defined€to€represent€specific€effectsĻmay€capture€multiple€influences€because€the€effects€are€present€in€only€one€study.€€ForĻexample,€Balson€et€al.€[3]€was€the€only€study€to€use€a€direct€question€to€elicit€WTP.€€As€weĻnoted€earlier,€it€also€focused€on€short€durations€of€visibility€change.€€Similarly,€Rowe,€dššArgeĻand€Brookshire€[24]€was€the€only€study€to€use€iterative€bidding€to€elicit€WTP€and€ChestnutĻand€Rowe€[5]€was€the€only€study€which€involved€a€mailed€survey.€€These€examples€of€study„¼ń`ń¼ń`ńspecific€factors€highlight€how€important€it€is€to€acknowledge€that€a€statistical€summaryĻinvolving€a€limited€number€of€studies€can€confound€specific€study€attributes€with€a€fixedĻeffect€that€could€be€included€to€identify€each€study.€€Thus,€a€summary€model€using€the€dataĻacross€studies€can€either€include€attributes€that€are€unique€to€studies€or€it€can€include€a€fixedĻeffect€factor,€but€not€both.€€Whichever€approach€to€distinguishing€features€of€the€studies€isĻchosen,€the€estimated€parameters€will€reflect€elements€of€the€other€omitted€factor.Ģą  ąTable€2€provides€eight€model€specifications€for€visibility€improvements€at€nationalĻparks.€€The€variable€names€used€in€Table€2€describe€the€hypothesized€influences€„€economicĻand€methodological€„€for€estimated€WTP.€€Each€of€these€effects€is€€a€qualitative€variableĻ(other€than€the€proportionate€change€in€visibility)€which€has€a€value€of€one€when€the€effect€isĻpresent.€€The€qualitative€variable€identifying€eastern€sites€is€entered€in€the€models€as€anĻinteraction€term€with€the€proportionate€change€in€visibility.€€The€default€category€forĻqualitative€descriptions€of€the€question€format€is€check„off€or€payment„card€format.€€BecauseĻour€estimates€are€based€on€a€pooled€sample€using€the€€mean€WTP€as€the€dependent€variable,Ļand€are€drawn€from€different€sample€sizes,€it€is€reasonable€to€assume€there€will€be€violationsŠ ź*:&* Šin€the€ideal€conditions€usually€maintained€with€ordinary€least€squares€(OLS).€€To€take€accountĻof€these€influences,€we€used€a€Huber€[19]€„€White€[29]€consistent€covariance€estimator€inĻcomputing€the€standard€errors€for€our€hypothesis€tests.€€This€approach€treats€each€study€as€theĻequivalent€of€a€sample€cluster€with€the€potential€for€œheteroscedasticity›€(i.e.€differences€in€errorĻvariance€across€clusters).€€Ģą  ąEach€model,€labeled€(œ1a›),€(œ1c›)€and€(2)€through€(6),€was€estimated€using€the€log€ofĻmean€WTP€(for€specified€subsets€of€the€full€sample€in€each€study)€specified€as€a€linearĻfunction€of€variables€describing€the€type€of€elicitation€procedure€and€location€of€the€interview.€ĻThe€models€labeled€(œ1b›)€and€(œ1c›)€correspond€to€the€Box„Cox€[4]€and€feasible€generalized€leastĻsquares€(œFGLS›)€estimates€for€the€most€complete€specification,€respectively.€€The€estimatedĻBox„Cox€parameter,€œšš›,€is€significantly€different€from€zero€suggesting€the€model€may€have€lessŠ ä4 Šcurvature€than€what€is€implied€by€the€semi„log€form.ņņ11óó€€The€Box„Cox€parameter€is€alsoŠ ¶ Šsignificantly€different€from€one,€implying€that€the€form€lies€between€linear€and€semi„log€byĻthis€criterion.€€While€the€coefficients€of€the€Box„Cox€form€are€not€directly€comparable€to€theĻother€estimates€(i.e.,€they€are€not€measures€of€the€percentage€change€in€WTP€with€a€change€inĻeach€independent€variable,€as€is€the€case€of€the€semi„log),€the€sign€and€statistical€significanceĻof€the€visibility€change€measures€and€all€other€variables€are€consistent€with€the€semi„logĻresults€using€either€OLS€with€the€Huber€estimates€of€coefficientsšš€standard€errors€or€the€œFGLS›Ļresults.€€The€œFGLS›€estimates€use€the€number€of€observations€in€each€sub„sample€providingĻthe€estimate€of€the€mean€WTP€to€correct€for€œheteroscedasticity›.€€Ģą  ąAll€of€the€models€in€Table€2€implicitly€assume€the€proportionate€change€in€visibleĻrange€is€an€adequate€measure€of€the€economic€commodity.€€This€formulation€implies€thatŠ ź*:&* Šrespondents€interpret€the€policy€as€providing€improvements€highly€substitutable€for€theĻoriginal€conditions.ņņ12óó€€The€remaining€models€in€Table€2€are€distinguished€by€the€variablesŠ ‚Ņ Šincluded€as€potential€determinants€of€mean€WTP€responses€and€the€studies€composing€theĻsample.€€Model€(œ1a›)€is€the€most€inclusive,€using€the€full€sample€and€each€of€the€explanatoryĻvariables€described€above.€€Model€(2)€also€uses€the€full€sample,€but€includes€a€limited€numberĻof€explanatory€variables.€€Model€(3)€is€our€preferred€specification€for€the€WTP€function€whenĻthe€Balson€et€al.€[3]€study€of€the€Grand€Canyon€is€excluded€from€the€data.€€Because€this€studyĻemphasizes€the€small€number€of€days€that€would€be€involved€in€the€visibility€changes,€oneĻmight€hypothesize€that€this€commodity€may€be€considered€differently€by€respondents.€ModelĻ(4)€is€our€preferred€specification€for€the€WTP€function€when€eastern€sites€are€not€included.€ĻAgain,€we€hypothesize€that€changes€in€visibility€conditions€at€eastern€sites€might€beĻconsidered€a€different€commodity,€and€as€such€arguably€should€not€be€considered€in€the€sameĻmeta€analysis€as€visibility€changes€at€western€parks.€€Models€(5)€and€(6)€exclude€both€theĻBalson€et€al.€œ[3]›€study€and€eastern€sites,€and€model€(5)€includes€a€more€explanatory€variablesĻthan€model€(6).€€These€models€illustrate€that€our€conclusions€about€the€link€between€WTP€andĻchanges€in€the€visible€range€are€not€sensitive€to€the€specified€determinants,€whether€or€not€theĻmost€restricted€sample€is€used.Ģą  ąRegardless€of€the€sample€composition€or€model€specification,€there€is€a€statisticallyĻsignificant,€positive€relationship€between€willingness€to€pay€and€proportionate€improvement€inĻthe€visible€range.€€The€effect€of€a€change€in€visible€range€on€WTP€is€smaller€in€magnitudeĻwhen€the€Balson€et€al.€[3]€study€is€excluded€from€the€data.€€This€result€is€consistent€with€theĻframing€of€the€visibility€change€used€in€Balson€et€al.€€It€was€the€only€study€in€our€sample€toŠ ź*:&* Šdescribe€the€number€of€days€when€the€visibility€change€would€take€place.€€Because€it€is€notĻpossible€to€capture€this€feature,€as€distinct€from€other€differences€in€this€studyššsĻimplementation€of€CV,€the€observed€response€in€WTP€to€proposed€visibility€change€includesĻanother€dimension€when€this€study€is€included€in€the€sample.€€Table€3€repeats€the€analysisĻwith€the€full€sample,€using€a€linear€form€with€an€intercept,€as€well€as€both€a€linear€and€semi„¼ń`ń¼ń`ńlog€specification€without€an€intercept.ņņ13óó€€Overall,€the€results€remain€unchanged€with€theseŠ Ź   Šmodels,€although€the€sign€changes€for€some€qualitative€variables€because€the€š šdefault€stateššĻfor€these€variables€has€changed.Ģą  ąThe€estimated€WTP€function€embodies€a€scope€response€that€is€due€to€the€specificationĻselected€for€the€function€and€the€measure€used€to€quantify€the€commodity€offered€toĻrespondents.€€Any€comparison€of€our€results€with€Diamondššs€scope€multiplier€(i.e.,€the€ratioĻof€the€WTP€for€one€quantity€change€in€comparison€to€the€WTP€for€another€smaller€quantityĻchange€of€the€same€commodity€š šshouldšš€exceed€the€ratio€of€the€quantities€offered)€isĻconditional€by€these€maintained€hypotheses.€€Nonetheless,€if€we€simply€compare€differentĻlevels€for€the€proportionate€change€in€visible€range,€models€with€estimated€coefficients€for€thisĻvariable€exceeding€about€.5€will€satisfy€the€Diamond€lower€bound.€€This€is€consistent€with€ourĻestimates€of€the€implied€substitution€between€the€change€offered€and€natural€conditions€(seeĻnote€12).€€It€would€not€be€satisfied€for€visibility€changes€at€the€Eastern€sites.€€However,€this€isĻnot€a€violation€of€the€scope€condition€because€it€can€as€easily€reflect€the€nature€of€visibilityĻchanges€in€that€context.Ģą  ąBefore€considering€a€second€measure€of€the€plausibility€of€our€estimatedĻresponsiveness€of€WTP€to€scope,€we€consider€œthe€effects€that€the›€factors€related€to€theŠ ź*:&* Šcommodity€have€on€the€WTP€estimates.€€In€an€earlier€meta€summary€of€CV€studies€forĻvisibility,€Chestnut€and€Rowe€[6]€indicated€that€eastern€sites€have€lower€values€than€theĻwestern€locations.€€This€distinction€is€also€supported€by€our€models,€suggesting€a€differentĻtype€of€complementarily€between€visible€range€and€the€scenic€vista€for€eastern€sites.€€TheĻWTP€to€avoid€a€decline€is€larger€than€WTP€to€acquire€an€improvement€over€what€areĻdescribed€as€current€conditions.€€This€effect€is€somewhat€sensitive€to€sample€composition€andĻthe€coefficient€estimates€are€only€statistically€significant€(at€the€ten€percent€level)€for€modelsĻthat€include€the€Balson€et€al.€data.Ģą  ąWhen€the€questions€are€linked€to€future€uses,€through€the€description€of€how€theĻchange€would€be€available€(at€the€time€of€the€next€visit€to€the€park),€the€WTP€responses€areĻsignificantly€lower.€€Questions€about€visibility€conditions€for€the€whole€region€lead€to€lowerĻWTP€responses€as€well,€but€these€findings€are€sensitive€to€the€sample€composition€and€to€theĻestimator€(e.g.€model€(œ1b›)).€€Moreover,€when€we€consider€models€excluding€an€intercept€(seeĻtable€3),€this€effect€becomes€positive€and€highly€statistically€significant.ņņ14óó€€If€this€variable€wereŠ ZŖ Štreated€as€a€measure€of€the€"amount"€of€improvement,€then€this€result€could€be€interpreted€asĻbeing€incompatible€with€one€dimension€of€a€scope€test.€However,€respondents€may€notĻinterpret€regional€changes€as€larger€than€park„specific€changes€(e.g.,€they€may€assume€thatĻpolicies€affecting€air€pollution€that€impacts€a€specific€park€would€also€affect€nearby€parksĻbecause€of€the€physical€processes€determining€visibility).€€Without€specific€information€onĻhow€people€conceptualize€š šmorešš€visibility€we€cannot€resolve€this€issue.ņņ15óó€€However,€it€isŠ F'–"& Šclear€that€the€positive€and€significant€relationship€between€WTP€and€the€proportionate€changeĻin€visibility€is€not€affected€by€the€treatment€of€this€variable.€Š ź*:&* Šą  ąThe€Balson€et€al.€[3]€study€was€the€only€one€to€use€a€direct€question€format.€€TheĻcoefficient€estimate€for€š šdirect€questionšš€indicates€MWTP€responses€were€significantly€lowerĻthan€other€studies,€but€this€effect€cannot€be€distinguished€from€the€lower€number€of€daysĻspecified€in€this€study€as€well.€€The€Rowe,€dššArge,€and€Brookshire€[24]€study€generallyĻelicited€responses€that€are€statistically€significantly€higher€than€the€other€studies.€€We€cannotĻsay€this€result€is€due€to€their€use€of€"iterative€bidding"€because€they€were€the€only€study€toĻuse€this€format.€€The€MWTP€values€from€respondents€interviewed€at€a€park€are€significantlyĻlower€than€responses€elicited€elsewhere.€€Of€course,€as€in€the€other€attributes,€this€effect€mayĻbe€confounded€by€the€issues€specific€to€the€individual€studies€and€not€a€reflection€of€theĻeffects€of€on„site€interviews.€€Only€two€studies€conducted€interviews€at€a€park€site,ϜMacFarland›,€œMalm›,€and€œMolenar›€[22]€and€Rowe,€dššArge,€and€Brookshire€[24]€(whichĻcontributed€3€observations€in€this€category).€€Ģą  ąFinally,€the€estimated€parameter€for€a€variable€identifying€respondents€as€residents€ofĻthe€state€where€the€park€(being€studied)€was€located€has€a€positive,€but€not€statisticallyĻsignificant€effect€on€models€(œ1a›)€through€(œ1c›).€€However,€in€the€log„linear€model€with€noĻintercept€(see€Table€3)€this€coefficient€estimate€is€statistically€significant,€suggestive€of€a€use„¼ń`ń¼ń`ńrelated€effect.Ģą  ąIn€terms€of€our€overall€objectives,€the€consistent€link€between€the€size€and€sign€of€theĻrelationship€between€the€visibility€change€and€the€WTP€is€not€affected€by€changes€in€theĻspecified€determinants€of€WTP,€the€functional€form,€the€sample€composition,€or€the€treatmentĻof€the€intercept.€€These€findings€imply€the€CV€estimates€for€visibility€change€are€related€to€theĻchanges€in€the€visible€range€described€with€the€photos€used€in€the€original€CV€studies.€Š ź*:&* ŠBecause€some€of€the€CV€analyses€(especially€Chestnut€and€Rowe€[5]€and€Balson€et€al.€[3])Ļwere€undertaken€with€respondents€who€could€be€expected€to€have€use€and€passive„use€(orĻnonuse)€values,€these€estimates€are€responsive€to€the€œNOAA›€Panelššs€initial€focus€onĻmeasuring€passive„use€values€and€are€not€the€result€of€a€large€sample€accentuating€theĻstatistical€significance€(Arrow€et€al.€[1]).Ģą  ąThe€statistical€association€between€WTP€and€changes€in€visible€range€may€not,€byĻitself,€suggest€the€WTP€estimates€are€"plausibly"€responsive€in€the€manner€Arrow€et€al.€[1]Ļpropose€for€scope€tests.€€A€second€approximate€gauge€of€the€plausibility€of€the€WTP€estimatesĻis€to€consider€how€important€the€various€aspects€of€the€economic€commodity€(or€object€ofĻchoice)€are€to€the€WTP€modelsšš€ability€to€detect€a€relationship€between€just€the€proportionateĻchange€in€visible€range€and€WTP.€€Focusing€solely€on€aspects€of€the€economic€commodity,ĻTable€4€compares€estimates€of€the€scope€effect€based€on€a€common€model€which€includesĻonly€factors€that€describe€the€economic€commodity,€but€which€vary€according€to€whichϜsubsample›€is€included€to€estimate€the€parameters.ņņ16óó€€While€this€was€implicitly€done€with€theŠ ZŖ Šmodels€in€Table€2,€we€also€changed€the€independent€variables€included€in€those€models,€thusĻconfounding€the€effects€of€sample€changes€with€œcovariate›€changes.€€Table€4€also€includesĻmeasures€of€the€base€visibility€condition€for€the€relevant€sample,€the€average€change€inĻvisibility€for€the€relevant€sample,€and€the€inter„œquartile›€ranges€for€these€statistics.Ģą  ąRecognizing€that€each€model€in€Table€4€has€omitted€variables,€we€focus€on€whether€orĻnot€factors€important€to€the€commodity€itself€are€omitted€and€how€these€omitted€factors€affectĻthe€estimates€of€WTP.€€Perhaps€the€most€important€omitted€factors€involve€descriptions€of€theĻamount€of€time€the€changes€in€visible€range€were€to€occur€and€the€vistas€that€are€affected€byŠ ź*:&* Šthese€changes.€All€studies€conveyed€different€information€about€the€timing€of€changes€inĻvisible€range.€€The€Balson,€et€al.€study€is€the€most€stark€example€because€it€focused€on€a€veryĻsmall€number€of€days€visibility€changes€were€to€occur€(as€little€as€10€days€in€the€summerĻmonths).€€The€second€factor,€the€vistas€affected€by€visibility€changes,€are€captured€at€a€veryĻsimple€level€with€the€interaction€term€between€visible€range€and€Eastern€sites.€€TheĻphotographs,€and€associated€vistas€used€to€describe€visibility€conditions€to€respondents€wereĻdifferent€for€all€studies,€even€those€focusing€partially€on€the€same€parks.€€One€way€to€attemptĻto€appraise€the€effects€of€both€these€factors€is€by€changing€the€sample€composition€used€toĻestimate€WTP.€€The€results€in€Table€4€suggest€including€the€Balson€et€al.€studyššs€estimates€orĻWTP€œresponses›€for€visibility€changes€at€Eastern€sites,€without€accounting€for€their€influence,Ļreduces€the€ability€of€the€proportionate€change€in€visibility€alone€to€explain€the€results.ņņ17óó€€TheŠ ä4 Šestimated€standard€error€(statistical€significance)€declines€dramatically€for€the€visibilityĻmeasure€when€we€attempt€to€include€Balson€et€al.€in€the€model.€€Further,€omitting€bothĻEastern€sites€and€Balson€et€al.€results€in€the€most€significant€association€between€WTP€andĻproportionate€changes€in€visible€range.€€This€is€what€we€would€expect€because€there€are€moreĻaspects€of€the€economic€commodity€that€are€different€and€unaccounted€for€in€the€models€forĻWTP.€€Examination€of€the€œmodelššs›€sensitivity€to€reasonable€variations€in€the€sampleĻcomposition,€hypothesized€to€reflect€variations€in€the€object€of€choice,€provides€our€secondĻgauge€of€the€plausibility€of€our€estimated€responsiveness€of€WTP€to€visibility€changes.€ĻOverall,€the€composite€results€of€our€meta€analysis€for€the€case€of€visibility€changes€do€seemĻto€be€responsive€to€the€œNOAA›€€panel's€call€for€both€statistical€significance€and€economicĻplausibility€in€the€scope€test.Š ź*:&* М‡›Ģņ ņIV.ą  ąņņImplicationsóóó óŠ ‚Ņ Šą  ąRecent€tests€of€economic€models€„„€whether€they€relate€to€choices€over€lotteries€(inĻlaboratory€experiments,€see€Harrison€[17]),€or€they€consider€monetary€measures€for€WTPĻbased€on€stated€preferences€(i.e.,€contingent€valuation),€have€highlighted€the€difficulty€inĻdeveloping€unambiguous€tests€for€consistent€(or€rational)€economic€behavior.€€In€each€case,Ļrecognition€of€the€set€of€maintained€assumptions€involved€should€condition€the€interpretationsĻthat€can€be€drawn€from€seemingly€direct€tests.€€We€have€proposed€an€alternative€criterion€thatĻacknowledges€the€potential€to€use€past€empirical€literature€to€quantitatively€evaluate€theĻevidence€for€or€against€a€hypothesis.€€This€practice€is€common€in€medical€research,€where€aĻcomposite€judgment€must€be€developed€from€an€sequence€of€trials,€each€testing€for€aĻtreatment€effect€under€different€conditions.€€Moreover,€such€composite€studies€often€serveĻsome€regulatory€action.Ģą  ąOur€analysis€suggests€that€it€may€be€possible€to€use€a€meta€summary€of€CV€studies€toĻmeet€a€comparable€objective€in€non„market€valuation.€€Of€course,€the€crucial€issue€in€applyingĻthe€method€is€a€common€metric€to€measure€the€commodity.€€Research€on€the€economic€valueĻof€changes€in€the€visible€range€meets€this€need,€and€is€one€of€the€simplest€cases€one€couldĻconsider.€€In€the€case€of€visibility€changes,€reasonably€close€links€have€been€developedĻbetween€the€technical€measurements€of€changes€in€the€emissions€of€air€pollutants€related€toĻvisibility,€the€changes€in€visible€range,€and€the€photographs€displaying€the€visibilityĻconditions.€€As€the€complexity€of€changes€in€environmental€resources€increases,€it€is€not€clearĻthe€same€strategy€can€be€used.€€Nonetheless,€for€this€case€our€results€indicate€clearŠ ź*:&* Šresponsiveness€of€CV€based€estimates€for€WTP€to€scope.€€Because€these€findings€indicate€thatĻsummary€models€also€fit€the€set€of€data€reasonably€well,€they€are€suggestive€of€a€consistentĻeconomic€relationship€between€WTP€and€the€proportionate€change€in€visible€range.Š  T¤ ŠÓ  ÓFOOTNOTESĢÓ¬ŽÓĢ1.ą  ąŌ‡XXXXŌThanks€are€due€Richard€Carson,€Laurie€Chestnut,€Ron€Cummings,€Peter€Diamond,€BobŠ T¤ ŠRowe€and€three€anonymous€referees€for€most€helpful€comments€on€earlier€drafts€and€to€PaulaĻRubio,€Brent€McLamb,€and€Carla€Skuce€for€preparing€and€editing€those€earlier€drafts€of€thisĻpaper.€€Partial€support€for€this€research€was€provided€by€the€UNC€Sea€Grant€Program€underĻGrant€#R/MRS„25.Ō#†XX]XXįŽ#ŌĢ2.ą  ąŌ‡XX]XXŌFor€an€overview€of€the€development€of€contingent€valuation,€see€Hanemann€[15,€16].Ō#†XX]XXø#ŌŠ n¾  Š3.ą  ąŌ‡XX]XXŌThis€interpretation€can€be€found€in€Diamond€and€Hausman€[11,€12],€as€well€as€inŠ @ ŠDiamond€et€al.€[13].€€For€example,€they€suggest:Ģą0  ąą ° ąą  ą".€.€.€internal€consistency€tests€(particularly€adding„up€tests)€are€required€to€assess€theĻreliability€and€validity€of€such€surveys€[CV€surveys].€€When€these€tests€have€beenĻdone,€contingent€valuation€has€come€up€short.€€Contingent€valuation€proponentsĻtypically€claim€that€the€surveys€used€for€these€tests€were€not€done€well€enough.€YetĻthey€have€not€subjected€their€own€surveys€to€such€tests€.€.€.€.€€There€is€a€history€ofĻanomalous€results€in€contingent€valuation€surveys€that€seems€closely€tied€to€theĻembedding€problem.€€Although€this€problem€has€been€recognized€in€the€literature€forĻover€a€decade,€it€has€not€been€solved.€€Thus,€we€conclude€that€current€contingentĻvaluation€methods€should€not€be€used€for€damage€assessment€or€for€benefit€costĻanalysis"€€(Diamond€and€Hausman€[12]€pp.€62„63,€bracketed€phrase€added).Ō#†XX]XXo‘#ŌŠ (#(# Š4.ą  ąŌ‡XX]XXŌThe€panel€was€chaired€by€Kenneth€Arrow€and€Robert€Solow.€€The€other€panelistsŠ )h$( Šincluded€Edward€Leamer,€Paul€Portney,€Roy€Radner,€and€Howard€Schuman.€€See€PortneyŠ ź*:&* Š[23]€for€a€discussion€of€the€NOAA€Panel's€assessment.€€The€full€text€of€the€NOAA€panelĻreport€is€in€the€ņņFederal€Registeróó€Vol.€58,€No.€10,€January€15,€1993,€pp.€4601„4614.Ō#†XX]XXĒ•#ŌŠ ‚Ņ Š5.ą  ąŌ‡XX]XXŌThis€general€point€has€been€carefully€illustrated€by€Harrison€[17]€in€his€critique€ofŠ XØ Šexperiments€testing€the€expected€utility€hypothesis.Ō#†XX]XX„—#ŌĢ6.ą  ąŌ‡XX]XXŌFor€instance,€contrast€the€following€example€describing€the€implications€of€assumingŠ ü L Šperfect€substitution€between€a€proposed€policy€to€change€a€resource€and€the€level€of€theĻresource.€€Assume€indirect€utility€is€quasi„linear€with€a€CES€sub„function€describing€the€roleĻof€the€environmental€resource€(e.g.€V(I,€b)€=€I€+€šš€[Bņņsóó„€bņņsóó]ņņ1/sóó€where€I€=€income,€B€=€theŠ rĀ  Špopulation€of€birds,€and€b€=€the€number€of€birds€lost€without€policy€intervention).€€In€thisĻcase,€the€ratio€of€WTP€to€save€two€different€amounts€of€birds,€say€100,000€verses€1,000,€canĻrange€from€about€95.48€when€the€elasticity€of€substitution€is€100€(which€is€comparable€to€theĻconsidering€the€implementation€of€the€policy€to€affect€the€birds€as€a€perfect€substitute€for€theĻinitial€status€of€the€resource)€to€only€1.5€for€when€the€elasticity€of€substitution€is€3.55.€€TheĻrationale€for€considering€alternatives€to€Diamondššs€perfect€substitution€assumption€followĻfrom€early€work€by€Hirshleifer€[18]€considering€the€effects€of€the€production€function€forĻpublic€goods€on€the€incentives€for€individual€action€contributing€to€production€of€a€publicĻgood.€€More€recently,€Cornes€[8]€has€demonstrated€that€use€of€a€CES€function€as€his€socialĻcomposition€function€offers€one€way€to€generalize€the€Hirshleifer€arguments.Ō#†XX]XX”˜#ŌĢ7.ą  ąŌ‡XX]XXŌSee€Chestnut€and€Rowe€[5]€and€Balson€et€al.€[3]€for€a€summary€of€this€evidence.€Š x%Č $ ŠRowe€and€ChestnutŌ#†XX]XX ž#ŌŌ‡XX]XXŌ€Ō#†XX]XXŲž#ŌŌ‡XX]XXŌ[25],€Stewart€et€al.€[27]€and€Stewart€et€al.€[28]€provide€support€for€theĻlinkage€between€perceptible€visible€range€and€visible€range€measured€in€miles.Ō#†XX]XXŸ#ŌĢ8.ą  ąŌ‡XX]XXŌSeveral€of€the€original€studies€considered€for€this€meta€analysis€valued€visibilityŠ ī*>&* Šchanges€in€urban€areas.€€These€studies€are€not€considered€here€because€they€sought€to€evaluateĻmultiple€interrelated€components€of€air€pollution,€including€visibility€changes€and€healthĻeffects.€€Synthesizing€these€studies€requires€a€consistent€treatment€of€the€health€effects,€whichĻrequires€more€consistency€of€the€health€effect€measures€than€appears€to€be€available€from€theĻstudies.€€For€this€reason,€and€because€the€health€effects€are€separate€from€the€scope€questionĻconsidered€here,€these€studies€were€not€included€in€the€final€analysis.Ō#†XX]XX #ŌĢ9.ą  ąŌ‡XX]XXŌThis€format€(measuring€the€commodity€as€the€proportionate€change€in€visibility,€andŠ œģ  Šdistinguishing€between€eastern€and€western€sites)€is€also€used€in€the€first€statistical€summaryĻof€CV€evidence€reported€by€Chestnut€and€Rowe€[6].Ō#†XX]XXŻ¢#ŌĢ10.ą  ąŌ‡XX]XXŌIn€each€study,€the€payment€mechanism€was€either€an€increased€entrance€fee€at€the€parkŠ b Šin€question€or€a€monthly€addition€to€the€household's€electricity€bill.€€A€qualitative€variableĻidentifying€the€payment€mechanism€was€not€included€in€the€final€analysis€because€it€wasĻidentified€as€being€highly€collinear€with€the€qualitative€variable€identifying€whether€or€notĻrespondents€were€asked€the€WTP€for€a€visibility€change€that€occurs€on€their€next€visit€to€theĻpark.Ō#†XX]XX(¤#ŌĢ11.ą  ąŌ‡XX]XXŌThe€general€form€for€the€Box„Cox€transformation€is€(yņņššó󀄀1)€/€šš€€with€šš€=€0€implyingŠ žN Šthe€dependent€variable€is€equivalent€to€ln€y€and€šš€=€1€implying€the€dependent€variable€is€y.€Š Š!   ŠThe€likelihood€ratio€tests€using€the€most€detailed€model€are€given€as€follows:Ģą  ąHņņ0óó:€€šš€=€0ą ø ąą  ąš/šņņ2óóņņ(1)óó€=€87.62ą Ą ąą  ąp„value€=€.000Š t%Ä $ Šą  ąHņņ0óó:€€šš€=€1ą ø ąą  ąš/šņņ2óóņņ(1)óó€=€120.72ą Ą ąą  ąp„value€=€.000Ō#†XX]XX\¦#ŌŠ F'–"& Š12.ą  ąŌ‡XX]XXŌTo€investigate€this€issue,€we€used€the€CES€index€function€following€the€argumentsŠ )h$( Šoutlined€in€note€6€to€characterize€the€proposed€change€in€visibility€for€each€observation€in€ourŠ ź*:&* Šsample.€€The€incremental€improvement€in€visibility€was€added€to€the€initial€visibilityĻconditions€if€the€situation€was€described€as€WTP€for€an€improvement€and€subtracted€if€theĻsituation€was€paying€to€avoid€a€decline.€€A€simple€grid€search€using€the€complete€modelĻspecification€and€full€sample€with€selected€values€indicated€that€values€with€the€highest€Rņņ2óóŠ & v Šimply€large€elasticities€of€substitution€(e.g.,€š%š€=€12.5€and€š%š€=€33).Ō#†XX]XX©#ŌŠ ų H Š13.ą  ąŌ‡XX]XXŌBased€on€the€argument€that€the€WTP€function€should€initiate€at€zero€when€there€is€noŠ Ź   Šchange€in€visibility,€Chestnut€and€Rowe€[6]€used€a€model€without€an€intercept€in€theirĻanalysis€of€a€smaller€set€of€the€CV€studies.€€Thus,€we€include€a€similar€analysis€forĻcomparison€purposes.Ō#†XX]XX¬#ŌĢ14.ą  ąŌ‡XX]XXŌThe€qualitative€variable€indicating€whether€the€visibility€change€was€for€the€wholeŠ b Šregion€was€also€entered€as€an€interaction€term€with€the€proportionate€change€in€visibility€inĻeach€model.€€This€effect€was€never€statistically€significant€at€any€conventional€level€ofĻsignificance.Ō#†XX]XXŽ­#ŌĢ15.ą  ąŌ‡XX]XXŌTo€interpret€this€shift€variable€(š šregion=1šš),€it€is€important€to€recognize€how€the€metaŠ ZŖ Šanalysis€is€using€the€CV€results.€€When€studies€considered€regional€changes€in€visibility,€theĻvariable€identifying€these€studies€implies€a€lower€intercept,€all€else€equal,€for€the€log€(WTP)Ļfunction.€€We€could€easily€interpret€this€variable€as€reflecting€a€substitution€effect€betweenĻparks,€and€not€a€scope€effect€of€the€commodity€offered.€This€is€why€resolution€of€this€issueĻcannot€be€accomplished€without€further€information€on€peopleššs€understanding€of€what€wasĻasked.Ō#†XX]XX Æ#ŌĢ16.ą  ąŌ‡XX]XXŌThe€independent€variables€used€in€all€four€models€are€the€variables€indicating€š štheŠ )h$( Šproportionate€change€in€visibility,šš€š šregion,šš€š šfuture€use,šš€and€the€interaction€term€betweenŠ ź*:&* Šš šeastern€sitesšš€and€š šthe€proportionate€change€in€visibility.ššŌ#†XX]XXÆ±#ŌĢ17.ą  ąŌ‡XX]XXŌLevy€et€al.€[21]€suggested€concerns€about€CVššs€ability€to€measure€visibility€benefits€inŠ ‚Ņ Šgeneral€terms.€€Their€evaluation€was€based€on€a€simple€comparison€of€extrapolations€from€theĻChestnut„Rowe€[5]€study€that€were€used€in€an€EPA€policy€evaluation€of€the€benefits€ofĻregulations€on€the€emissions€of€the€Navajo€Generating€Station€versus€what€would€be€impliedĻby€the€results€of€Balson€et€al.ššs€[3]€pilot€study.€€While€they€were€careful€to€identify€theĻmultiple€considerations€in€interpreting€each€study,€the€authors€did€not€consider€pooling€theĻestimates.€€Our€meta€analysis€confirms€that€the€large€differences€in€the€benefit€estimates€seemĻto€arise€from€an€economic€characteristic€of€the€situations€described€by€Balson€et€al.Ļdistinguishing€the€amount€of€the€economic€commodity€being€offered.ą  ąŌ#†XX]XX6³#ŌŠ b ŠŠ  ä4 ŠŃ°ŃѰŃŲŲÓ  Óņ ņŪŪŪŪŪŪŪŪÜÜÜÜÜÜÜÜŪŪÜÜŪŪŪŪÜÜÜÜŪŪŪŪŪŪŃ7€XXdéXXdé7ŃÓÓTable€I:€Summary€of€CV€Studies€for€Visibility€at€National€Parksņņó óņņaóóóóŠ  ŠŌ*hƒ dhŌŌ,¹ōd"ŌŌ,¬Ōd"ŌŌ,¬Ōd"ŌŌ,ˆŌd"ŌŌ,ßŌ”"ŌŌ+  ŌŠ 1€" ńé €‘€‚Ø‚ƒƒ„„1Šņ ņAuthorsó óŠ 4€%™‘ „„……4Šņ ņSummary€Statisticsó óņņņņbóóóóŠ 3€$™‘ „„3Šņ ņLocationó óŠ 4€!™‘ „„4Šņ ņType€of€Surveyó óŠ N€=!™‘ Ø ‘  €q€‚ˆ‚ƒƒ„„……NŠĢĢRowe,Ļd'Arge,€andĻBrookshireĻ[1980]€Š A€2'— 33 „„AŠOŠ 0€!  „„0Š6Š 6€%  „„……6ŠĢĢĢNavahoĢRecreationĢAreaŠ P€=.ĻĒ " „„……PŠĢĢĢIn€person€interviewsĻadministered€toĻhouseholds€in€areaŠ VL9ĻĒ " ˆ q  €q€‚ˆ‚ƒƒ……VŠŠ &€ óė „„&ŠŌ)xdŌŠ 0€!{s „„0Š$9.27Š &{s ……&ŠŠ  óė ……ŠŠ 4*óė ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ d\ „„&ŠŌ)xlŌŠ 0€!ģä „„0Š$8.64Š &ģä ……&ŠŠ  d\ ……ŠŠ 4*d\ ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ ÕĶ  „„&ŠqŠ 0€!]U  „„0Š$6.83„$10.82Š &]U  ……&ŠŠ  ÕĶ  ……ŠŠ 4*ÕĶ  ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ F >  „„&ŠŌ)vdŌŠ 0€!Ī Ę  „„0Š0.50Š &Ī Ę  ……&ŠŠ  F >  ……ŠŠ D€3F >  ˆ q  €‘€‚Ø‚ƒƒ„„……DŠĢMacFarland,ĢMalm,€andĢMolenarĢ[1983]Š A€2'ū    „„AŠOŠ 0€!_ W  „„0Š8Š 6€%_ W  „„……6ŠĢĢĢGrand€Canyon€andĻMesa€VerdeĻNational€ParksŠ P€=.$"   „„……PŠĢĢĢIn€person€interviewsĻadministered€toĢvisitors€to€the€areaŠ VL9$"   Ø ‘  €q€‚ˆ‚ƒƒ……VŠŠ &€ H @  „„&ŠŌ)xdŌŠ 0€!Š Č  „„0Š$2.75Š &Š Č  ……&ŠŠ  H @  ……ŠŠ 4*H @  ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ ¹ ±  „„&ŠŌ)xlŌŠ 0€!A9  „„0Š$2.68Š &A9  ……&ŠŠ  ¹ ±  ……ŠŠ 4*¹ ±  ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ *" „„&ŠqŠ 0€!²Ŗ „„0Š$1.69„$3.73Š &²Ŗ ……&ŠŠ  *" ……ŠŠ 4**" ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ ›“ „„&ŠŌ)vdŌŠ 0€!# „„0Š1.18Š &# ……&ŠŠ  ›“ ……ŠŠ D€3›“ ˆ q  €‘€‚Ø‚ƒƒ„„……DŠĢSchulze,ĢBrookshire,ĢWalther,Ģand€KelleyĢ[1983]Š A€2'A999  „„AŠOŠ 0€!“¬ „„0Š20Š 6€%“¬ „„……6ŠĢĢĢGrand€Canyon,ĻMesa€Verde,€andĻZion€National€ParksŠ P€=.yq"    „„……PŠĢIn€person€interviewsĻadministered€toĢhouseholds€in€ĻAlbuquerque,€LosĻAngeles,€Denver,Ļand€ChicagoŠ VL9:2"00!  Ø ‘  €q€‚ˆ‚ƒƒ……VŠŠ &€ • „„&ŠŌ)xdŌŠ 0€!% „„0Š$8.50Š &% ……&ŠŠ  • ……ŠŠ 4*• ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€  „„&ŠŌ)xlŌŠ 0€!–Ž „„0Š$7.00Š &–Ž ……&ŠŠ   ……ŠŠ 4* ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ w „„&ŠqŠ 0€!’ „„0Š$4.42„11.67Š &’ ……&ŠŠ  w ……ŠŠ 4*w ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ šč „„&ŠŌ)vdŌŠ 0€!xp  „„0Š0.79Š &xp  ……&ŠŠ  šč ……ŠŠ D€3šč ˆ q  €‘€‚Ø‚ƒƒ„„……DŠĢĢChestnut,ĢRoweĻ[1990a]Š A€2'­„&  6 „„AŠOŠ 0€! " „„0Š72Š 6€% " „„……6ŠĢĢGrand€Canyon,ĻYosemite,€andĻShenandoah,ĢNational€ParksŠ P€=.ĪĘ'"  9 „„……PŠĢMail€with€telephoneĻfollow„up€surveyĻadministered€toĢhouseholds€inĻArizona,€Virginia,ĻCalifornia,€NewĻYork,€and€MissouriŠ VL9PH)"VV: Ø ‘  €q€‚ˆ‚ƒƒ……VŠŠ &€ ņź# „„&ŠŌ)xdŌŠ 0€!zr$ „„0Š$4.35Š &zr$ ……&ŠŠ  ņź# ……ŠŠ 4*ņź# ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ c[% „„&ŠŌ)xlŌŠ 0€!ėć& „„0Š$4.20Š &ėć& ……&ŠŠ  c[% ……ŠŠ 4*c[% ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ ŌĢ' „„&ŠqŠ 0€!\T( „„0Š$3.15„$5.48Š &\T( ……&ŠŠ  ŌĢ' ……ŠŠ 4*ŌĢ' ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ E=) „„&ŠŌ)vdŌŠ 0€!ĶÅ* „„0Š0.62Š &ĶÅ* ……&ŠŠ  E=) ……ŠŠ D€3E=) ˆ q  €‘€‚Ø‚ƒƒ„„……DŠĢĢBalson,Ģ€et.€al.Ļ[1990]Š A€2'%ś#0  O „„AŠOŠ 0€!^!V , „„0Š10Š 6€%^!V , „„……6ŠĢĢĢGrand€CanyonĻNational€ParkŠ P€=.b$Z#0"ććR „„……PŠĢIn€person€interviewsĻadministered€throughĻa€pilot€studyĻconducted€in€St.ĻLouis€and€San€DiegoĻCountiesŠ VL9ä%Ü$2"00S Ø ‘  €q€‚ˆ‚ƒƒ……VŠŠ &€ G"?!- „„&ŠŌ)xdŌŠ 0€!Ļ"Ē!. „„0Š$0.46Š &Ļ"Ē!. ……&ŠŠ  G"?!- ……ŠŠ 4*G"?!- ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ ø#°"/ „„&ŠŌ)xlŌŠ 0€!@$8#0 „„0Š$0.12Š &@$8#0 ……&ŠŠ  ø#°"/ ……ŠŠ 4*ø#°"/ ˆ q  €q€‚ˆ‚ƒƒ……4ŠŠ &€ )%!$1 „„&ŠqŠ 0€!±%©$2 „„0Š$0.007€„€$0.97Š &±%©$2 ……&ŠŠ  )%!$1 ……ŠŠ 0&)%!$1 ˆ q  €”€‚ˆ‚……0ŠŠ (€ š&’%3 „„(ŠŌ)vdŌŠ 2€!"'&4 „„2Š0.955Š &"'&4 ……&ŠŠ  š&’%3 ……ŠŠ'š&’%3 ˆ0 ”   'ŠĢņņņņaóóóó€€A€detailed€description€of€these€studies€is€available€from€the€authors€upon€request.Š $)(7 Šņņņņbóóóó€€O€=€number€of€observations€taken€from€the€study;€Ō)xdŌ€=€mean€of€the€mean€willingness€to€payŠ  *)8 Šper€month€used€for€the€meta€analysis€in€1990€dollars;€Ō)xlŌ€=€median€value€of€willingness€to€pay;Ļq€=€interquartile€range€of€mean€willingness€to€pay;€Ō)vdŌ€=€mean€proportionate€change€in€visibility.Š  ß+×*: ŠŃhŃŃhŃŃŃŃTR½“Ų'33Ų'Ų'3Letter (Landscape)’3Ų'TŃŅG°ŅŅG°ŅŌ‡ō ō]XXŌą@z 7Čąņ ņTable€II:€€Summary€of€WTP€Estimates€for€Improved€Visibility€at€National€Parks:€€Semilog€Modelsņņņņaóó€óóó óˆŠ h ŠŌ*Šƒ dd¹ōd"¬Ōd"¬Ōd"ˆŌd"ßŌ”"ŠŌŌ,š”d ŌŌ,Ątd ŌŌ,8tdEŌŌ,ĄtdEŌŌ,ĄtdEŌŌ,ĄtdEŌŌ,8tdEŌŌ,ŲtdEŌŌ,8tdEŌŌ,Ąt”EŌŌ+  ŌŠ 7€& 1É €P€‚ˆ‚ƒƒ„„……7ŠĢIndependent€VariablesŠ G€6'  h „„……GŠMeanĻValuesŠ I€6'   „„……IŠModelsŠ H>+¹Q" ˆ P  €P€‚ˆ‚ƒƒ……HŠŠ   ……ŠŠ &€  „„&Š(1aˆ)Š 0€! ” „„0Š(1bˆ)Š 0€! ” „„0Š(1cˆ)Š 0€! ” „„0Š(2ˆ)Š 0€! ” „„0Š(3ˆ)Š 0€! ” „„0Š(4ˆ)Š 0€! ” „„0Š(5ˆ)Š 4€! ” „„4Š(6ˆ)Š H€9! ” ˆ P  €€‚ˆ‚ƒƒ„„HŠInterceptŠ 0€!Yń „„0ŠŠ 0€!Yń „„0Š1ˆ.326Ģ(5ˆ.67)Š 0€!!¹ „„0Š1ˆ.911Ģ(4ˆ.71)Š 0€!!¹ „„0Š1ˆ.326Ģ(3ˆ.35)Š 0€!!¹ „„0Š0ˆ.916Ģ(4ˆ.31)Š 0€!!¹ „„0Š1ˆ.091Ģ(13ˆ.44)Š 0€!!¹ „„0Š0ˆ.919Ģ(4ˆ.41)Š 0€!!¹ „„0Š1ˆ.102Ģ(9ˆ.65)Š 4€!!¹ „„4Š1ˆ.131Ģ(20ˆ.55)Š H€9!!¹ ˆ   €€‚ˆ‚ƒƒ„„HŠProportionate€changeĻin€visibilityņņņņbóóóóŠ 0€!9Ń  „„0Š0.709Š 0€!q   „„0Š1ˆ.127Ģ(2ˆ.41)Š 0€!9Ń  „„0Š1ˆ.213Ģ(6ˆ.11)Š 0€!9Ń  „„0Ј.806Ģ(6ˆ.72)Š 0€!9Ń  „„0Š1ˆ.127Ģ(2ˆ.39)Š 0€!9Ń  „„0Š0ˆ.575Ģ(21ˆ.824)Š 0€!9Ń  „„0Š1ˆ.112Ģ(2ˆ.30)Š 0€!9Ń  „„0Š0ˆ.555Ģ(11ˆ.20)Š 4€!9Ń  „„4Š0ˆ.543Ģ(18ˆ.36)Š H€9!9Ń  ˆ   €ą€‚ˆ‚ƒƒ„„HŠEastern€Park*Ļproportionate€changeĻin€visibilityŠ 0€! ±  „„0Š0.156ņņņņcóóóóŠ 0€!‰ !  „„0Š„0ˆ.814Ģ(„2ˆ.52)Š 0€!Q é  „„0Š„ˆ.892Ģ(„4ˆ.46)Š 0€!Q é  „„0Š„ˆ.583Ģ(„3ˆ.06)Š 0€!Q é  „„0Š„0ˆ.843Ģ(„2ˆ.57)Š 0€!Q é  „„0Š„0ˆ.420Ģ(14ˆ.10)Š 0€!Q é  „„0ЈР0€!‰ !  „„0ЈР4€!‰ !  „„4ЈРH€9!‰ !  ˆ ą  €€‚ˆ‚ƒƒ„„HŠDecline€in€visibilityĻ(=1)Š 0€!1 É  „„0Š0.304Š 0€!i   „„0Š0ˆ.408Ģ(1ˆ.91)Š 0€!1 É  „„0Ј.607Ģ(3ˆ.40)Š 0€!1 É  „„0Ј.482Ģ(2ˆ.26)Š 0€!1 É  „„0Š0ˆ.496Ģ(2ˆ.18)Š 0€!1 É  „„0Š0ˆ.416Ģ(1ˆ.33)Š 0€!1 É  „„0Š0ˆ.478Ģ(1ˆ.53)Š 0€!1 É  „„0Š0ˆ.448Ģ(1ˆ.23)Š 4€!1 É  „„4Š0ˆ.447Ģ(1ˆ.23)Š H€9!1 É  ˆ   €€‚ˆ‚ƒƒ„„HŠAsked€region€(=1)Š 0€!  „„0Š0.661Š 0€!  „„0Š„0ˆ.468Ģ(„1ˆ.86)Š 0€!Iį  „„0Š„ˆ.744Ģ(„1ˆ.96)Š 0€!Iį  „„0Š„ˆ.306Ģ(„0ˆ.84)Š 0€!Iį  „„0ЈР0€!  „„0Š0ˆ.026Ģ(0ˆ.49)Š 0€!Iį  „„0ЈР0€!  „„0Š0ˆ.028Ģ(0ˆ.424)Š 4€!Iį  „„4ЈРH€9!  ˆ   €€‚ˆ‚ƒƒ„„HŠIterative€bidding€(=1)Š 0€!™1 „„0Š0.052Š 0€!™1 „„0Š0ˆ.116Ģ(0ˆ.76)Š 0€!ał „„0Ј.321Ģ(0ˆ.67)Š 0€!ał „„0Ј.116Ģ(0ˆ.23)Š 0€!ał „„0ЈР0€!™1 „„0Š0ˆ.770Ģ(2ˆ.79)Š 0€!ał „„0Š0ˆ.571Ģ(1ˆ.85)Š 0€!ał „„0Š0ˆ.732Ģ(2ˆ.378)Š 4€!ał „„4Š0ˆ.716Ģ(2ˆ.14)Š H€9!ał ˆ   €€‚ˆ‚ƒƒ„„HŠDirect€question€(=1)Š 0€!±I „„0Š0.078Š 0€!±I „„0Š„4ˆ.616Ģ(„14ˆ.42)Š 0€!y „„0Š„4ˆ.252Ģ(„9ˆ.06)Š 0€!y „„0Š„4ˆ.330Ģ(„9ˆ.34)Š 0€!y „„0Š„4ˆ.206Ģ(„20ˆ.37)Š 0€!y „„0ЈР0€!±I „„0Š„4ˆ.196Ģ(„19ˆ.92)Š 0€!y „„0ЈР4€!±I „„4ЈРH€9!±I ˆ   €€‚ˆ‚ƒƒ„„HŠPark€interview€(=1)Š 0€!Éa „„0Š0.096Š 0€!Éa „„0Š„0ˆ.621Ģ(„4ˆ.88)Š 0€!‘) „„0Š„ˆ.964Ģ(„3ˆ.31)Š 0€!‘) „„0Š„ˆ.661Ģ(„2ˆ.38)Š 0€!‘) „„0ЈР0€!Éa „„0Š„0ˆ.775Ģ(„7ˆ.41)Š 0€!‘) „„0Š„0ˆ.693Ģ(„9ˆ.08)Š 0€!‘) „„0Š„0ˆ.764Ģ(„7ˆ.303)Š 4€!‘) „„4Š0ˆ.776Ģ(„8ˆ.01)Š H€9!‘) ˆ   €€‚ˆ‚ƒƒ„„HŠSample€composed€ofĻresidents€(=1)Š 0€!©A  „„0Š0.261Š 0€!įy „„0Š0ˆ.148Ģ(1ˆ.00Š 0€!©A  „„0Ј.200Ģ(1ˆ.25)Š 0€!©A  „„0Ј.146Ģ(0ˆ.79)Š 0€!©A  „„0ЈР0€!įy „„0ЈР0€!įy „„0ЈР0€!įy „„0ЈР4€!įy „„4ЈРH€9!įy ˆ   €ą€‚ˆ‚ƒƒ„„HŠQuestion€framed€forĻvisibility€change€atĻrevisit€to€site€(=1)Š 0€!‰!$ „„0Š0.174Š 0€!ł‘" „„0Š„1ˆ.041Ģ(„2ˆ.14)Š 0€!ĮY# „„0Š„1ˆ.212Ģ(„2ˆ.65)Š 0€!ĮY# „„0Š„ˆ.674Ģ(„1ˆ.43)Š 0€!ĮY# „„0Š„0ˆ.849Ģ(„1ˆ.87)Š 0€!ĮY# „„0ЈР0€!ł‘" „„0Š„0ˆ.558Ģ(„1ˆ.64)Š 0€!ĮY# „„0ЈР4€!ł‘" „„4ЈРF€5!ł‘" ˆ ą  €Ų€‚ˆ‚„„FŠ€€€€€€€€€€€€€€€€€€€€n:€€ĢRņņ2:Š ”9' ŠóóBox„Cox€(šš)Š 9€(i(3 „„9Š.115ĢĢĢŠ 2€!1É) „„2Š115ˆĢ0ˆ.7255̈̈Р2€!1É) „„2Š115ˆĢˆ.711ņņdóóŠ ”9' Ј.291̈Р2€!1É) „„2Š115ˆĢˆ.738ņņdŠ ”9' ŠóóˆŠ 2€!i( „„2Ј.115Ģ0ˆ.703Š 2€!”9' „„2Š106ˆĢ0ˆ.408Š 2€!”9' „„2Š97ˆĢ0ˆ.725Š 2€!”9' „„2Š88ˆĢ0ˆ.428Š 4€!”9' „„4Š88ˆĢ0ˆ.428Š1'%”9' ˆ0 Ų  1ŠņņņņÓ€ÓaóóóóNumbers€in€parentheses€for€models€(1a)€through€(6)€are€ratios€of€coefficients€to€Huber„White€consistent€variance€estimates.€€With€models€(1b)€and€(1c)€they€correspond€to€theŠ )Į+ Šratio€of€the€estimated€coefficient€to€the€relevant€standard€error€in€each€case.ņņŠ U ķ- ŠņņbóóóóThis€variable€is€scaled€by€100€for€these€estimates€so€it€is€the€proportionate€change€in€the€miles€of€visibility€in€absolute€magnitude€„„€[revised€„€initial]/initial.ņņŠ ! / ŠņņcóóóóFraction€of€observations€for€visibility€changes€at€eastern€sites.ņņŠ ­"E!1 ŠņņdóóóóThe€Rņņ2óó€statistics€computed€for€these€models€are€not€comparable€to€OLS€because€they€are€not€measured€as€the€same€dependent€variables.€€Ō#†XX?ōōPŚ#ŌŠ Ł#q"3 ЇÓ€œūÓŃŃŃTRMt3Ų'½“Ų'33Ų'Letter (Portrait)Ų'3Letter (Landscape)TŃѰhŃҰGŅҰGŅņ ņTable€III:€€Analysis€of€the€Specification€Sensitivity€Analysis€for€WTP€Modelsó óŠ Č ŠŌ*߃ d dš”d Ątd 8tdEĄtdEĄtdEĄtdE8tdEŲtdE8tdEĄt”EßŌŌ, Äd ŌŌ,Q¤d`ŌŌ,Ą”d`ŌŌ,Q””`ŌŌ+  ŌŠ 7€& ±é €²€‚Ø‚ƒƒ„„……7ŠĢĢIndependent€VariablesŠ I€6'+c²² „„……IŠą@mc—ąModelsˆŠ A7$Y‘ Ø ²  €Z€‚ˆ‚ƒƒ……AŠŠ &€ c› „„&Šą@ĄIąSemilogˆĢą@FąNo€InterceptˆŠ .€!Ō  „„.ŠLinearĢWith€InterceptŠ ;€(Ō " „„;Šą@ł|’ąLinearˆĢą@y’ąNo€InterceptˆŠ H€9!Ō  ˆ Z  €Z€‚ˆ‚ƒƒ„„HŠInterceptŠ 0€!E } „„0Š„ˆŠ 5€(E }U „„5Š5ˆ.359Ģ(2ˆ.68)Š ;€(. fU „„;Š„ˆĢˆŠ O€@(. fU ˆ Z  €J€‚x‚ƒƒ„„OŠProportionate€change€inĻvisibilityŠ 0€!x °  „„0Š1ˆ.056Ģ(5ˆ.09)Š 2(x ° U „„2Š2ˆ.946Ģ(4ˆ.16)Š .€x ° U „„.Š3ˆ.490Ģ(4ˆ.25)Š O€@(x ° U x J  €J€‚x‚ƒƒ„„OŠEastern€Park*€ProportionateĢChange€in€VisibilityŠ 0€!Āś  „„0Š„0ˆ.771Ģ(„4ˆ.62)Š 2(Āś U „„2Š„2ˆ.307Ģ(„3ˆ.07)Š .€Āś U „„.Š„2ˆ.499Ģ(„3ˆ.65)Š O€@(Āś U x J  €J€‚x‚ƒƒ„„OŠDecline€(=1)Š 0€!#[  „„0Š0ˆ.671Ģ(1ˆ.47)Š 2( D U „„2Š2ˆ.364Ģ(1ˆ.13)Š .€ D U „„.Š3ˆ.474Ģ(1ˆ.28)Š O€@( D U x J  €J€‚x‚ƒƒ„„OŠAsked€Region€(=1)Š 0€!m„ „„0Š0ˆ.763Ģ(8ˆ.90)Š 2(VŽU „„2Š„2ˆ.376Ģ(„1ˆ.07)Š .€VŽU „„.Š2ˆ.260Ģ(7ˆ.02)Š O€@(VŽU x J  €J€‚x‚ƒƒ„„OŠIterative€Bidding€(=1)Š 0€!·ļ „„0Š1ˆ.155Ģ(3ˆ.16)Š 2( ŲU „„2Š1ˆ.568Ģ(4ˆ.29)Š .€ ŲU „„.Š5ˆ.331Ģ(2ˆ.29)Š O€@( ŲU x J  €J€‚x‚ƒƒ„„OŠDirect€Question€(=1)Š 0€!9 „„0Š„3ˆ.345Ģ(„16ˆ.11)Š 2(ź"U „„2Š„7ˆ.716Ģ(„4ˆ.59)Š .€ź"U „„.Š„2ˆ.877Ģ(„3ˆ.67)Š O€@(ź"U x J  €J€‚x‚ƒƒ„„OŠPark€Interview€(=1)Š 0€!Kƒ „„0Š„0ˆ.579Ģ(„4ˆ.81)Š 2(4lU „„2Š„3ˆ.354Ģ(„7ˆ.92)Š .€4lU „„.Š„3ˆ.230Ģ(„8ˆ.86)Š O€@(4lU x J  €J€‚x‚ƒƒ„„OŠSample€Composed€of€residentsĻ(=1)Š 0€!~¶  „„0Š0ˆ.226Ģ(2ˆ.42)Š 2(~¶ U „„2Š0ˆ.358Ģ(0ˆ.42)Š .€~¶ U „„.Š0ˆ.662Ģ(0ˆ.92)Š O€@(~¶ U x J  €J€‚x‚ƒƒ„„OŠQuestion€framed€for€visibilityĻchange€at€revisit€to€site€(=1)Š 0€!Č# „„0Š0ˆ.334Ģ(1ˆ.25)Š 2(Č#U „„2Š„2ˆ.780Ģ(„1ˆ.42)Š .€Č#U „„.Š1ˆ.848Ģ(2ˆ.09)Š M€<(Č#U x J  €Š€‚ˆ‚„„MŠn:ĢRņņ2óó:Š 9€(""Z&3 „„9Š115Ģ0.821Š 9€(""Z&" „„9Š116Ģ0.425Š ;€(""Z&" „„;Š116Ģ0.780Š8.,""Z&" ˆ0 Š  8ŠŃh°ŃŃŃŅG°ŅŅG°ŅŃČhŃņ ņń`ńŠ ;#s( Šń`ńŠ ;#s( ŠŃTRw3Ų' LetterMt3Ų'3Ų'Letter (Portrait)3Ų'Letter (Portrait)TŃTable€IV:€Sample€Composition€and€WTP„„Visible€Range€ResponsesĢó óĢŌ*yƒ d d Äd Q¤d`Ą”d`Q””`yŌŌ,,d4 ŌŌ,ŖDD"ŌŌ,éD4"ŌŌ,ÜD4"ŌŌ,‘Dd"ŌŌ+  ŌŠ 7€& šŅ €Æ€‚~‚ƒ%ƒ„„……7Šņ ņĢSampleó óŠ E€6'9;;ü) „„……EŠņ ņVisibility€Coefficientsņņņņaóóó óóóŠ 9€(P „„……9Šņ ņĢInitialĢConditionsņņbóóó óŠ I€6'ź"ÆÆk „„……IŠņ ņĢChangeņņņņbóóó óóóŠ OE29;;l ~% Æ  €N€‚^‚ƒ5ƒ……OŠŠ *€ I „„*Šņ ņProportionateŠ §ß ŠChangeó óŠ 2€! Č „„2Šņ ņEast€*Š §ß ŠProportionateĻChangeó óŠ &y ± ……&ŠŠ  I ……ŠŠ >€/I ^5 N  €>€‚^‚ƒ%ƒ„„>ŠĢFull€sampleĢ(n€=€115)Š 0€!Ē ’  „„0ŠĢ.685Ģ(2.34)Š 0€!Ē ’  „„0ŠĢ„.353Ģ(„2.08)Š 0€!Ē ’  „„0ŠĢ76.1Ģ[56€„€96]Š 4€!Ē ’  „„4ŠĢ43.8Ģ[25€„€50]Š H€9!Ē ’  ^% >  €'€‚^‚ƒ%ƒ„„HŠĢOmitting€BalsonĢ€€€et.€al.Ģ(n€=€106)Š 0€!ī&  „„0ŠĢ.660Ģ(34.97)Š 7€(= " „„7ŠĢ„.498Ģ(„16.72)Š 7€(= " „„7ŠĢ87.3Ģ[70€„€96]Š 4€!=  „„4ŠĢ48.5Ģ[25€„€59]Š H€9!=  ^% '  €>€‚^‚ƒ%ƒ„„HŠĢOmitting€Eastern€SitesĢ(n€=€97)Š 0€!,d „„0ŠĢ.661Ģ(2.02)Š 0€!,d „„0ŠĢ„„„„Š 0€!C{ „„0ŠĢ75.4Ģ[56€„€96]Š 4€!,d „„4ŠĢ41.6Ģ[25€„€50]Š J€9!,d ^% >  €@€‚^‚ƒUƒ„„JŠĢOmitting€EasternĢ€€€Sites€and€BalsonĢ€€€et.€al.Ģ(n€=€88)Š 2€!<t „„2ŠĢ.636Ģ(54.01)Š 2€!j¢ „„2ŠĢ„„„„Š 2€!¹ „„2ŠĢ87.7Ģ[70€„€96]Š 4€!j¢ „„4ŠĢ46.3Ģ[25€„€59]Š1'%j¢ ^U @   1ŠĢņņņņÓÓaóóóóThe€numbers€in€parentheses€are€ratios€of€the€estimated€coefficient€to€the€Huber€standard€errors€for€the€semi„logŠ c› Šmodels€relating€WTP€to€visibility€conditions,€region€and€change€to€be€relevant€to€a€future€visit€with€an€intercept.ĢņņņņbóóóóThe€first€statistic€corresponds€to€the€mean€in€miles€and€the€values€in€the€brackets€the€inter„quartile€range€forŠ  ?! Šthe€relevant€sample.ņ ņŠ Ł!# Šó óÓ½ÓŠ  «#ć% ŠŲŲŲŲŲŲńańҰGŅҰGŅѰhŃѰČŃńańŃ  ŃŌ€&&]XXŌÜ ÜŌ€"&& &&ŌŌ‡"XX &"&ŌŪŪŪ ŪÜ!ÜÜ ÜŪŪŃ€#ĶŃŃ7€ÓžXXdéXXdé7Ńą@Jģąņ ņReferencesó óˆŠ ° ŠĢą0  ąą ° ą1.€€K.€Arrow,€E.€Leamer,€H.€Schuman€and€R.€Solow,€Comments€on€NOAA€Proposed€Rule€onĻNatural€Resource€Damage€Assessment€69€(1994)Ō#†"&&]X"Xˆ#ŌŌ‡"XX &"&Ō.Šk»(#(# ŠĢą0  ąą ° ą2.€€K.€Arrow,€R.€Solow,€P.€Portney,€E.€Leamer,€R.€Radner€and€H.€Schuman,€ņņReport€of€NOAAŠ =  ŠPanel€on€Contingent€Valuationóó€58€Fed€Reg€46€(1993)Ō#†"&&]X"X'!#ŌŌ‡"XX &"&Ō.Š( x(#(# ŠĢą0  ąą ° ą3.€€W.€E.€Balson,€R.T.€Carson,€M.B.€Conaway,€B.€Fischhoff,€W.M.€Hanemann,€A.€Hulse,€R.J.ĻKopp,€K.M.€Martin,€R.C.€Mitchell,€J.€Molenar,€S.€Presser,€and€P.A.€Ruud,€ņņDevelopmentŠ å 5  Šand€Design€of€a€Contingent€Valuation€Survey€for€Measuring€the€Public's€Value€forĻVisibility€Improvements€at€the€Grand€Canyon€National€Parkóó,€Draft€report,€Decision€Focus,Š ¹  ŠInc.,€Los€Altos,€California,€September€(1990).Š (#(# ŠĢą0  ąą ° ą4.€€G.€E.€P.€Box€and€D.€R.€Cox,€š šAn€Analysis€of€Transformations,šš€ņņJournal€of€the€RoyalŠ vĘ  ŠStatistical€Societyóó,€Series€B,€26,€211„243€(1964).Ša± (#(# ŠĢą0  ąą ° ą5.€€L.G.€Chestnut.€and€R.€D.€Rowe,€ņņPreservation€Value€for€Visibility€Protection€at€the€NationalŠ 5… ŠParksóó,€Draft€Final€Report€to€U.S.€Environmental€Protection€Agency,€RCG/Hagler€Bailley,Š  p ŠInc.€February€(1990a).Š (#(# ŠĢą0  ąą ° ą6.€€„„„„„„,"Economic€Valuation€of€Changes€in€Visibility:€A€State€of€the€Science€Assessment€forĻNAPAP"€in€ņņAcidic€Deposition:€State€of€Science€and€Technologyóó,€Report€No.€27,€ņņMethodsŠ Ę Šfor€Valuing€Acidic€Deposition€and€Air€Pollution€Effectsóó,€Washington,€D.C.,€NAPAP€Š ± Š(1990b).€€Š (#(# ŠĢą0  ąą ° ą7.€€R.G.€Cummings,€D.€S.€Brookshire,€and€W.€D.€Schulze,€ņņValuing€Environmental€Goods:€€AnŠ n¾ ŠAssessment€of€the€Contingent€Valuation€Method,€óóTotowa,€N.J.,€Rowman€and€Allanheld,Š Y© ŠPublishers€(1986).Ō#†"&&]X"XF"#ŌŌ‡"XX &"&ŌŠD”(#(# Š€Ģą0  ąą ° ą8.€€R.€Cornes,€š šDyke€Maintenance€and€Other€Stories:€Some€Neglected€Types€of€Public€Goods,ššŠ  f ŠņņQuarterly€Journal€of€Economicsóó€108,€February,€259„272€(1993).Š’ O(#(# ŠĢą0  ąą ° ą9.€€W.H.€Desvousges,€F.€R.€Johnson,€R.€W.€Dunford,€K.€J.€Boyle,€S.€P.€Hudson,€K.€N.€Wilson,Š Ó"#! ŠņņMeasuring€Nonuse€Damages€Using€Contingent€Valuation:€€An€Experimental€EvaluationŠ ¼# " Šof€Accuracyóó,€RTI€Monograph€92„1,€Research€Triangle€Park,€N.C.,€Research€TriangleŠ „$õ# ŠInstitute,€(1992).Š (#(# ŠĢą0  ąą ° ą10.€€P.€A.€Diamond,€"Testing€the€Internal€Consistency€of€Contingent€Valuation€Surveys."€ņņJournalŠ b'²"& Šof€Environmental€Economics€and€Managementóó,€(in€press).ŠM(#'(#(# ŠĢą0  ąą ° ą11.€€P.€A.€Diamond,€and€J.€A.€Hausman,€š šOn€Contingent€Valuation€Measurement€of€NonuseĻValuesšš€in€(J.€A.€Hausman,€editor)€ņņContingent€Valuation:€A€Critical€Assessmentóó,Š  +Z&* ŠAmsterdam,€North€Holland,€(1993).Šõ+E'+(#(# ŠŃ#ĶŃŲŲŲ!ŲŃ  чą0  ąą ° ą12.€€P.€A.€Diamond€and€J.€A.€Hausman,€"Contingent€Valuation:€Is€Some€Number€Better€ThanĻNo€Number?"€ņņJournal€of€Economic€Perspectivesóó€8,€45„64€(1994).€Š™é(#(# ŠńańŃ#ĶŃŲŲŲ!ŲŃ  ŃńańĢą0  ąą ° ą13.€P.€A.€DiamondŌ#†"&&]X"X)#ŌŌ‡"XX &"&Ō,€J.€A.€Hausman,€G.€K.€Leonard,€and€M.€A.€Denning,€"Does€ContingentŠ m½ ŠValuation€Measure€Preferences?€Experimental€Evidence,"€in€(J.€A.€Hausman€editor)ĻņņContingent€Valuation:€€A€Critical€Assessmentóó,€Amsterdam,€North€Holland,€(1993).Š? (#(# ŠĢńańŃ#ĶŃŲŲŲ!ŲŃ  Ńńańą0  ąą ° ą14.€€W.€M.€Hanemann,€"Willingness€to€Pay€Versus€Willingness€to€Accept:€How€Much€Can€TheyĻDiffer,"€ņņAmerican€Economic€Reviewóó€81,€June,€635„647€(1991).Šü L(#(# ŠĢą0  ąą ° ą15.€€„„„„„,"Preface:€€Notes€on€the€History€of€Environmental€Valuation€in€the€USA"€in€StaleĻNavrud,€editor,€ņņPricing€the€Environment:€€The€European€Experienceóó,€Oxford,€U.K.:€Š ¹  ŠOxford€Univ.€Press,€(1992).Š (#(# ŠĢą0  ąą ° ą16.€€„„„„„,"Valuing€the€Environment€Through€Contingent€Valuation,"€ņņJournal€of€EconomicŠ vĘ  ŠPerspectivesóó€8,€Fall,€19„44€(1994).Ša± (#(# ŠĢą0  ąą ° ą17.€€G.€Harrison,€"Expected€Utility€Theory€and€the€Experimentalists,"€working€paper€90„04,ĻUniversity€of€South€Carolina,€September,€1990.€Š (#(# ŠĢą0  ąą ° ąŪŪŪŪ18.€€J.€Hirshleifer,€š šFrom€Weakest€Link€to€Best€Shot:€The€Voluntary€Provision€of€Public€Goods,ššĻņņPublic€Choice€óó41€(3):€371„386€(1983)€and€correction€in€ņņPublic€Choiceóó€(1985).ŪŪÜÜŠŁ)(#(# ŠĢą0  ąą ° ą19.€€P.€J.€Huber,€"The€Behavior€of€Maximum€Likelihood€Estimates€Under€Non„standardĻConditions."€Proceedings€of€the€Fifth€Symposium€on€Mathematical€Statistics€andĻProbability,€Berkeley,€Calif.:€University€of€California€Press€(1967).€Š (#(# ŠĢą0  ąą ° ą20.€€P.€O.€Johansson,€ņņThe€Economic€Theory€and€Measurement€of€Environmental€Benefitsóó.Š Q” ŠCambridge:€Cambridge€University€Press,€€57„70€(1987).Š (#(# ŠĢą0  ąą ° ą21.€€D.€S.€Levy,€J.€€K.€Hammett,€N.€Duan,€T.€Downes„Le€Guin€and€D.€Friedman,€ņņConceptualŠ  ^ Šand€Statistical€Issues€in€Contingent€Valuation:€Estimating€the€Value€of€Altered€VisibilityĻin€the€Grand€Canyonóó.€€Santa€Monica,€Calif.:€Rand€(1995).Šā!2 (#(# Šą0  ąą ° ąŠ (#(# Šą0  ąą ° ą22.€€K.€MacFarland,€W.€Malm,€and€J.€Molenar,€"An€Examination€of€Methodologies€for€AssessingĻthe€Value€of€Visibility."€in€(Robert€Rowe€and€Lauraine€Chestnut€editors)ÜÜÜ܀ņņManaging€AirŠ Ÿ$ļ# ŠQuality€and€Scenic€Resources€at€National€Parks€and€Wilderness€Areasóó,€Boulder,€CO.:€Š Š%Ś $ ŠWestview€Press€(1983).Š (#(# ŠĢą0  ąą ° ą23.€€P.€R.€Portney,€"The€Contingent€Valuation€Debate:€€Why€Economists€Should€Care",€ņņJournalŠ G(—#' Šof€Economics€Perspectivesóó€8,€Fall,€3„18€(1994).Š2)‚$((#(# Šā āŠ *m%) Šą0  ąą ° ą24.€€R.€D.€Rowe,€R.€d'Arge,€and€D.€Brookshire,€"An€Experiment€on€the€Economic€Value€ofĻVisibility,"€ņņJournal€of€Environmental€Economics€and€Managementóó€7,€€January,€1„19Š ™é Š(1980).Š (#(# ŠĢą0  ąą ° ą25.€€R.€D.€Rowe,€and€L.€G.€Chestnut,€ņņThe€Value€of€Visibility:€Economic€Theory€and€ApplicationsŠ V¦ Šfor€Air€Pollution€Controlóó,€Cambridge,€Mass,€Abt€Books€(1982).ŠA ‘(#(# ŠĢą0  ąą ° ą26.€€W.€D.€Schulze,€D.€Brookshire,€E.€Walther,€and€K.€Kelly,€€ņņThe€Benefits€of€PreservingŠ  e ŠVisibility€in€the€National€Parklands€of€the€Southwestóó€VIII,€draft€report€prepared€for€theŠ  P ŠU.S.€Environmental€Protection€Agency,€Washington,€D.C.€(1983).Š (#(# ŠĢą0  ąą ° ą27.€€T.€K.€Stewart,€P.€Middleton,€and€D.€Ely,€"Judgments€of€Visual€Air€Quality:€Reliability€andĻValidity,"€ņņJournal€of€Environmental€Management€óó3,€129„145€(1983).Цö (#(# ŠĢą0  ąą ° ąą0(#(#ąą ° ą28.€€T.€K.€Stewart,€€P.€Middleton,€M.€Downton,€and€D.€Ely,€€"Judgments€of€Photographs€VersusĻField€Observations€in€Studies€of€Perception€and€Judgment€of€the€Visual€Environment,"€ĻņņJournal€of€Environmental€Psychologyóó€4,€283„302€(1984).ŠLœ(#(# ŠĢą0  ąą ° ą29.€€H.€White,€€"A€Heteroscedasticity„Consistent€Covariance€Matrix€Estimator€and€a€Direct€TestĻfor€Heteroscedasticity,"€ņņEconometricaóó€48,€May,€817„830€(1980).€Š Y(#(# ŠŌ#†"&&]X"Xn/#ŌĢņ ņńbńĢĢńbńó óā ā