ÿWPCV‰ ìΰû áÄø×=ÃL=¬>Ø»>“?`#FîƒG§qK°L­ÈL²uM 0U'N 0ã|N 0Å_O 0¿$P 0BãP 0¹%Q 0³ÞQ 0­‘R 0§>S 0¡åS 0ņT 0ÑKU 0³V 0­ÏV 0§|W 0¡#X 0›ÄX 0•_Y 0ôY 0ÂZ 0ÎC[ 0°\ 0ªÁ\ 0¤k] 0ž^ 0˜­^ 0’E_ 0Š×_ 0²a`¾a 0¾Ñc 0¾d 0>Me 0ø‹e 0âƒf 0âeg 0âGh 0â)i 0â j 0ºíj 0Ò§k 02yl 0ü«l B§m 0:Äm 0Yþm 0_Wn 0 ¶n 0‚Ào 0Bq B*Hs D/rs D+¡s 0Ìs D/Òu 0Îv 06Ïv 0w 0: x 0^Gx 0<¥x 1eáxU6FyU*|y 1u¦y 72z 0cMzͰzÔ}{ùQ}ýJ~ÇG¨€€Ú¶€A‚‚ Ñ„¥q…q…%†††††††††††††††††††††††;‡˜\\P2PPRS01V1\HP5SI-2-C237-1-P2ÈÈ,,,,ÈÈ0(ÖÃ9 Z‹6Times New Roman RegularX(üœ$¡¡ÔUSUS.,Ô&b9/Ï tWé./ïeBackup3|x#ÿU‹ÿÀÀÀ Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú3Ú  ÚóóÝ  ÝÔ_Ô€€Based€on€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.(.(3Ä($¤¤Ý ƒüœ!ÝÔUSUS.,ÔÝ  Ýà  àòòÚ  Ú0Ú  Úóó(#Ã$òòÚ  Ú0Ú  Úóó ¹Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú4Ú  ÚóóÝ  ÝÔ_Ô€€Holt€and€Sherman€[1986]€find€that€overall€product€quality€improves€under€bundling. ¾Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú5Ú  ÚóóÝ  ÝÔ_Ô€€As€in€many€instances,€there€are€exceptions.€€If€the€consumer€has€more€than€oneÐ ° Ðtelevision€set€or€if€the€set€has€picture„in„picture€capability€then€the€consumer€can€watch€moreÐ œì Ðthan€a€single€channel€at€any€given€time.€€As€a€general€rule,€however,€just€one€channel€is€watchedÐ ˆØ Ðat€a€time. ÕÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú6Ú  ÚóóÝ  ÝÔ_Ô€€The€competitive€issue€is€one€that€is€currently€being€hotly€debated€in€the€presence€of€theÐ ° Ðsignificant€inroads€being€made€by€Direct€Broadcast€Satellite€(DBS)€service€in€the€United€States.€Ð œì ÐBrown€and€Uri€[2003]€explore€this€issue€in€detail€and€conclude€that€cable€systems€still€lack€aÐ ˆØ Ðsubstantial€competitor. ±Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú10Ú  ÚóóÝ  ÝÔ_Ô€€Profit€is€just€the€difference€between€revenue€and€the€cost€of€production. ”Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú9Ú  ÚóóÝ  ÝÔ_Ô€€This€is€just€the€marginal€cost€of€production. šÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú11Ú  ÚóóÝ  ÝÔ_Ô€€Ô_ÔAlthoughÔ_Ô€there€are€10,146€cable€systems€in€the€United€States,€the€ownership€of€cableÐ ° Ðoperations€is€relatively€concentrated.€€Additionally,€there€are€ownership€ties€between€cableÐ œì Ðsystems€and€related€firms.€€For€example,€there€are€vertical€relationships€between€cable€companiesÐ ˆØ Ðand€program€suppliers,€horizontal€concentration€among€cable€companies,€and€clustered€cableÐ tÄ Ðsystems€whereby€cable€companies€consolidate€ownership€within€a€geographic€area.€€Each€ofÐ `° Ðthese€relationships€has€implications€for€the€cost€of€supplying€cable€service.€€Ð L œ Ðà  àAs€cable€system€deployment€and€the€number€of€subscribers€grew,€the€ownership€structureÐ 8 ˆ Ðof€the€industry€changed.€€Early€on,€the€cable€industry€was€characterized€by€privately€owned,€smallÐ $ t Ðsystems€scattered€throughout€the€United€States.€€The€expanding€market€for€cable€service€in€theÐ  ` Ð1980s,€however,€attracted€the€interest€of€large€media€companies€(Crandall€and€Ô_ÔFurchtgottÔ_Ô„RothÐ ü L  Ð[1996]).€€These€companies€purchased€cable€systems€throughout€the€United€States,€emerging€asÐ è 8  Ðlarge€national€cable€owners€known€as€multiple€systems€operators€(Ô_ÔMSOsÔ_Ô).€€Several€of€theseÐ Ô$  ÐÔ_ÔMSOsÔ_Ô€also€invested€in€the€development€of€cable€programming€and€thereby€establishedÐ À  Ðownership€ties€between€two€vertically€integrated€markets€„€companies€producing€and€supplyingÐ ¬ü  Ðcable€programming€and€companies€purchasing€that€programming€and€delivering€it€to€subscribers.€Ð ˜è  ÐNearly€98.7€percent€of€the€cable€systems€responding€to€the€òò2002€FCC€Annual€Cable€Price€SurveyóóÐ „Ô  Ðindicated€that€they€were€owned€by€a€Ô_ÔMSOÔ_Ô.Ð pÀ Ðà  àIn€recent€years€cable€systems€have€engaged€in€a€clustering€strategy€in€order€to€consolidateÐ \¬ Ðtheir€systems€in€and€around€specific€cities€or€regions.€€Cable€companies€can€obtain€increasedÐ H˜ Ðeconomies€of€scale€from€clustering€as€compared€to€having€noncontiguous€cable€systems€that€areÐ 4„ Ðmore€geographically€diffuse€(Ford€and€Jackson€[1997]€and€General€Accounting€Office€[1999]).€Ð  p ÐThe€clustering€approach€enables€firms€to€consolidate€facilities€for€receiving€and€transmittingÐ  \ Ðprogramming,€reduce€the€number€of€repair€crews,€have€regional€service€centers,€and€reduce€theÐ øH Ðsize€of€management€(Ô_ÔHinderyÔ_Ô€[1998]).€€Additionally,€clustering€can€serve€to€enhance€advertisingÐ ä4 Ðrevenues€(Ô_ÔHinderyÔ_Ô€[1998]€and€Park€Center€Investment€Research€[2003]).€€Just€over€81.1€percentÐ Ð   Ðof€the€cable€systems€surveyed€on€the€òò2002€FCC€Annual€Cable€Price€Surveyóó€were€part€of€a€cluster.Ð ° Ðà  àThe€various€ownership€interrelationships€that€exist€in€the€cable€industry€ostensiblyÐ œì Ðprovide€efficiencies€to€cable€companies€that€result€in€reduced€costs€of€providing€cable€service.€Ð ˆØ ÐFor€example,€relatively€larger€cable€providers€may€realize€reduced€programming€costs€and€alsoÐ tÄ Ðhave€cost€savings€in€management€and€related€overhead€functions€(Ô_ÔEmmonsÔ_Ô€and€Ô_ÔPragerÔ_Ô€[1997]). Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú12Ú  ÚóóÝ  ÝÔ_Ô€€See,€e.g.,€Beard€et€al.€[2001]€and€Mayo€and€Ô_ÔOtsukaÔ_Ô€[1991]€for€a€discussion€of€demandÐ ° Ðelasticities€for€cable€service. £Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú7Ú  ÚóóÝ  ÝÔ_Ô€€The€approach€was€pioneered€by€Waugh€[1926]€and€Court€[1936]. Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú8Ú  ÚóóÝ  ÝÔ_Ô€€Freemanððs€main€contribution€was€in€introducing€assumptions€to€simplify€the€estimationÐ ° Ðof€marginal€bid€functions.€€These€assumptions€include€assuming€that€the€quantity€of€eachÐ œì Ðavailable€characteristic€is€fixed€(permitting€ordinary€least€squares€estimation)€and€that€the€supplyÐ ˆØ Ðof€each€characteristic€is€infinitely€elastic€(Ô_ÔBartikÔ_Ô€[1987]). œÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú13Ú  ÚóóÝ  ÝÔ_Ô€€The€appropriate€adjustments€will€be€discussed€later. Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú14Ú  ÚóóÝ  ÝÔ_Ô€€Section€623(k)€was€adopted€as€Section€3(k)€of€the€1992€Cable€Act,€Pub.€L.€No.€102„Ð ° Ð385,€106€Stat.€1460,€codified€at€47€U.S.C.€ðð€543(k)òò.óó Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú15Ú  ÚóóÝ  ÝÔ_Ô€€The€1992€Cable€Act€defines€basic€cable€service€as€the€tier€of€service€that€includes€theÐ ° Ðretransmission€of€local€television€broadcast€signals€(47€U.S.C.€ðð€522)òòóó.€€Cable€programmingÐ œì Ðservice€is€defined€as€any€video€programming€other€than€video€programming€carried€on€the€basicÐ ˆØ Ðservice€tier,€and€video€programming€offered€on€a€per€channel€or€per€program€basis.€€EquipmentÐ tÄ Ðrefers€to€a€converter€box,€remote€control,€and€other€equipment€necessary€to€access€programming.€ ÇÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú2Ú  ÚóóÝ  ÝÔ_Ô€€Local€channel€refers€to€those€channels€that€carry€local€broadcast€stations€(either€throughÐ ° Ðmust„carry€requirement€or€retransmission€agreement),€public,€educational,€or€governmentÐ œì Ðprogramming,€commercial€leased€access,€and€other€programming€that€originates€locally.€€TheÐ ˆØ Ðterm€satellite€channels€refers€to€nationally„delivered€channels€on€networks€that€are,Ð tÄ Ðpredominately,€delivered€by€satellite€to€the€cable€Ô_ÔheadendÔ_Ô.€€Satellite€channels€include€majorÐ `° Ðregional€sports€networks.€ aÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú16Ú  ÚóóÝ  ÝÔ_Ô€€Cable€systems€cannot€be€subject€to€rate€regulation€in€areas€where€the€FCC€has€made€aÐ ° Ðfinding€of€ð ðeffective€competition.ðð€€A€cable€system€is€subject€to€effective€competition€when€anyÐ œì Ðone€of€the€following€conditions€is€met:€(1)€Fewer€than€30€percent€of€the€households€in€itsÐ ˆØ  Ðfranchise€area€subscribe€to€the€cable€service€of€a€cable€system.€(2)€The€franchise€area€is€served€byÐ ° Ðat€least€two€unaffiliated€multichannel€video€programming€distributors€each€of€which€offersÐ œì Ðcomparable€programming€to€at€least€50€percent€of€the€households€in€the€franchise€area€and€theÐ ˆØ Ðnumber€of€households€subscribing€to€multichannel€video€programming€other€than€the€largestÐ tÄ Ðmultichannel€video€programming€distributor€exceeds€15€percent€of€the€households€in€theÐ `° Ðfranchise€area.€(3)€A€multichannel€video€programming€distributor,€operated€by€the€franchisingÐ L œ Ðauthority€for€that€franchise€area,€offers€video€programming€to€at€least€50€percent€of€theÐ 8 ˆ Ðhouseholds€in€the€franchise€area.€(4)€A€local€exchange€carrier€or€its€affiliate€(or€any€multichannelÐ $ t Ðvideo€programming€distributor€using€the€facilities€of€such€carrier€or€its€affiliate)€offers€videoÐ  ` Ðprogramming€services€directly€to€subscribers€by€any€means€(other€than€direct„to„home€satelliteÐ ü L  Ðservices)€in€the€franchise€area€of€an€unaffiliated€cable€operator€which€is€providing€cable€serviceÐ è 8  Ðin€that€franchise€area,€but€only€if€the€video€programming€services€so€offered€in€that€area€areÐ Ô$  Ðcomparable€to€the€video€programming€services€provided€by€the€unaffiliated€cable€operator€in€thatÐ À  Ðarea.€€In€other€franchise€areas,€local€communities€have€the€authority€to€regulate€the€rates€of€theÐ ¬ü  Ðbasic€service€tier€and€equipment,€but€may€or€may€not€choose€to€exercise€that€authority. ®Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú17Ú  ÚóóÝ  ÝÔ_Ô€€The€term€service€tier€means€a€cable€service€for€which€the€operator€charges€a€separateÐ ° Ðrate.€€The€major€Ô_ÔCPSTÔ_Ô€tier€typically€meets€two€criteria:€(1)€offers€the€greatest€number€of€channelsÐ œì Ðamong€the€Ô_ÔCPSTÔ_Ô€tiers,€and€(2)€has€the€highest€number€of€subscribers€among€the€Ô_ÔCPSTÔ_Ô€tiers.€Ð ˆØ ÐCable€systems€require€subscribers€to€purchase€the€Ô_ÔBSTÔ_Ô€in€order€to€purchase€the€Ô_ÔCPSTÔ_Ô.€ ÈÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú18Ú  ÚóóÝ  ÝÔ_Ô€€Local€channel€refers€to€those€channels€that€carry€local€broadcast€stations€(eitherÐ ° Ðthrough€must„carry€requirement€or€retransmission€agreement),€public,€educational,€orÐ œì Ðgovernment€programming,€commercial€leased€access,€and€other€programming€that€originatesÐ ˆØ Ðlocally.€€The€term€satellite€channels€refers€to€nationally„delivered€channels€on€networks€that€are,Ð tÄ Ðpredominately,€delivered€by€satellite€to€the€cable€Ô_ÔheadendÔ_Ô.€€Satellite€channels€include€majorÐ `° Ðregional€sports€networks.€ ‘Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú20Ú  ÚóóÝ  ÝÔ_Ô€€Federal€Communications€Commission€[2003].ô\  `‹&Times New Roman ¹Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú24Ú  ÚóóÝ  ÝÔ_Ô€50,000€subscribers€is€the€median€size€of€the€cable€systems€included€in€the€survey. ¢Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú25Ú  ÚóóÝ  ÝÔ_Ô€€This€is€done€in€each€instance€where€the€test€is€performed. ÎÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú26Ú  ÚóóÝ  ÝÔ_Ô€€This€latter€specification€is€referred€to€as€the€Ô_ÔsemilogÔ_Ô€specification. †Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú28Ú  ÚóóÝ  ÝÔ_Ô€€One€of€the€details€worthy€of€note€is€the€impact€that€the€presence€of€Ô_ÔcollinearityÔ_Ô€amongÐ ° Ðthe€exogenous€variables€has€on€the€parameter€estimates.€€Considerable€time€was€spent€onÐ œì Ðmitigating€the€influence€of€Ô_ÔcollinearityÔ_Ô.€€This€was€done€by€a€judicious€examination€of€the€data€forÐ ˆØ Ðthe€presence€of€a€high€degree€of€correlation€between€the€exogenous€variables€individually€or€inÐ tÄ Ðconcert.€€When€Ô_ÔcollinearityÔ_Ô€appeared€to€be€an€insurmountable€problem,€the€variables€were€eitherÐ `° Ðcombined€or€one€or€more€deleted€from€the€specification.€€One€disappointing€result€was€theÐ L œ Ðgeneral€lack€of€finding€any€sort€of€statistically€significant€relationships€between€the€variablesÐ 8 ˆ Ðincluded€in€the€large€set€of€supplementary€data€compiled€from€the€2000€Census€of€PopulationÐ $ t ÐSummary€Files€provided€by€the€U.S.€Census€Bureau€and€the€analog€and€digital€cable€priceÐ  ` Ðequations.€€These€supplementary€data€were€quite€comprehensive€consisting€of€information€onÐ ü L  Ðsuch€things€as€school€enrollment,€educational€attainment,€marital€status,€disability,€languageÐ è 8  Ðspoken,€nativity,€employment€status,€commuting€patterns,€occupation,€income,€type€of€housingÐ Ô$  Ðunit,€number€of€occupants,€house€heating€fuel,€mortgage€status,€rent,€and€tenure€of€occupants.Ð À  Ð VÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú29Ú  ÚóóÝ  ÝÔ_Ô€€In€the€sense€being€used€here,€robust€estimates€are€estimates€that€are€not€extremelyÐ ° Ðsensitive€to€the€functional€specification€or€to€which€variables€are€included€or€excluded€from€theÐ œì Ðspecification.€€ äÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú19Ú  ÚóóÝ  ÝÔ_Ô€€There€is€a€possibility€that€the€number€of€DBS€subscribers€was€systematicallyÐ ° Ðoverestimated€by€the€cable€systems€surveyed.€€Based€on€the€results€of€the€survey,€for€2002€theÐ œì Ðratio€of€the€number€of€DBS€subscribers€to€cable€subscribers€is€0.35.€€Other€data€sources€reportÐ ˆØ Ðthis€ratio€to€be€somewhat€smaller€for€2002.€€Ô_ÔVeronisÔ_Ô€Ô_ÔSuhlerÔ_Ô€[2001],€for€example,€report€a€value€ofÐ tÄ Ð0.29€while€the€òòNinth€Annual€Reportóó€of€the€Federal€Communications€Commission€[2002]€gives€aÐ `° Ðvalue€of€0.27.€€The€number€of€DBS€subscribers€in€each€cable€systemððs€franchise€area€is€notÐ L œ Ðavailable.€It€is€estimated€by€the€cable€systems€on€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€AnÐ 8 ˆ Ðexact€count€was€not€required. Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú30Ú  ÚóóÝ  ÝÔ_Ô€€The€correlation€coefficients€uniformly€exceeded€0.75. ¦Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú32Ú  ÚóóÝ  ÝÔ_Ô€€GEMS€Television€is€a€women's€Spanish„language€cable€channel.€€ £Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú33Ú  ÚóóÝ  ÝÔ_Ô€€CN8€is€the€Ô_ÔComcastÔ_Ô€Network. ¨Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú39Ú  ÚóóÝ  ÝÔ_Ô€€The€Canales€Group€is€a€Spanish€„€language€programming€channel.€€*+ ÿÿ(_26Ô2ÔÔ3  Ôà..à*G+G ÿÿ(_25Ó ¨ý ÓÓ ° ÓÓ/%` ` ¸ hÀpÈ xÐ /ÓÔ2ÔÔ3  Ôà..àÓ/%` ` ¸ hÀpÈ xÐ /ÓÓ ¨ý ÓÓ ° Ó*8+8 ÿÿ(_24Ó  ÓÓ,"¸ ¸ hÀpÈ xÐ ,ÓÔ2ÔÔ3  Ôà..àÓ,"¸ ¸ hÀpÈ xÐ ,ÓÓ  Ó*5+5 ÿÿ(_23Ó ` ÓÓ) hÀpÈ xÐ )ÓÔ2ÔÔ3  Ôà..àÓ) hÀpÈ xÐ )ÓÓ ` Ó<ÿÿ:Footnote Refòòóó*2+2 ÿÿ(_22Ó ¸ ÓÓ&hhÀpÈ xÐ &ÓÔ2ÔÔ3  Ôà..àÓ&hhÀpÈ xÐ &ÓÓ ¸ Ó*/+/ ÿÿ(_21Ó  ÓÓ#ÀÀpÈ xÐ #ÓÔ2ÔÔ3  Ôà..àÓ#ÀÀpÈ xÐ #ÓÓ  Ó*,+, ÿÿ(_20Ó h ÓÓ pÈ xÐ ÓÔ2ÔÔ3  Ôà..àÓ pÈ xÐ ÓÓ h Ó*)+) ÿÿ(_19Ó À ÓÓppÈ xÐ ÓÔ2ÔÔ3  Ôà..àÓppÈ xÐ ÓÓ À Ó*&+& ÿÿ(_18Ó  ÓÓÈÈ xÐ ÓÔ2ÔÔ3  Ôà..àÓÈÈ xÐ ÓÓ  Ó*AA ÿÿ(_17Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÔ2ÔÔ3  Ô€Ó5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó*GG ÿÿ(_16Ó ¨ý ÓÓ ° ÓÓ/%` ` ¸ hÀpÈ xÐ /ÓÔ2ÔÔ3  Ô€Ó/%` ` ¸ hÀpÈ xÐ /ÓÓ ¨ý ÓÓ ° Ó*88 ÿÿ(_15Ó  ÓÓ,"¸ ¸ hÀpÈ xÐ ,ÓÔ2ÔÔ3  Ô€Ó,"¸ ¸ hÀpÈ xÐ ,ÓÓ  Ó*55 ÿÿ(_14Ó ` ÓÓ) hÀpÈ xÐ )ÓÔ2ÔÔ3  Ô€Ó) hÀpÈ xÐ )ÓÓ ` Ó*22 ÿÿ(_13Ó ¸ ÓÓ&hhÀpÈ xÐ &ÓÔ2ÔÔ3  Ô€Ó&hhÀpÈ xÐ &ÓÓ ¸ Ó*// ÿÿ(_12Ó  ÓÓ#ÀÀpÈ xÐ #ÓÔ2ÔÔ3  Ô€Ó#ÀÀpÈ xÐ #ÓÓ  Ó*,, ÿÿ(_11Ó h ÓÓ pÈ xÐ ÓÔ2ÔÔ3  Ô€Ó pÈ xÐ ÓÓ h Ó*)) ÿÿ(_10Ó À ÓÓppÈ xÐ ÓÔ2ÔÔ3  Ô€ÓppÈ xÐ ÓÓ À Ó(&& ÿÿ&_9Ó  ÓÓÈÈ xÐ ÓÔ2ÔÔ3  Ô€ÓÈÈ xÐ ÓÓ  Ó(AA ÿÿ&_8Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÔ2ÔÔ3  ÔÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó(GG ÿÿ&_7Ó ¨ý ÓÓ ° ÓÓ/%` ` ¸ hÀpÈ xÐ /ÓÔ2ÔÔ3  ÔÓ/%` ` ¸ hÀpÈ xÐ /ÓÓ ¨ý ÓÓ ° Ó(88 ÿÿ&_6Ó  ÓÓ,"¸ ¸ hÀpÈ xÐ ,ÓÔ2ÔÔ3  ÔÓ,"¸ ¸ hÀpÈ xÐ ,ÓÓ  Ó(55 ÿÿ&_5Ó ` ÓÓ) hÀpÈ xÐ )ÓÔ2ÔÔ3  ÔÓ) hÀpÈ xÐ )ÓÓ ` Ó(22 ÿÿ&_4Ó ¸ ÓÓ&hhÀpÈ xÐ &ÓÔ2ÔÔ3  ÔÓ&hhÀpÈ xÐ &ÓÓ ¸ Ó(// ÿÿ&_3Ó  ÓÓ#ÀÀpÈ xÐ #ÓÔ2ÔÔ3  ÔÓ#ÀÀpÈ xÐ #ÓÓ  Ó(,, ÿÿ&_2Ó h ÓÓ pÈ xÐ ÓÔ2ÔÔ3  ÔÓ pÈ xÐ ÓÓ h Ó()) ÿÿ&_1Ó À ÓÓppÈ xÐ ÓÔ2ÔÔ3  ÔÓppÈ xÐ ÓÓ À Ó&&& ÿÿ$_Ó  ÓÓÈÈ xÐ ÓÔ2ÔÔ3  ÔÓÈÈ xÐ ÓÓ  Ó0AAÿÿ.NormalÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó ´Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú21Ú  ÚóóÝ  ÝÔ_Ô€€A€cable€system€is€defined€as€the€area€served€by€a€single€Ô_ÔheadendÔ_Ô.€€€€A€Ô_ÔheadendÔ_Ô€is€theÐ ° Ðcontrol€center€of€a€cable€television€system,€where€incoming€signals€are€amplified,€converted,Ð œì Ðprocessed,€and€combined€into€a€common€cable€along€with€any€original€Ô_ÔcablecastingÔ_Ô,€forÐ ˆØ Ðtransmission€to€subscribers.€€A€system€operator€is€the€individual,€organization,€company,€or€otherÐ tÄ Ðentity€that€operates€a€cable€television€system.<AAÿÿ:Definition TÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó<AAÿÿ:Definition LÓ , ÓÓ5+Ü Ü` ¸ hÀpÈ xÐ 5ÓÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Ó8ÿÿ6Definitionòòóó(hhÿÿ&H1Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Óò òÔ  ÔÔ‡°°°ÔÔ‡ôôôÔó óÔ  ÔÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó(]]ÿÿ&H2Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Óò òÔ‡„„„ÔÔ‡ôôôÔó óÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó(]]ÿÿ&H3Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Óò òÔ‡¼¼¼ÔÔ‡ôôôÔó óÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó(]]ÿÿ&H4Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Óò òÔ‡XXXÔÔ‡ôôôÔó óÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó(]]ÿÿ&H5Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Óò òÔ‡ôôôÔó óÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  ÓÔ‡ôôôÔ(]]ÿÿ&H6Ó  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Óò òÔ‡ÔÔ‡ôôôÔó óÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó2DDÿÿ0AddressÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓòòóóÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó8MMÿÿ6BlockquoteÓ , ÓÓ , ÓÓ5+Ü Ü` ¸ hÀpÈ xÐ 5ÓÓ  ÓÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5Ó,ÿÿ*CITEòòóó,dlÿÿ*CODEÔ‡ôôôÔÔK‡h<þ6X9`(‹Courier NewKÔÔ‡ôôôÔÔS‡ô\  `‹&Times New RomanSÔ d4ÿÿ2Emphasisòòóó6ÿÿ4HyperlinkóóòòÔ ÿ ÔóóÔ  Ô<ÿÿ:FollowedHypeóóòòÔ €€ ÔóóÔ  Ô4goÿÿ2Keyboardò òÔ‡ôôôÔÔK‡h<þ6X9`(‹Courier NewKÔÔ‡ôôôÔÔS‡ô\  `‹&Times New RomanSÔó ó<Ÿ§ÿÿ:PreformattedÓ  ÓÓ/%° °Ïî ,KkŠ©È %#/ÓÔ‡ôôôÔÔK‡h<þ6X9`(‹Courier NewKÔÔ‡ôôôÔÔS‡ô\  `‹&Times New RomanSÔÓ/%° °Ïî ,KkŠ©È %#/ÓÓ  Ó<Ûïÿÿ:z©Bottom of Ó7€](€X7ÓÓ  ÓÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÔ‡ÔÔ?‡i%ä2¼A`‹Arial?ÔÓ  ÓÔ‡ôôôÔÔS‡ô\  `‹&Times New RomanSÔÓ7€^(€X7ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó&ÿÿ  d d\)1dxd'dxd<Ûïÿÿ:z©Top of ForÓ7€`(€X7ÓÓ  ÓÓ  ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÔ‡ÔÔ?‡i%ä2¼A`‹Arial?ÔÓ  ÓÔ‡ôôôÔÔS‡ô\  `‹&Times New RomanSÔÓ7€^(€X7ÓÓ5+° °` ¸ hÀpÈ xÐ 5ÓÓ  Ó\)2dxd0KSÿÿ.SampleÔK‡h<þ6X9`(‹Courier NewKÔÔS‡ô\  `‹&Times New RomanSÔ0ÿÿ.Strongò òó ó8dlÿÿ6TypewriterÔ‡ôôôÔÔK‡h<þ6X9`(‹Courier NewKÔÔ‡ôôôÔÔS‡ô\  `‹&Times New RomanSÔ4ÿÿ2Variableòòóó: ÿÿ8HTML MarkupÔB  ÔÔ ÿ ÔÔ  Ô2 ÿÿ0CommentÔB  Ô456789:; ÿÿ35;AGMSY_11.1.1.1.1.1.1.1.<þ6X9`(‹Courier New%ä2¼A`‹Arial3#37=CIQYag­­1.a.i.(1)(a)(i)1)a)ji)(;3£$´´Ô2#ÔÚ  Ú0Ú  Ú.Ô3  Ôà0  à ÃÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú34Ú  ÚóóÝ  ÝÔ_Ô€€NOGGIN€is€an€educational€channel€focusing€on€pre„Ô_ÔschoolersÔ_Ô. ÊÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú1Ú  ÚóóÝ  ÝÔ_Ô€€Industry€Analysis€Division,€Media€Bureau,€Federal€Communication€Commission,Ð ° ÐWashington,€DC.€€The€views€expressed€are€those€of€the€authors€and€do€not€necessarily€representÐ œì Ðthe€policies€of€the€Federal€Communications€Commission€or€the€views€of€other€FederalÐ ˆØ ÐCommunications€Commission€staff€members. ïÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú23Ú  ÚóóÝ  ÝÔ_Ô€€TRIO€broadcasts€original€stage€specials€and€popular€arts€television€from€the€UnitedÐ ° ÐKingdom,€Canada,€and€Australia. óÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú38Ú  ÚóóÝ  ÝÔ_Ô€€E!€Style€specializes€in€coverage€of€popular€entertainment€„€celebrity€interviews,€news,Ð ° Ðand€behind„the„scenes€features. ½Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú40Ú  ÚóóÝ  ÝÔ_Ô€€Ô_ÔEWTNÔ_Ô€is€Americaððs€largest€religious€cable€network. žÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú41Ú  ÚóóÝ  ÝÔ_Ô€€ZEE€TV€is€the€most€popular€satellite€channel€in€India. ÐÝ ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú37Ú  ÚóóÝ  ÝÔ_Ô€€Recall€that€the€data€on€the€number€of€DBS€subscribers€are€reported€by€the€cableÐ ° Ðsystems.€€There€appears€to€be€slight€measurement€error€in€these€data.€€This€obviously€raises€theÐ œì Ðquestion€of€the€reliability€of€the€coefficient€estimate€on€the€DBS€penetration€variable€(GreeneÐ ˆØ Ð[1997]€and€Uri€[1995]).€€€€ 7Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú36Ú  ÚóóÝ  ÝÔ_Ô€€In€the€dominant€firm„competitive€fringe€model€it€is€assumed€that€there€is€a€single€firmÐ ° Ðor€jointly€acting€group€of€firms€holding€a€dominant€position€and€several€smaller€price„takingÐ œì Ðfirms,€typically€referred€to€as€the€competitive€fringe.€€Because€of€its€position,€the€dominant€firmÐ ˆØ Ðis€modeled€as€selecting€a€price€that€the€fringe€firms€take€as€given€in€deciding€how€much€toÐ tÄ Ðsupply.€€ –Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú31Ú  ÚóóÝ  ÝÔ_Ô€€Their€impact€is€captured€in€the€constant€term. ›Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú27Ú  ÚóóÝ  ÝÔ_Ô€€The€test€statistic€is€distributed€as€a€t„statistic. Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú22Ú  ÚóóÝ  ÝÔ_Ô€€Overbuilding€occurs€when€more€than€one€cable€operator€provides€service€in€a€specificÐ ° Ðgeographic€area.€€Each€cable€operator€has€its€own€proprietary€cable€wires. Ý ƒ Ä($ÝÔUSUS.,Ôà  àòòÚ  Ú35Ú  ÚóóÝ  ÝÔ_Ô€€By€definition€a€cluster€is€a€group€of€systems€in€close€geographic€proximity€owned€byÐ ° Ðthe€same€cable€operator.€€That€is,€clustered€systems€are€owned€by€a€multiple€systems€operatorÐ œì Ð(Ô_ÔMSOÔ_Ô).€€A€Ô_ÔMSOÔ_Ô€is€a€company€that€operates€multiple€cable€systems.€€But€also,€it€is€only€theÐ ˆØ Ðlargest€operators€that€cluster.Ý ƒüœ!ÝÔUSUS.,ÔÝ  ÝÔ_ÔÌà  àÌCable€Service€and€Its€Implicit€Marginal€ValuationÐ ˆØ Ðà  àà ` àÌÌÌNoel€D.€UriÐ 8 ˆ ÐÌandÐ  ` Ðà  àÌKeith€Brown׃n×Ý ƒ #ÃÝòòÚ  Ú1Ú  ÚóóÝ  Ý×  ×Ð è 8  ÐÐ  Ô$  ÐÓ  ÓAbstractÐ ° ÐÌÓ  Óà  àThe€objective€is€to€estimate€the€implicit€marginal€value€of€channel€offerings€by€cableÐ ˆØ Ðsystems€in€the€United€States.€€Cable€service€is€an€example€of€a€bundled€commodity€and,€as€such,Ð tÄ Ðpresents€some€unique€issues€that€must€be€addressed€when€endeavoring€to€estimate€the€implicitÐ `° Ðmarginal€valuation€of€individual€channels.€€Hedonic€analysis€is€a€convenient€framework€in€whichÐ L œ Ðto€explore€the€relationship€between€the€price€of€cable€service€and€the€channels€offered€by€a€cableÐ 8 ˆ Ðsystem.€€Two€separate€hedonic€equations€are€estimated€„€one€for€the€price€of€analog€cable€serviceÐ $ t Ðand€one€for€the€price€of€digital€cable€service.€€With€regard€to€the€price€of€analog€cable€service,€theÐ  ` Ðimplicit€marginal€valuation€is€statistically€significant€and€positive€for€just€fifteen€of€106€channelÐ ü L  Ðofferings€and€negative€for€seven€channels.€€When€the€level€of€statistical€significance€is€raised€toÐ è 8  Ð10€percent,€an€additional€seven€channels€have€a€positive€implicit€marginal€valuation€and€anÐ Ô$  Ðadditional€three€have€a€negative€implicit€marginal€valuation.€€For€the€price€of€digital€cableÐ À  Ðservice,€the€implicit€marginal€valuation€is€statistically€significant€and€positive€for€just€eleven€ofÐ ¬ü  Ð105€channel€offerings€and€negative€for€five€channels.€€When€the€level€of€statistical€significance€isÐ ˜è  Ðraised€to€10€percent,€an€additional€four€channels€have€a€positive€implicit€marginal€valuation€andÐ „Ô  Ðan€additional€three€have€a€negative€implicit€marginal€valuation.€Ð pÀ ÐÌÐ  H˜ ÐØØÑ8€÷äXXdìdÈ8ÑÓÓòòIntroductionóóÐ ° Ðà  àCable€service€is€an€example€of€a€bundled€commodity.€€Consumers€purchase€a€group€ofÐ ˆØ Ðchannels€(called€tiers)€from€a€cable€system€without€an€option€of€choosing€only€those€channelsÐ `° Ðthey€prefer.€€The€only€programming€available€on€a€single€item€basis€is€pay„per„view.€€EvenÐ 8 ˆ Ðpremium€channels€such€as€Home€Box€Office€and€the€Movie€Channel€are€typically€offered€only€asÐ  ` Ða€bundle.€€This€bundling€approach€makes€it€difficult€to€determine€the€implicit€marginal€value€ofÐ è 8  Ðindividual€channel€offerings.€€In€what€follows€a€unique€data€set€will€be€used€in€conjunction€with€aÐ À  Ðmodified€hedonic€model€to€estimate€the€implicit€price€of€individual€channels€provided€by€a€cableÐ ˜è  Ðsystem.Ð pÀ Ðà  àThe€pricing€schedule€used€by€most€cable€systems€consists€of€a€basic€tier€(referred€to€asÐ H˜ ÐÔ_ÔBSTÔ_Ô)€that€typically€consists€of€local€stations€(e.g.,€broadcast€channels)€plus€a€few€satelliteÐ  p Ðchannels€and€public,€educational,€and€government€access€(PEG)€channels€if€they€are€carried.Ô_Ô׃×Ý ƒ #ÃÝòòÚ  Ú2Ú  ÚóóÝ  Ý×  ×Ô_Ô€€Ð øH ÐAdditionally,€most€systems€offer€other€tiers,€designated€cable€programming€service€tiers€(Ô_ÔCPSTÔ_Ô).€Ð Ð  ÐThe€major€Ô_ÔCPSTÔ_Ô€typically€consists€of€satellite„delivered€channels€(e.g.,€Ô_ÔMSNBCÔ_Ô,€Ô_ÔCNNÔ_Ô,€theÐ ¨ø ÐWeather€Channel,€and€Ô_ÔESPNÔ_Ô).€€About€90€percent€of€cable€subscribers€take€both€the€Ô_ÔBSTÔ_Ô€andÐ €Ð Ðmajor€Ô_ÔCPSTÔ_Ô€while€the€remaining€10€percent€take€Ô_ÔBSTÔ_Ô€only.׃ ×Ý ƒ #ÃÝòòÚ  Ú3Ú  ÚóóÝ  Ý×  ×€€Both€Ô_ÔBSTÔ_Ô€and€Ô_ÔCPSTÔ_Ô€are€relativelyÐ X ¨ Ðclose€to€pure€bundling€as€consumers€must€purchase€either€the€bundle€or€nothing.Ð 0"€  Ðà  àThe€traditional€explanations€for€bundling€include€the€presence€of€economies€of€scope,Ð ° Ðcomplementarity€in€consumption,€and€as€a€response€to€uncertainty€in€production€(Cabral€[2000]Ð ˆØ Ðand€Holt€and€Sherman€[1986]).׃×Ý ƒ #ÃÝòòÚ  Ú4Ú  ÚóóÝ  Ý×  ×€€Another€explanation€is€that€a€firm€bundles€services€in€an€effortÐ `° Ðto€extract€consumer€surplus€and€thereby€increase€its€profit€(Adams€and€Ô_ÔYellenÔ_Ô€[1976]€and€VarianÐ 8 ˆ Ð[1989]).€€Each€of€these€explanations€have€been€used€to€characterize€the€pricing€approach€adoptedÐ  ` Ðby€the€cable€industry€(see,€e.g.,€Ô_ÔHinderyÔ_Ô€[1998],€Mayo€and€Ô_ÔOtsukaÔ_Ô€[1991],€and€Ô_ÔRubinovitzÔ_ÔÐ è 8  ÐÔ_Ô[1993]).Ô_Ô€€Ð À  Ðà  àBundling€a€service€instead€of€offering€its€components€separately€is€profitable€if€demandÐ ˜è  Ðby€a€consumer€is€negatively€correlated€across€goods€and/or€services€or€if€consumer€demand€isÐ pÀ Ðheterogeneous€(Ô_ÔSchmalenseeÔ_Ô€[1984]).€€For€an€individual€consumer€(subscriber),€the€demand€forÐ H˜ Ðcable€service€is€òòde€factoóó€negatively€correlated€because€watching€one€channel€necessarilyÐ  p Ðprecludes€the€subscriber€from€watching€another€channel€at€the€same€time.׃|×Ý ƒ #ÃÝòòÚ  Ú5Ú  ÚóóÝ  Ý×  ×€€Moreover,€across€allÐ øH Ðsubscribers€demand€is€heterogeneous.€€Consumers€subscribe€to€cable€service€in€order€to€haveÐ Ð  Ðaccess€to€different€types€of€programming€not€available€via€over„the„air€broadcasts€because€theyÐ ¨ø Ðvalue€the€service€that€cable€programming€provides.€€It€is€a€simple€observation€that€individualÐ €Ð Ðsubscribers€are,€in€general,€different€from€one€another€and€that€the€relative€valuation€of€cableÐ X ¨ Ðservice€is€different€for€each€one.€€The€price€of€cable€service€is€just€the€sum€of€consumersððÐ 0"€  Ðwillingness€to€pay€to€have€access€to€channels€they€value€plus€what€these€subscribers€must€pay€forÐ $X" Ðthose€channels€they€do€not€value€(Owen€and€Ô_ÔWildmanÔ_Ô€[1985]).€€The€implicit€prices€(i.e.,€theÐ ° Ðimplicit€marginal€valuation)€of€the€channels€that€subscribers€value€can€be€estimated€using€aÐ ˆØ Ðhedonic€approach.€€The€use€of€the€hedonic€approach€to€estimate€the€implicit€value€or€theÐ `° Ðwillingness€to€pay€of€the€average€subscriber€for€individual€channels€requires€modification€of€theÐ 8 ˆ Ðtraditional€hedonic€technique€because€of€the€Ô_ÔnonlinearityÔ_Ô€in€the€price€function€and€due€to€theÐ  ` Ðgeneral€absence€of€competition€on€the€supply€side€on€the€supply€side€of€the€market.׃×Ý ƒ #ÃÝòòÚ  Ú6Ú  ÚóóÝ  Ý×  ×€€Ð è 8  ÐòòSome€Theoretical€ConstructsóóÐ À  Ðà  àHedonic€analysis€presents€a€convenient€framework€in€which€to€explore€the€relationshipÐ ˜è  Ðbetween€the€price€of€cable€service€and€the€channels€offered€by€a€cable€system.׃@×Ý ƒ #ÃÝòòÚ  Ú7Ú  ÚóóÝ  Ý×  ×€€Hedonic€analysisÐ pÀ Ðis€based€on€the€premise€that€a€differentiated€good€or€service€can€be€described€by€an€objectivelyÐ H˜ Ðmeasured€set€of€characteristics.€€Because€market€price€is€expressed€as€a€function€of€the€good€orÐ  p Ðserviceððs€characteristics,€the€availability€of€objective€measures€of€the€good€or€serviceððsÐ øH Ðcharacteristics€combined€with€observations€on€market€prices€enable€the€estimation€of€the€implicitÐ Ð  Ðmarginal€valuation€or€shadow€price€of€each€characteristic.€€Consequently,€the€market€price€can€beÐ ¨ø Ðdecomposed€into€the€components€that€correspond€to€the€characteristic€of€the€good€or€serviceÐ €Ð Ð(Ô_ÔGrilichesÔ_Ô€[1971]€and€Cowling€and€Ô_ÔCubbinÔ_Ô€[1972]).Ð X ¨ Ðà  àRosen€[1974]€presents€a€hedonic€approach€defined€by€two€sides€of€the€market€„Ð 0"€  Ðconsumers€(demand)€and€producers€(supply).€€The€original€approach€was€enhanced€by€FreemanÐ $X" Ð[1979].׃×Ý ƒ #ÃÝòòÚ  Ú8Ú  ÚóóÝ  Ý×  ×€€A€representative€consumer€maximizes€utility€subject€to€a€budget€constraint.€€Utility€is€aÐ ° Ðfunction€of€the€good€or€service€of€interest€and€a€composite€commodity€consisting€of€all€otherÐ ˆØ Ðgoods€and€services.€€A€bid€function,€Ô_ÔððÔ_Ô,€is€defined.€€It€is€a€function€of€the€characteristics€of€theÐ `° Ðgood€or€service,€utility€of€the€individual€consumer,€and€income.€€The€partial€derivative€of€Ô_ÔððÔ_Ô€withÐ 8 ˆ Ðrespect€to€a€given€characteristic€is€defined€to€be€the€implicit€marginal€valuation€that€the€consumerÐ  ` Ðplaces€on€the€characteristic,€holding,€of€course,€utility,€income€and€all€other€characteristicsÐ è 8  Ðunchanged.€€The€partial€derivative€gives€the€ð ð...€additional€amount€that€must€be€paid€by€anyÐ À  Ðhousehold€to€move€to€a€bundle€with€a€higher€level€of€that€characteristic,€other€things€equalððÐ ˜è  Ð(Freeman€[1993,€p.€372]).€€The€bid€functions€will€be€tangent€to€the€price€function€for€differentÐ pÀ Ðcombinations€of€the€characteristics€of€the€good€or€service€and€utility.€€An€individual€willÐ H˜ Ðmaximize€utility€by€moving€along€the€price€schedule€until€the€marginal€willingness€to€pay€isÐ  p Ðequal€to€the€implicit€marginal€valuation€(price)€of€the€characteristics.€€That€is,€in€equilibrium€aÐ øH Ðconsumerððs€marginal€bid€for€a€characteristic€equals€the€marginal€price€of€the€characteristic€of€theÐ Ð  Ðgood€or€service€of€interest.€€Ð ¨ø Ðà  àAn€analogous€approach€is€used€to€specify€the€supply€side€of€the€market.€€A€perfectlyÐ €Ð Ðcompetitive€firm€is€assumed€to€maximize€profit.€€From€conventional€microeconomic€theory,Ð X ¨ Ðprofit€maximization€occurs€when€the€marginal€revenue€from€an€additional€unit€of€output€is€equalÐ 0"€  Ðto€the€marginal€cost€of€supplying€that€unit€(Martin€[1993]).׃@×Ý ƒ #ÃÝòòÚ  Ú9Ú  ÚóóÝ  Ý×  ×€€A€firm€has€an€offer€function,€Ô_ÔððÔ_Ô.€Ð $X" ÐThis€is€a€function€of€the€characteristics€of€the€good€or€service€being€produced,€revenue,€and€theÐ ° Ðcost€of€production.€€The€offer€functions€will€be€tangent€to€the€price€function€for€differentÐ ˆØ Ðcombinations€of€the€characteristics€of€the€good€or€service€and€profit.׃×Ý ƒ #ÃÝòòÚ  Ú10Ú  ÚóóÝ  Ý×  ×€€That€is,€in€equilibrium€aÐ `° Ðfirmððs€offer€function€for€a€characteristic€equals€the€marginal€price€of€the€characteristic€of€theÐ 8 ˆ Ðgood€or€service€of€interest.€€à  àà h àà À àÐ  ` Ðà  àThe€marginal€cost€of€production€will€equal€the€implicit€marginal€valuation€of€theÐ è 8  Ðcharacteristics€of€the€good€or€service€where€the€bid€functions€and€the€offer€functions€are€bothÐ À  Ðtangent€to€the€price€function.€€The€difference€among€consumers€with€respect€to€their€income€andÐ ˜è  Ðutility€and€the€variations€among€firms€with€regard€to€the€production€technology,€productiveÐ pÀ Ðefficiency,€and€cost€of€inputs€results€in€a€vector€of€bids€and€offers€that€produces€an€equilibriumÐ H˜ Ðprice€schedule€for€the€characteristics€of€the€good€or€service.Ð  p Ðà  àUsing€the€conventional€hedonic€approach€proposed€by€Rosen€and€refined€by€Freeman€toÐ øH Ðmodel€cable€service€necessitates€a€number€of€adjustments.€€First,€consider€the€demand€side.€€€TheÐ Ð  Ðhedonic€estimation€problem€is€not€due€to€demand„supply€interaction.€€An€individual€consumerððsÐ ¨ø Ðdecision€cannot€affect€suppliers€in€the€hedonic€model€because€an€individual€consumer€does€notÐ €Ð Ðaffect€the€price.Ð X ¨ Ðà  àThere€is€a€demand„side€estimation€problem€but€from€another€source.€€The€Ô_ÔnonlinearityÔ_Ô€ofÐ 0"€  Ðthe€hedonic€price€function€allows€the€consumer€to€endogenously€choose€both€quantities€andÐ $X" Ðimplicit€marginal€prices.€€The€result€is€that€the€disturbance€term€on€the€marginal€bid€functionÐ à%0!$ Ðconsists€of€both€an€unobserved€ð ðtastesðð€components€and€a€purely€random€component€(Ô_ÔBartikÔ_ÔÐ ¸'#& Ð[1986]).€€The€marginal€bid€function,€however,€can€be€consistently€estimated€with€correctÐ )à$( Ðinstruments.€€If€the€estimation€problem€is€due€to€unobserved€tastes,€any€set€of€variables€thatÐ ° Ðexogenously€shifts€the€consumerððs€budget€constraint€will€be€appropriate€because€the€budgetÐ ˆØ Ðconstraint€shift€is€correlated€with€the€consumerððs€choice€of€the€characteristics€of€the€good€orÐ `° Ðservice€yet€uncorrelated€with€unobserved€tastes.€€€Ð 8 ˆ Ðà  àNext,€consider€the€supply€side.€€An€important€assumption€in€the€hedonic€formulation€isÐ  ` Ðthat€perfect€competition€between€firms€exists.€€From€this€it€follows€that€the€prices€of€theÐ è 8  Ðcharacteristics€of€a€good€or€service€equals€their€marginal€cost€of€production.€€This€equality€doesÐ À  Ðnot€necessarily€hold€in€noncompetitive€markets.€€The€market€for€cable€service€is,€in€general,€notÐ ˜è  Ðcompetitive€(Brown€and€Uri€[2003]).Ô_Ô׃×Ý ƒ #ÃÝòòÚ  Ú11Ú  ÚóóÝ  Ý×  ×Ô_Ô€€Hence,€there€is€no€reason€to€expect€that€the€marginal€costÐ pÀ Ðof€production€of€a€characteristic€will€equal€the€price€of€that€characteristic.€€Ð ° Ðà  àIn€a€noncompetitive€market,€the€behavior€of€the€firm€changes€relative€to€that€of€a€firm€in€aÐ ˆØ Ðcompetitive€market.€€The€firm€is€still€assumed€to€have€as€its€objective€the€maximization€of€profit.€Ð `° ÐHence,€the€characteristic€of€a€good€or€service€will€still€be€produced€up€to€the€point€where€itsÐ 8 ˆ Ðmarginal€cost€of€production€is€equal€to€its€marginal€revenue.€€With€a€downward€sloping€demandÐ  ` Ðcurve,׃,×Ý ƒ #ÃÝòòÚ  Ú12Ú  ÚóóÝ  Ý×  ×€price€will€exceed€marginal€revenue.€€Thus,€in€equilibrium€price€will€exceed€the€marginalÐ è 8  Ðcost€of€production€of€the€characteristic.€€The€price€charged€to€the€consumer€is€still€the€sum€of€theÐ À  Ðrepresentative€consumerððs€implicit€marginal€valuations€of€the€characteristics€of€the€good€orÐ ˜è  Ðservice.€€The€utility€associated€with€the€consumption€of€that€good€or€service,€however,€will€beÐ pÀ Ðlower€because€a€smaller€quantity€will€be€consumed.Ð H˜ Ðà  àFor€a€perfectly€competitive€market,€the€price€charged€is€equal€to€the€marginal€cost€ofÐ  p Ðproduction€and€will€equal€the€sum€of€the€implicit€marginal€valuations€of€the€characteristics€of€theÐ øH Ðgood€or€service.€€As€noted,€the€noncompetitive€price,€however,€will€exceed€the€marginal€cost€ofÐ Ð  Ðproduction€implying€that€the€price€charged€by€the€noncompetitive€firm€and€the€implicit€marginalÐ ¨ø Ðvaluation€by€the€consumer€are€on€different€indifference€curves.€€Hence,€the€standard€practice€ofÐ €Ð Ðregressing€solely€the€characteristics€of€cable€service€on€price€will€yield€biased€estimates.€€Instead,Ð X ¨ Ðit€is€necessary€to€account€for€supplier€traits€including€the€degree€of€monopoly€power€possessed€byÐ 0"€  Ðthe€firm€in€order€to€obtain€meaningful€estimates€of€the€implicit€valuation€of€the€characteristics€ofÐ ° Ðcable€service.Ð ˆØ Ðà  àÔ_ÔFeenstraÔ_Ô€[1995]€has€shown€that€the€difference€between€the€price€of€a€good€or€service€in€aÐ `° Ðnoncompetitive€market€and€the€competitive€price€can€be€effectively€addressed€in€a€hedonic€modelÐ 8 ˆ Ðby€introducing€a€measure€to€reflect€the€degree€of€market€power€measured€possessed€by€the€firm.€Ð  ` ÐHe€recommends€the€use€of€a€measure€computed€as€the€difference€between€the€price€and€theÐ è 8  Ðmarginal€cost€of€production€and€nominally€refers€to€this€as€the€markup.€€Controlling€for€marketÐ À  Ðpower€will€result€in€estimates€of€the€implicit€prices€of€the€characteristics€of€cable€serviceÐ ˜è  Ðreflecting€both€the€implicit€marginal€valuation€of€cable€service€and€the€marginal€cost€ofÐ pÀ Ðproduction.€€Because€consistent€data€are€not€available€on€the€marginal€cost€of€production€variousÐ H˜ Ðsupplier€traits€will€be€used€as€proxies.€€These€include€the€type€and€size€of€the€cable€system,Ð  p Ðwhether€the€cable€system€is€subject€to€regulation,€whether€it€is€confronted€with€effectiveÐ øH Ðcompetition,€whether€it€is€part€of€a€cluster,€whether€it€is€owned€by€a€multiple€systems€operator,Ð Ð  Ðand€its€capacity.Ð ¨ø Ðà  àThere€is€one€final€issue€involving€the€estimation€of€the€implicit€marginal€valuation€of€theÐ €Ð Ðcharacteristics€of€a€good€or€service.€€It€is€often€difficult,€if€not€impossible,€to€obtain€consistentÐ X ¨ Ðestimates€of€the€coefficients€on€the€demand€and€supply€functions€that€are€the€basis€of€the€hedonicÐ 0"€  Ðinference€(Ô_ÔEppleÔ_Ô€[1987]).€€This€is€not€of€concern€here,€however,€since€the€sole€objective€is€toÐ $X" Ðobtain€consistent€and€unbiased€estimates€of€the€implicit€marginal€valuation€of€the€channelÐ à%0!$ Ðofferings€by€cable€systems€from€the€reduced€form€equation€(i.e.,€the€price€equation)€and€not€inÐ ¸'#& Ðidentifying€the€structural€demand€and€supply€equations€per€se.€€Hence,€consistent€and€unbiasedÐ )à$( Ðestimates€of€the€coefficients€of€the€reduced€form€equation€can€be€obtained€using€ordinary€leastÐ h+¸&* Ðsquares€with€appropriate€adjustments.׃×Ý ƒ #ÃÝòòÚ  Ú13Ú  ÚóóÝ  Ý×  ×Ð ° ÐòòData€IssuesóóÐ ˆØ Ðà  àSection€623(k)€of€the€Communications€Act,€as€amended€by€the€Cable€TelevisionÐ `° ÐConsumer€Protection€and€Competition€Act€of€1992,ÑdÈÑ׃|×Ý ƒ #ÃÝòòÚ  Ú14Ú  ÚóóÝ  Ý×  ×€requires€the€FCC€to€publish€annually€aÐ 8 ˆ Ðstatistical€report€on€cable€prices,€or€more€specifically,€average€rates€for€the€delivery€of€basic€cableÐ  ` Ðservice,€cable€programming€service,€and€equipment.׃¸×Ý ƒ #ÃÝòòÚ  Ú15Ú  ÚóóÝ  Ý×  ×€€The€requirement€to€compare€the€price€ofÐ è 8  Ðcable€service€for€systems€where€effective€competitive€has€been€found€and€the€price€of€cableÐ À  Ðservice€where€effective€competition€has€not€been€found€is€important€given€the€objectives€of€theÐ ˜è  Ð1992€Cable€Act.€€The€primary€data€used€in€this€study€rely€on€the€results€of€the€survey€conductedÐ pÀ Ðas€a€result€of€this€requirement.Ð H˜ Ðà  àThe€òò2002€FCC€Annual€Cable€Price€Surveyóó€requested€data€from€a€sample€of€cable€systemsÐ  p Ðas€of€July€1,€2002€and€July€1,€2001.€€The€òò2002€FCC€Annual€Cable€Price€Surveyóó€was€structured€toÐ øH Ðallow€the€FCC€to€compare€prices€charged€by€two€groups€of€cable€systems:€(1)€systems€that€areÐ Ð  Ðdeemed€to€face€effective€competition€(nominally€referred€to€as€the€competitive€group);€and€(2)Ð ¨ø Ðsystems€that€do€not€face€effective€competition€(the€noncompetitive€group).׃¸ ×Ý ƒ #ÃÝòòÚ  Ú16Ú  ÚóóÝ  Ý×  ×€€Cable€systems€inÐ €Ð Ðthe€competitive€group€are€limited€to€geographic€areas€where€a€cable€operator€has€sought€andÐ ° Ðobtained€a€FCC€finding€of€effective€competition.€€For€these€purposes€the€FCCððs€formal€legalÐ ˆØ Ðdecisions€were€used€as€a€basis€to€determine€whether€effective€competition€exists€based€on€theÐ `° Ðstatutory€definition€of€that€term.€Ý‚.ÿÿÿÝòòÝ  ÝÝ‚.ÿÿÿ7UƒøÝóóÝ  ÝÐ 8 ˆ Ðà  àThe€òò2002€FCC€Annual€Cable€Price€Surveyóó€collected€information€about€average€monthlyÐ  ` Ðrates€for€the€basic€service€tier€(Ô_ÔBSTÔ_Ô)€and€major€cable€programming€service€tier€(Ô_ÔCPSTÔ_Ô).Ô_Ô׃d×Ý ƒ #ÃÝòòÚ  Ú17Ú  ÚóóÝ  Ý×  ×Ô_Ô€€TheÐ è 8  ÐÔ_ÔBSTÔ_Ô€typically€consists€of€local€stations€(e.g.,€broadcast€channels)€plus€a€few€satellite€channelsÐ À  Ðand€public,€educational,€and€government€access€(PEG)€channels€if€the€they€are€carried.Ô_Ô׃x×Ý ƒ #ÃÝòòÚ  Ú18Ú  ÚóóÝ  Ý×  ×Ô_Ô€€TheÐ ˜è  Ðmajor€Ô_ÔCPSTÔ_Ô€typically€consists€of€satellite„delivered€channels.€€About€90€percent€of€cableÐ ° Ðsubscribers€take€both€the€Ô_ÔBSTÔ_Ô€and€major€Ô_ÔCPSTÔ_Ô€while€the€remaining€10€percent€take€Ô_ÔBSTÔ_Ô€only.€Ð ˆØ ÐThe€price€of€Ô_ÔBSTÔ_Ô€service€plus€Ô_ÔCPSTÔ_Ô€service€will€be€used€in€the€analysis€of€the€price€of€analogÐ `° Ðcable€service.€€In€addition€data€were€collected€for€the€most€highly€subscribed€digital€tier€ofÐ 8 ˆ Ðservice.€€This€price€will€be€used€in€the€analysis€of€the€price€of€digital€cable€service.€€InformationÐ  ` Ðwas€also€collected€on€the€complete€channel€offerings€of€the€cable€system€for€both€analog€andÐ è 8  Ðdigital€cable€service,€the€average€monthly€charge€for€equipment,€consisting€of€an€analogÐ À  Ðaddressable€converter€and€remote€control€and€digital€converter€plus€remote€control.€€The€òò2002Ð ˜è  ÐFCC€Annual€Cable€Price€Surveyóó€further€sought€information€needed€to€determine€average€ratesÐ pÀ Ðper€channel€and€to€explain€changes€in€rates.€€Finally,€information€was€gathered€on€other€factorsÐ H˜ Ðthat€affect€cable€prices€and€competition€in€the€multichannel€video€programming€service€marketÐ  p Ðsuch€as€the€cable€systemððs€best€estimate€of€the€number€of€subscribers€to€DBS€service€in€theÐ øH Ðfranchise€area,׃%×Ý ƒ #ÃÝòòÚ  Ú19Ú  ÚóóÝ  Ý×  ×€as€well€as€the€availability€of€advanced€services€such€as€Internet€access,€cableÐ Ð  Ðtelephony,€and€interactive€programming.€€A€summary€of€the€major€findings€of€the€òò2002€FCCÐ ¨ø ÐAnnual€Cable€Price€Surveyóó€for€all€cable€systems€surveyed,€and€separately€for€the€competitive€andÐ €Ð Ðnoncompetitive€groups,€is€available€elsewhere€and€not€reproduced€here.׃ð×Ý ƒ #ÃÝòòÚ  Ú20Ú  ÚóóÝ  Ý×  ×Ð X ¨ Ðà  àTo€compare€average€monthly€rates€of€competitive€and€noncompetitive€cable€systems,׃G×Ý ƒ #ÃÝòòÚ  Ú21Ú  ÚóóÝ  Ý×  ×€aÐ ° Ðseparate€sample€was€selected€from€each€group.€€These€samples€included€282€of€the€356€systemsÐ ˆØ Ðin€the€competitive€group€and€473€of€the€9,790€systems€in€the€noncompetitive€group.€€To€ensureÐ `° Ðthat€the€samples€were€representative€and€to€gain€more€precise€estimates,€both€groups€wereÐ 8 ˆ Ðstratified€into€subgroups€(or€strata)€and€a€portion€of€the€sample€from€each€stratum€was€selected.€Ð  ` ÐThe€competitive€group€was€divided€according€to€the€test€by€which€effective€competition€wasÐ è 8  Ðdetermined€and€the€noncompetitive€group€according€to€the€number€of€subscribers€in€eachÐ À  Ðoperatorððs€cable€system.Ð ˜è  Ðà  àCompetitive€systems€were€divided€into€five€strata€based€upon€the€test€for€which€effectiveÐ pÀ Ðcompetition€was€determined.€€Cable€systems€meeting€the€overbuild€test€were€further€subdividedÐ H˜ Ðinto€two€strata:€(1)€Ô_ÔwirelineÔ_Ô€overbuild€and€(2)€DBS€overbuild.׃z,×Ý ƒ #ÃÝòòÚ  Ú22Ú  ÚóóÝ  Ý×  ×€€The€remaining€systems€in€theÐ  p Ðcompetitive€group€were€divided€into€three€strata€as€follows:€(1Ô‡Â5ÂXXÔòò)Ô#†X÷äXÂÂ5Îf#Ôóó€low€penetration,€(2)€municipal,€Ð øH Ðand€(3)€LEC.€€The€LEC€stratum€consists€of€both€the€incumbent€cable€systems€that€competed€withÐ Ð  Ðan€affiliate€of€a€LEC€at€the€time€that€a€finding€of€effective€competition€was€made€and€the€LECÐ ¨ø Ðaffiliates.€€The€other€strata,€except€for€the€DBS€stratum,€similarly€consist€of€the€incumbent€cableÐ €Ð Ðsystems€as€well€as€the€relevant€competitors.€€The€DBS€stratum€includes€only€the€incumbentÐ X ¨ Ðbecause€monthly€rates€of€DBS€operators€are€not€part€of€the€òò2002€FCC€Annual€Cable€PriceÐ 0"€  ÐSurveyóó.Ð ° Ðà  àFor€the€LEC,€Ô_ÔwirelineÔ_Ô€overbuild,€DBS€overbuild,€and€municipal€strata,€all€95,€37,€42,€andÐ ˆØ Ð14€systems,€respectively,€were€included€because€of€the€relatively€small€number€of€systems€inÐ `° Ðeach€of€these€four€strata.€€A€random€sample€was€chosen€for€the€low€penetration€stratum€becauseÐ 8 ˆ Ðthat€subgroup€had€a€large€number€of€systems,€168€in€all,€from€which€94€were€randomly€selectedÐ  ` Ðfor€that€stratum.€€This€resulted€in€a€total€of€282€systems€for€the€competitive€group.Ð è 8  Ðà  àNoncompetitive€systems€also€were€divided€into€five€strata.€€The€number€of€cable€systemsÐ À  Ðselected€from€each€stratum€depended€on€the€number€of€subscribers€nationwide€in€that€stratum.€€AÐ ˜è  Ðsample€of€systems€not€stratified€by€size€would€have€placed€a€disproportionately€greater€emphasisÐ pÀ Ðon€smaller€systems€relative€to€the€number€of€subscribers€the€smaller€systems€serve.€€The€veryÐ H˜ Ðlarge€stratum€includes€systems€serving€more€than€50,000€subscribers€in€a€single€community.€€TheÐ  p Ðlarge€stratum€contains€systems€serving€more€than€50,000€subscribers,€but€with€no€individualÐ øH Ðcommunity€exceeding€50,000€subscribers.€€The€medium€stratum€is€comprised€of€systems€servingÐ Ð  Ðfrom€10,001€through€50,000€subscribers.€€The€small€stratum€includes€systems€serving€from€1,001Ð ¨ø Ðthrough€10,000€subscribers,€and€the€very€small€stratum€includes€systems€with€1,000€or€fewerÐ €Ð Ðsubscribers.€Ð X ¨ Ðà  àThe€high€proportion€of€subscribers€nationwide€represented€by€the€very€large€stratumÐ 0"€  Ðresulted€in€the€selection€of€all€99€cable€systems€in€that€stratum.€€Other€selections€include€109€ofÐ $X" Ðthe€169€large€systems;€153€of€the€888€medium„sized€systems;€72€of€the€2,717€small€systems;€andÐ à%0!$ Ð40€of€the€5,917€very€small€systems.€€Because€of€the€low€proportion€of€subscribers€nationwideÐ ¸'#& Ðrepresented€by€very€small€systems,€the€formula€for€calculating€sample€size€initially€producedÐ )à$( Ðfewer€selections€from€that€stratum.€€The€number€of€selections€was€adjusted€upward€to€40Ð h+¸&* Ðsystems,€however,€to€ensure€that€there€were€a€sufficient€number€of€observations€from€that€stratumÐ ° Ðfor€valid€statistical€inference.Ð ˆØ Ðà  àOf€the€755€Survey€questionnaires€mailed€to€cable€systems€from€both€groups,€respondentsÐ `° Ðcompleted€694€questionnaires.€€Of€the€61€incomplete€questionnaires,€cable€systems€explainedÐ 8 ˆ Ðthat€the€requisite€data€were€unavailable€for€22€systems€that€had€recently€been€sold€or€combinedÐ  ` Ðwith€other€systems,€and€12€questionnaires€were€undeliverable.€€Competitive€cable€systemsÐ è 8  Ðsubmitted€262€of€the€completed€questionnaires.€€Of€these,€163€systems€had€direct€competition€inÐ À  Ðtheir€geographic€area,€with€91€meeting€the€LEC€test,€72€meeting€the€overbuild€test€(with€31€in€theÐ ˜è  ÐÔ_ÔwirelineÔ_Ô€overbuild€subgroup€and€41€in€the€DBS€overbuild€subgroup),€and€13€served€a€communityÐ pÀ Ðin€which€the€municipality€owned€one€of€the€systems€(thereby€meeting€the€municipal€test).€€Of€theÐ H˜ Ðremaining€respondents€in€the€competitive€group,€86€served€less€than€30€percent€of€households€inÐ  p Ðtheir€service€area€(thereby€meeting€the€low€penetration€test).€Noncompetitive€cable€systemsÐ øH Ðsubmitted€the€remaining€432€responses.Ð Ð  Ðà  àThe€òò2002€FCC€Annual€Cable€Price€Surveyóó€data€are€keyed€to€the€five€digit€Zip€codeÐ ¨ø Ðassociated€with€the€greatest€number€of€subscribers€in€the€franchise€area.€€This€allows€for€mergingÐ €Ð Ðthe€data€from€the€òò2002€FCC€Annual€Cable€Price€Surveyóó€with€social,€economic,€housing,€andÐ X ¨ Ðgeographic€information€from€auxiliary€sources.€€These€supplementary€data€measure€subscriberÐ 0"€  Ðtraits€that€potentially€serve€to€influence€both€the€demand€for€cable€service.€€Ð $X" Ðà  àData€on€the€social,€economic,€and€housing€characteristics€of€subscribers€by€five€digit€ZipÐ à%0!$ Ðcode€were€obtained€from€the€2000€Census€of€Population€Summary€Files€provided€by€the€U.S.Ð ¸'#& ÐCensus€Bureau.€€These€data€are€quite€comprehensive€consisting€of€information€on€schoolÐ )à$( Ðenrollment,€educational€attainment,€marital€status,€disability,€language€spoken,€country€of€birth,Ð h+¸&* Ðemployment€status,€commuting€patterns,€occupation,€income,€type€of€housing€unit,€number€ofÐ ° Ðoccupants,€house€heating€fuel,€rent€versus€own,€mortgage€status,€tenure€of€occupancy,€and€so€on.Ð ˆØ ÐòòPreliminary€Empirical€IssuesóóÐ `° Ðà  àTo€accurately€portray€cable€service,€the€analog€and€digital€components€need€to€beÐ 8 ˆ Ðconsidered€separately.€€DÔ‡X÷äXXX÷äÔigital€cable€systems€provide€a€higher€quality€signal€and€hence€a€betterÐ  ` Ðvideo€image€than€is€possible€with€analog€cable€service.€€Cable€systems€use€digital€technology€toÐ è 8  Ðcompress€video€signals,€allowing€more€than€one€program€service€to€be€carried€in€the€bandwidthÐ À  Ðspace€normally€required€for€one€analog€program€service.€€Typically,€the€signal€is€sent€to€the€homeÐ ˜è  Ðand€decompressed€in€the€set„top€box€for€display€on€the€television€(National€Cable€and€TelevisionÐ pÀ ÐAssociation€[2003]).€€Finally,€digital€cable€service€offers€channels€not€available€via€analog€cableÐ H˜ Ðservice.€€For€example,€based€on€results€from€the€òòÔ#†X÷äXXX÷ä~#Ô2002€FCC€Annual€Cable€Price€SurveyóóÔ‡X÷äXXX÷äÔ,€digitalÐ  p Ðcable€channel€offerings€not€available€through€analog€cable€service€include€Discovery€Ô_ÔEspanolÔ_Ô,Ð øH ÐÔ_ÔEuroNewsÔ_Ô,€Rio€de€la€Plata,€Saigon€Broadcasting€Network,€Sundance€Movie€Channel,€TV€Asia,Ð Ð  ÐTV5€(French),€TRIO,׃o×Ý ƒ #ÃÝòòÚ  Ú23Ú  ÚóóÝ  Ý×  ×€the€Word€Network.Ð ¨ø ÐÔ#†X÷äXXX÷äC#Ôà  àThere€are€a€number€of€empirical€issues€that€must€be€addressed€before€presenting€theÐ €Ð Ðestimation€results€including€the€need€to€use€weighted€least€squares€estimation€and€the€presence€ofÐ X ¨ ÐÔ_ÔheteroscedasticityÔ_Ô,€the€appropriate€functional€specification,€and€the€existence€of€outliers€in€theÐ 0"€  Ðdata.€€Because€the€survey€collected€information€from€cable€systems€of€widely€disparate€sizes,€theÐ $X" Ðpotential€for€Ô_ÔheteroscedasticÔ_Ô€error€terms€exists€(Davidson€and€MacKinnon€[1993]).€€The€priceÐ à%0!$ Ðequations€for€analog€cable€service€and€digital€cable€service€are€considered€separately.€€TheÐ ¸'#& ÐÔ_ÔGoldfeldÔ_Ô„Ô_ÔQuandtÔ_Ô€test€is€used€to€test€for€Ô_ÔheteroscedasticityÔ_Ô€with€the€sample€divided€between€cableÐ ° Ðsystems€with€more€that€50,000€subscribers€and€those€with€less€than€50,000€(Ô_ÔGoldfeldÔ_Ô€and€Ô_ÔQuandtÔ_ÔÐ ˆØ Ð[1965]).׃ ×Ý ƒ #ÃÝòòÚ  Ú24Ú  ÚóóÝ  Ý×  ×€€That€is,€the€observations€are€divided€into€two€groups€in€such€a€way€that€under€the€nullÐ `° Ðhypothesis€of€Ô_ÔhomoscedasticityÔ_Ô,€the€disturbance€variances€will€be€the€same€while€under€theÐ 8 ˆ Ðalternative€hypothesis,€the€disturbance€variances€will€differ€systematically.€€With€a€linear€priceÐ  ` Ðequation€specification€for€analog€cable€service€and€the€middle€200€observations€dropped€in€orderÐ è 8  Ðto€increase€the€power€of€the€test,׃!|×Ý ƒ #ÃÝòòÚ  Ú25Ú  ÚóóÝ  Ý×  ×€the€computed€Ô_ÔGoldfeldÔ_Ô„Ô_ÔQuandtÔ_Ô€test€statistic€is€2.37.€€For€aÐ À  ÐÔ_ÔsemilogÔ_Ô€specification€(i.e.,€the€log€of€the€price€of€analog€cable€service€is€used€as€the€dependentÐ ˜è  Ðvariable),€the€computed€test€statistic€is€3.15.€€The€critical€value€is€F(146,€91)€=€1.32€at€the€5Ð pÀ Ðpercent€level€of€significance.€€Alternately,€with€a€linear€price€equation€specification€for€digitalÐ H˜ Ðcable€service€and€the€middle€200€observations€dropped,€the€computed€Ô_ÔGoldfeldÔ_Ô„Ô_ÔQuandtÔ_Ô€testÐ  p Ðstatistic€is€2.16.€€For€a€Ô_ÔsemilogÔ_Ô€specification,€the€computed€test€statistic€is€2.65.€€The€criticalÐ øH Ðvalue€is€F(59,€71)€=€1.51€at€the€5€percent€level€of€significance.€€Thus,€in€all€instances€the€nullÐ Ð  Ðhypothesis€of€Ô_ÔhomoscedasticityÔ_Ô€is€rejected€at€the€5€percent€level.€€Consequently,€weighted€leastÐ ¨ø Ðsquares€will€be€used€where€the€weight€for€the€analog€price€equation€is€the€number€of€analog€cableÐ €Ð Ðsubscribers€in€a€system€and€the€weight€for€the€digital€price€equation€will€be€the€number€of€digitalÐ X ¨ Ðcable€subscribers€in€a€system.€€The€usual€Ô_ÔEickerÔ_Ô„White€asymptotic€formula€is€used€in€computingÐ 0"€  Ðthe€robust€estimate€of€the€variance„covariance€matrix€(Davidson€and€MacKinnon€[1993]€andÐ $X" ÐWhite€[1980]).Ð à%0!$ Ðà  àNext€consider€the€issue€of€the€appropriate€functional€specification.€€It€is€typically€the€caseÐ ¸'#& Ðthat€the€functional€specification€used€in€hedonic€studies€specifies€price€as€a€linear€function€of€aÐ ° Ðset€of€explanatory€variables€or€that€the€dependent€variable€(price)€is€transformed€by€logòòeóó€(i.e.,Ð ˆØ ÐNapierian€logarithms)€before€empirically€estimating€the€relationship.Ô_Ô׃"×Ý ƒ #ÃÝòòÚ  Ú26Ú  ÚóóÝ  Ý×  ×Ô_Ô€€Consideration€of€theÐ `° Ðappropriate€specification€will€be€thus€limited€to€these€two€alternatives.€€In€order€to€determineÐ 8 ˆ Ðwhether€one€specification€is€preferred€to€another,€a€straightforward€Ô_ÔnonnestedÔ_Ô€test€is€used.€€TheÐ  ` Ðtest€is€applied€to€both€the€analog€cable€price€equation€and€the€digital€cable€price€equation.€€TheÐ è 8  Ðtest€chosen€is€the€J„test€developed€by€Davidson€and€MacKinnon€[1981,€1993].€€The€basic€idea€ofÐ À  Ðthe€test€is€to€embed€a€competing€regression€functional€specification€in€a€more€general€one€andÐ ˜è  Ðthen€test€the€original€model€against€it.€€Each€of€the€specifications€is€considered€in€Ô_ÔpairwiseÔ_ÔÐ pÀ Ðfashion€with€each€of€the€functional€specifications€alternately€serving€as€the€correct€specificationÐ H˜ Ð(i.e.,€the€null€hypothesis).€€For€the€analog€price€equation€with€the€linear€specification€serving€asÐ  p Ðthe€null€hypothesis,€the€computed€test€statistic€is€9.36.€€With€the€Ô_ÔsemilogÔ_Ô€specification€serving€asÐ øH Ðthe€null€hypothesis,€the€computed€test€statistic€is€8.91.€€For€the€digital€price€equation€with€theÐ Ð  Ðlinear€specification€serving€as€the€null€hypothesis,€the€computed€test€statistic€is€5.90.€€With€theÐ ¨ø ÐÔ_ÔsemilogÔ_Ô€specification€serving€as€the€null€hypothesis,€the€computed€test€statistic€is€5.92.€€TheÐ €Ð Ðcritical€value€at€the€5€percent€level€is€1.96.׃x|×Ý ƒ #ÃÝòòÚ  Ú27Ú  ÚóóÝ  Ý×  ×€€Thus,€in€all€instances,€the€computed€test€statisticÐ X ¨ Ðexceeds€the€critical€value.€€This€implies€that€there€is€no€clearly€preferred€functional€specificationÐ 0"€  Ðfor€either€the€analog€price€equation€or€the€digital€price€equation.€€That€is,€there€is€no€empiricalÐ $X" Ðbasis€for€selecting€one€functional€specification€over€another.€€Hence,€the€Ô_ÔsemilogÔ_Ô€specification€isÐ à%0!$ Ðarbitrarily€chosen.€à ¸ àà  àÐ ¸'#& Ðà  àThere€is€one€final€issue€to€be€considered€before€turning€to€the€estimation€of€the€priceÐ ° Ðequations.€€This€involves€the€presence€of€outliers€in€the€data.€€Regression€diagnostics€is€a€tool€forÐ ˆØ Ðassessing€the€quality€and€reliability€of€regression€estimates.€€It€is€used€here€to€test€for€the€presenceÐ `° Ðof€outliers.€€It€is€especially€useful€for€cross€sectional€data€(the€type€being€used€here)€in€helping€toÐ 8 ˆ Ðdetermine€in€a€systematic€way€the€location€of€data€points€that€are€either€unusual€or€inordinatelyÐ  ` Ðinfluential€(Greene€[1997]).€€Ð è 8  Ðà  àRegression€diagnostic€tests€were€developed€by€Ô_ÔBelsleyÔ_Ô€et€al.€[1980].€€The€testsÐ À  Ðsystematically€search€for€unusual€or€influential€data.€€That€is,€regression€diagnostics€look€forÐ ˜è  Ðobservations€that€lie€outside€patterns€set€by€other€data,€or€those€that€strongly€influence€theÐ pÀ Ðregression€results.€€The€impact€of€such€data€points€is€rarely€apparent€from€even€a€close€inspectionÐ H˜ Ðof€the€basic€data€series.€€The€basis€of€regression€diagnostics€is€an€analysis€of€the€response€ofÐ  p Ðcoefficient€estimates€to€controlled€perturbations€of€the€inputs€including€the€parameters€to€beÐ øH Ðestimated,€error€and€model€specification,€and€the€ordering€of€the€data.€€Four€separate€regressionÐ Ð  Ðdiagnostic€tests€are€available€including€Ô_ÔRSTUDENTÔ_Ô,€Ô_ÔHATDIAGÔ_Ô,€Ô_ÔCOVRATÔ_Ô,€and€Ô_ÔDFFITSÔ_Ô.€€TheÐ ¨ø Ðinterested€reader€is€referred€to€Ô_ÔBelsleyÔ_Ô€et€al.€[1980]€for€information€on€the€distribution€of€the€testÐ €Ð Ðstatistics.Ð X ¨ Ðà  àBoth€the€analog€cable€price€equation€and€the€digital€cable€price€equation€were€subjectedÐ 0"€  Ðto€the€regression€diagnostic€tests€individually.€€An€observation€was€judged€to€be€an€outlier€if€itÐ $X" Ðfailed€two€or€more€of€the€tests.€€For€the€analog€cable€price€equation,€twenty€four€observationsÐ à%0!$ Ðwere€judged€to€be€outliers€and€for€the€digital€cable€price€equation,€eighteen€observations€wereÐ ¸'#& Ðjudged€to€be€outliers.€€There€were€no€overlaps€in€the€observations€determined€to€be€outliers.€€ThatÐ )à$( Ðis,€there€was,€for€example,€no€observation€that€was€judged€to€be€an€outlier€for€both€analog€cableÐ h+¸&* Ðprice€and€digital€cable€price.€€Moreover,€a€closer€inspection€of€the€observations€judged€to€beÐ ° Ðoutliers€did€not€reveal€any€obvious€problems€such€as€a€misplaced€decimal€point€in€recording€theÐ ˆØ Ðdata€or€a€data€transposition€error.€Ð `° Ðà  àFor€each€of€the€price€equations,€a€qualitative€(dummy)€variable€is€introduced€to€mark€theÐ 8 ˆ Ðoutliers.€€The€variable€is€defined€to€equal€zero€if€two€or€more€of€the€regression€diagnostics€testsÐ  ` Ðare€passed€and€one€if€two€or€more€are€failed.€€Defined€this€way,€the€impact€of€the€outliers€will€beÐ è 8  Ðcorrected€only€for€their€impact€on€the€estimate€of€the€constant€(intercept)€term.Ð À  ÐòòEstimation€ResultsóóÐ ˜è  Ðà  àA€considerable€amount€of€preliminary€analysis€went€into€determining€the€finalÐ pÀ Ðspecifications€for€the€analog€cable€price€equation€and€the€digital€cable€price€equation€in€order€toÐ H˜ Ðreflect€appropriately€subscriber€traits€and€cable€system€traits.€€There€are€few€insights€to€beÐ  p Ðgleaned€from€recounting€all€of€the€details€of€this€preliminary€analysis.׃#×Ý ƒ #ÃÝòòÚ  Ú28Ú  ÚóóÝ  Ý×  ×€€Simply€note€that€theÐ øH Ðâ âobjective€has€been€to€obtain€coefficient€estimates€that€are€theoretically€consistent€and€that€areÐ Ð  Ðcredible€and€robust׃$×Ý ƒ #ÃÝòòÚ  Ú29Ú  ÚóóÝ  Ý×  ×€but€that€also€can€give€some€insight€into€the€nature€and€extent€of€the€implicitÐ ° Ðâ âmarginal€valuation€of€individual€channel€offerings€by€cable€systems.€€Finally,€the€specificationsÐ ˆØ Ðare€data€dependent.€€That€is,€they€have€been€developed€based€on€the€data€collected€on€the€òò2002Ð `° ÐFCC€Annual€Cable€Price€Surveyóó.€€Ð 8 ˆ Ðà  àFor€the€final€estimation,€not€all€694€observations€could€be€used.€€For€nine€of€theÐ  ` Ðobservations,€the€Zip€code€information€was€either€missing€or€inaccurate€and€could€not€beÐ è 8  Ðcorrected.€€These€observations€were€dropped€from€the€data€set€for€the€analog€cable€priceÐ À  Ðequation.€€Additionally,€there€were€103€cable€systems€that€did€not€provide€digital€cable€service€inÐ ˜è  Ð2002.€€It€was€necessary€to€drop€these€observations€from€the€data€set€for€the€digital€cable€priceÐ pÀ Ðequation.€€The€majority€(i.e.,€70.1€percent)€were€either€very€small€or€small€noncompetitiveÐ H˜ Ðsystems€or€LEC€systems€in€the€competitive€group.€€This€gave€a€total€of€685€observations€used€inÐ  p Ðthe€estimation€of€the€analog€cable€price€equation€and€582€observations€used€in€the€estimation€ofÐ øH Ðthe€digital€price€equation.€€By€dropping€these€observations,€any€inferences€with€regard€to€theÐ Ð  Ðbehavior€of€subscribers€and€cable€systems€must€be€qualified€accordingly.€€This€understanding€isÐ ¨ø Ðimplicit€and€the€point€is€not€pursued€in€the€subsequent€discussion.Ð €Ð Ðà  àSeveral€channel€offerings€were€also€eliminated€from€the€data€set.€€Eleven€channels€fromÐ X ¨ Ðthe€analog€cable€price€equation€data€set€and€fourteen€channels€from€the€digital€cable€priceÐ 0"€  Ðequation€data€set€were€offered€by€fewer€than€ten€cable€systems.€€Moreover,€these€channelÐ $X" Ðofferings€were€relatively€highly€collinear׃&T×Ý ƒ #ÃÝòòÚ  Ú30Ú  ÚóóÝ  Ý×  ×€(i.e.,€only€a€few€cable€systems€offered€these€channelsÐ à%0!$ Ðand€these€cable€systems€tended€to€offer€most€of€the€eleven€(analog)€or€fourteen€(digital)Ð ° Ðchannels).€€To€avoid€any€problems€of€statistical€inference,€these€channels€were€simply€droppedÐ ˆØ Ðfrom€the€data€sets.€€Finally,€all€cable€systems€offer€the€broadcast€network€channels.€€Hence,€theseÐ `° Ðare€not€explicitly€included.׃w×Ý ƒ #ÃÝòòÚ  Ú31Ú  ÚóóÝ  Ý×  ×€€Ð 8 ˆ Ðà  àThe€coefficient€estimates€for€the€analog€cable€price€equation€and€the€digital€cable€priceÐ  ` Ðequation€were€obtained€using€weighted€least€squares€with€correction€for€Ô_ÔheteroscedasticityÔ_Ô.€€TheÐ è 8  Ðweight€for€the€analog€cable€price€equation€was€the€number€of€analog€cables€service€subscribersÐ À  Ðand€the€weight€for€the€digital€cable€price€equation€was€the€number€of€digital€cable€subscribers.€Ð ˜è  ÐThe€estimation€results€for€the€analog€cable€price€equation€are€presented€in€Table€1€and€theÐ pÀ Ðestimation€results€for€the€digital€cable€price€equation€are€reported€in€Table€2.€€A€completeÐ H˜ Ðdescription€of€the€channels€and€their€offerings€(i.e.,€the€type€of€programming)€can€be€found€inÐ  p ÐNational€Cable€and€Telecommunications€Association€[2003]).€€A€description€of€the€variablesÐ øH Ðreflecting€cable€system€(supplier)€and€subscriber€(consumer)€traits€along€with€their€sources€isÐ Ð  Ðgiven€in€the€Appendix.€€The€results€are€fairly€revealing.€€With€regard€to€the€price€of€analog€cableÐ ¨ø Ðservice,€the€implicit€marginal€valuation€is€statistically€significantly€different€from€zero€at€the€5Ð €Ð Ðpercent€level€and€positive€for€just€fifteen€of€106€channel€offerings€but€it€is€also€negative€for€sevenÐ X ¨ Ðchannels.€€When€the€level€of€statistical€significance€is€raised€to€10€percent,€an€additional€sevenÐ 0"€  Ðchannels€have€a€positive€implicit€marginal€valuation€and€an€additional€three€have€a€negativeÐ $X" Ðimplicit€marginal€valuation.€€Among€the€channels€that€have€a€positive€implicit€marginal€valuationÐ à%0!$ Ðare€the€Animal€Planet€Network,€Bravo,€the€Food€Network,€GEMS€Television,׃'|×Ý ƒ #ÃÝòòÚ  Ú32Ú  ÚóóÝ  Ý×  ×€and€Shop€NBC.€Ð ¸'#& ÐEach€of€these€channel€offerings€has€an€implicit€value€of€about€three€to€four€percent€of€the€price€ofÐ ° Ðanalog€cable€service.€€The€channels€with€the€largest€implicit€value€of€about€seven€to€eight€percentÐ ˆØ Ðinclude€BET€Jazz,€CN8,׃(×Ý ƒ #ÃÝòòÚ  Ú33Ú  ÚóóÝ  Ý×  ×€and€Ô_ÔToonÔ_Ô€Disney.€€Channels€with€a€negative€implicit€marginalÐ `° Ðvaluation€include€Ô_ÔCNNÔ_Ô€Headline€News,€Court€TV,€the€Fox€News€Channel,€and€NOGGIN.Ô_Ô׃m|×Ý ƒ #ÃÝòòÚ  Ú34Ú  ÚóóÝ  Ý×  ×Ô_ԀР8 ˆ ÐThese€channels€have€a€negative€implicit€marginal€valuation€of€between€three€and€nine€percent€ofÐ  ` Ðthe€price€of€analog€cable€service.€€Ð è 8  Ðà  àJust€as€interesting€as€the€coefficients€that€are€statistically€significant€are€the€results€thatÐ À  Ðindicate€the€relatively€large€number€of€channels€that€have€neither€a€positive€nor€a€negativeÐ ˜è  Ðimplicit€marginal€valuation.€€Seventy€five€of€the€analog€cable€channels€offered€have€noÐ pÀ Ðstatistically€significant€implicit€marginal€valuation.€€These€channels€include€the€majority€of€theÐ H˜ Ðnews€channels,€the€foreign€language€channels,€and€the€sports€channels.€€Ð  p Ðà  àVariables€introduced€to€reflect€subscriber€and€cable€system€traits€are,€in€general,€notÐ øH Ðstatistically€significant.€€A€large€number€of€variables€were€considered€in€preliminary€analysis€inÐ Ð  Ðorder€to€capture€the€unobserved€tastes€of€subscribers.€€These€variables€included€schoolÐ ¨ø Ðenrollment,€educational€attainment,€marital€status,€disability,€language€spoken,€employmentÐ €Ð Ðstatus,€commuting€patterns,€occupation,€income,€type€of€housing€unit,€number€of€occupants,Ð X ¨ Ðhouse€heating€fuel,€mortgage€status,€rent,€and€tenure€of€occupants.€€Only€a€few€of€these€variablesÐ 0"€  Ðwere€retained€in€the€final€specification€since€almost€all€did€not€prove€to€be€statisticallyÐ $X" Ðsignificant.€€Of€those€retained,€only€median€income€and€the€poverty€level€are€statisticallyÐ à%0!$ Ðsignificantly€different€from€zero€at€the€five€percent€level.€€As€median€income€increases€so€doesÐ ¸'#& Ðthe€price€of€analog€cable€service.€€The€sign€on€the€poverty€measure,€however,€is€different€thanÐ ° Ðwhat€would€be€expected€òòa€priorióó.€€It€makes€no€sense€to€conclude€that€a€higher€incidence€ofÐ ˆØ Ðpoverty€is€associated€with€a€higher€price€for€analog€cable€service.€€This€is€an€anomalous€resultÐ `° Ðthat€remains€to€be€explained.€€Ð 8 ˆ Ðà  àOnly€a€couple€of€the€variables€introduced€to€control€for€market€power€are€statisticallyÐ  ` Ðsignificant.€€The€coefficient€estimates€indicate€that€competitive€cable€systems€that€are€owned€byÐ è 8  ÐÔ_ÔLECsÔ_Ô€and€municipalities€have€a€lower€price€for€analog€cable€service.€€With€regard€to€theÐ À  Ðstatistical€insignificance€of€other€factors€that€might€potentially€impact€the€price€of€analog€cableÐ ˜è  Ðservice,€there€are€a€couple€of€issues€to€note.€€First,€most€cable€systems€in€the€sample€are€both€partÐ pÀ Ðof€a€cluster€and€are€affiliated€with€a€Ô_ÔMSOÔ_Ô.Ô_Ô׃‘×Ý ƒ #ÃÝòòÚ  Ú35Ú  ÚóóÝ  Ý×  ×Ô_Ô€€Hence,€the€lack€of€finding€of€any€statisticalÐ H˜ Ðsignificance€for€these€variables€is€most€likely€an€artifact€of€there€being€not€enough€variation€inÐ  p Ðthese€explanatory€variables€across€the€entire€sample.€€Also,€system€capacity€measured€in€terms€ofÐ øH Ðmegahertz€does€not€appear€to€be€significant€in€explaining€the€price€of€analog€cable€service.€€AÐ Ð  Ðlarger€capacity€should€lead€to€a€higher€cost€and€a€higher€price.€€This€is€not€indicated€in€the€dataÐ ¨ø Ðfrom€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€This€result,€however,€is€also€probably€a€functionÐ €Ð Ðof€there€not€being€enough€variation€across€cable€systems€in€the€sample.€€Ð X ¨ Ðà  àThe€penetration€of€DBS€in€a€franchise€area€has€a€quite€significant€and€negative€effect€onÐ 0"€  Ðthe€price€analog€cable€service.€€This€measure€captures€the€degree€of€competition€that€cableÐ $X" Ðsystems€as€the€dominant€firm€face€from€Ô_ÔnoncableÔ_Ô€providers,€the€fringe€firm€(Brown€and€UriÐ à%0!$ Ð[2003]).׃u×Ý ƒ #ÃÝòòÚ  Ú36Ú  ÚóóÝ  Ý×  ×€€A€one€percent€increase€in€the€number€of€DBS€subscribers€relative€to€the€total€numberÐ ° Ðof€subscribers€will€result€in€about€a€0.10€percent€reduction€in€the€price€analog€cable€service.׃t,×Ý ƒ #ÃÝòòÚ  Ú37Ú  ÚóóÝ  Ý×  ×Ð ˆØ Ðà  àWith€regard€to€the€price€of€digital€cable€service,€the€implicit€marginal€valuation€isÐ `° Ðstatistically€significantly€different€from€zero€at€the€5€percent€level€and€positive€for€just€eleven€ofÐ 8 ˆ Ð105€channel€offerings€but€it€is€also€negative€for€five€channels.€€When€the€level€of€statisticalÐ  ` Ðsignificance€is€raised€to€10€percent,€an€additional€four€channels€have€a€positive€implicit€marginalÐ è 8  Ðvaluation€and€an€additional€three€have€a€negative€implicit€marginal€valuation.€€Among€theÐ À  Ðchannels€that€have€a€positive€implicit€marginal€valuation€are€Discovery€Ô_ÔEspanolÔ_Ô,€E!€Style,׃p@ ×Ý ƒ #ÃÝòòÚ  Ú38Ú  ÚóóÝ  Ý×  ×€FineÐ ˜è  ÐLiving,€the€National€Geographic€Channel,€and€Ô_ÔToonÔ_Ô€Disney€Ô_ÔEspanolÔ_Ô.€€Each€of€these€channelÐ pÀ Ðofferings€has€an€implicit€marginal€valuation€of€about€six€to€seven€percent€of€the€price€of€digitalÐ H˜ Ðcable€service.€€Channels€with€a€negative€implicit€marginal€valuation€include€Bravo,€CanalesÐ  p ÐGroup,׃)| ×Ý ƒ #ÃÝòòÚ  Ú39Ú  ÚóóÝ  Ý×  ×€Ô_ÔCNNÔ_Ô€in€Ô_ÔEspanolÔ_Ô,€Ô_ÔEWTNÔ_Ô,׃qÌ ×Ý ƒ #ÃÝòòÚ  Ú40Ú  ÚóóÝ  Ý×  ×€Tech€TV,€and€ZEE€TV.׃r×Ý ƒ #ÃÝòòÚ  Ú41Ú  ÚóóÝ  Ý×  ×€€These€channels€have€a€negativeÐ øH Ðimplicit€marginal€valuation€of€between€two€and€five€percent€of€the€price€of€digital€cable€service.€€Ð Ð  Ðà  àAs€in€the€case€of€analog€service,€equally€interesting€as€the€coefficients€that€are€statisticallyÐ ° Ðsignificant€are€the€results€that€indicate€the€relatively€large€number€of€digital€cable€channels€thatÐ ˆØ Ðhave€neither€a€positive€nor€negative€implicit€marginal€valuation.€€Eighty€three€of€the€digital€cableÐ `° Ðservice€channels€offered€have€no€statistically€significant€implicit€marginal€valuation.€€TheseÐ 8 ˆ Ðchannels€include€most€of€the€news€channels,€a€large€number€of€sports€channels,€the€familyÐ  ` Ðchannels,€and€the€shopping€networks.€Ð è 8  Ðà  àAs€was€the€case€for€the€analog€price€cable€service€equation,€variables€introduced€to€reflectÐ À  Ðsubscriber€and€cable€system€traits€are,€in€general,€not€statistically€significant.€€Only€a€few€of€theÐ ˜è  Ðvariables€considered€in€preliminary€analysis€were€retained€in€the€final€specification€since€almostÐ pÀ Ðall€did€not€prove€to€be€statistically€significant.€€Of€those€retained€to€reflect€subscriber€traits,€justÐ H˜ Ðmedian€income€proved€to€be€statistically€significantly€different€from€zero€at€the€5€percent€levelòòóó.€€Ð  p Ðà  àJust€two€of€the€variables€introduced€to€reflect€market€power€are€statistically€significantlyÐ øH Ðdifferent€from€zero€at€the€5€percent€level.€€Competitive€cable€systems€that€are€owned€byÐ Ð  Ðmunicipalities€have€a€lower€price€for€digital€cable€service.€€Also,€cable€systems€subject€to€localÐ ¨ø Ðregulation€have€a€higher€price€than€those€not€subject€to€local€regulation.€€This€is€not€surprising.€Ð €Ð ÐWhile€rates€for€Ô_ÔBSTÔ_Ô€service€are€regulated,€those€for€Ô_ÔCPSTÔ_Ô€service€are€not.€€Any€revenue€shortfallÐ X ¨ Ðis€more€than€made€up€for€by€the€combined€Ô_ÔBSTÔ_Ô€plus€Ô_ÔCPSTÔ_Ô€price.€€With€regard€to€the€statisticalÐ 0"€  Ðinsignificance€of€other€variables€reflecting€market€power€of€a€cable€system€that€might€potentiallyÐ $X" Ðimpact€the€price€of€digital€cable€service,€the€same€concerns€discussed€for€analog€cable€serviceÐ à%0!$ Ðapply.Ð ¸'#& Ðà  àThe€penetration€of€DBS€in€a€franchise€area€has€a€quite€significant€and€negative€effect€onÐ )à$( Ðthe€price€digital€cable€service.€€A€one€percent€increase€in€DBS€penetration€results€in€about€a€1.39Ð h+¸&* Ðpercent€reduction€in€the€price€of€digital€cable€service.Ð ° ÐòòComparisonóóÐ ˆØ Ðà  àThere€are€few€studies€that€attempt€to€decompose€explicitly€the€price€of€cable€service€intoÐ `° Ðits€component€parts.€€The€data€issues€are€typically€insurmountable.€€Nevertheless,€the€studies€thatÐ 8 ˆ Ðhave€endeavored€to€do€this€rely€on€data€from€Warren€Publishing€[annual].€€This€source,€however,Ð  ` Ðdoes€not€contain€the€level€of€detail€on€cable€systemððs€channel€offerings€found€in€the€òò2002€FCCÐ è 8  ÐAnnual€Cable€Price€Surveyóó.€€There€is€only€a€single€study€that€has€results€directly€comparable€toÐ À  Ðthose€obtained€here.€€Ô_ÔAnstineÔ_Ô€[2001]€presents€estimates€of€the€implicit€marginal€price€of€channelÐ ˜è  Ðgroupings€in€the€framework€of€a€conventional€hedonic€model€specification€with€an€adjustmentÐ pÀ Ðfor€cable€system€market€power€based€on€data€for€1991.€€The€results€are€obtained€using€ordinaryÐ H˜ Ðleast€squares€with€no€attention€to€the€potential€econometric€shortcomings€in€the€data€and€theÐ  p Ðspecification€that€have€been€of€concern€here.€€It€is€found€that€sports€networks€as€a€group,€newsÐ øH Ðnetworks€as€a€group,€and€family€networks€as€a€group€all€have€positive€implicit€marginal€values.€Ð Ð  ÐThe€results€are€somewhat€at€odds€with€those€obtained€here.€€The€difference€can€be€attributed€to€aÐ ¨ø Ðnumber€of€factors€including€different€data€sets€and€time€periods.€€Ð €Ð ÐòòTwo€Final€QuestionsóóÐ X ¨ Ðà  àThere€are€two€questions€that€remain€to€be€addressed.€€First,€is€there€an€explanation€of€whyÐ 0"€  Ðsubscribers€place€a€nonzero€implicit€marginal€valuation€on€relatively€few€of€a€cable€systemððsÐ $X" Ðchannel€offerings€for€both€analog€and€digital€cable€service?€€Second,€if€the€estimates€obtainedÐ à%0!$ Ðhere€do€accurately€measure€the€implicit€marginal€valuation€of€individual€channels,€what€are€theÐ ¸'#& Ðimplications?Ð )à$( Ðà  àConcerning€the€first€question,€one€possible€reason€that€most€individual€channels€do€notÐ h+¸&* Ðhave€a€statistically€significant€positive€or€negative€implicit€marginal€valuation€relates€to€bundling.€Ð ° ÐAs€noted€previously,€cable€service€is€an€example€of€pure€bundling€because€the€subscriber€cannotÐ ˆØ Ðpurchase€channel€offerings€individually€from€a€cable€system.€€Moreover,€a€cable€systemÐ `° Ðostensibly€constructs€the€bundle€to€appeal€to€all€or,€at€last,€most€potential€subscribers.€€TheÐ 8 ˆ Ðsubscriber€interested€in€news€and,€say€not€sports,€as€well€as€the€subscriber€interested€in€sports€andÐ  ` Ðnot€news€are€included€as€subscribers€to€the€same€service.€€To€have€access€to€any€one€channel€on€aÐ è 8  Ðtier,€a€subscriber€must€purchase€the€entire€tier.€€That€is,€to€be€able€to€view€one€channel,€access€toÐ À  Ðall€channels€on€a€tier€must€be€paid€for.€€The€values€estimated€here€represent€the€average€implicitÐ ˜è  Ðmarginal€valuation€of€channels€across€all€subscribers€and€cable€systems€for€Ô_ÔBSTÔ_Ô€plus€Ô_ÔCPSTÔ_ÔÐ pÀ Ðservice.€€This€includes€subscribers€who€value€a€channel€offering€relatively€highly,€relativelyÐ H˜ Ðlowly,€or€not€at€all€with€the€result€that,€on€average,€most€channels€have€no€nonzero€implicitÐ  p Ðmarginal€valuation.Ð øH Ðà  àWhat€are€the€implications€of€the€finding€that€relatively€few€of€the€analog€and€digital€cableÐ Ð  Ðchannels€offered€by€cable€systems€have€a€positive€implicit€marginal€value?€€It€certainly€appearsÐ ¨ø Ðthat€subscribersðð€utility€could€be€enhanced€if€a€different€pricing€structure€were€adopted.€€That€is,Ð €Ð Ðadoption€of€an€òòa€la€carteóó€approach€to€pricing€both€analog€and€digital€cable€service€where€theÐ X ¨ Ðchannels€offered€by€a€cable€system€are€unbundled€would€allow€subscribers€to€pay€for€just€theÐ 0"€  Ðchannels€on€which€they€place€a€positive€implicit€marginal€value€and€would€result€in€enhancedÐ $X" Ðsubscriber€well„being.€€Such€an€approach€has€some€potential€benefits€and€well€as€pitfalls€forÐ à%0!$ Ðcable€systems.€€On€the€benefit€side,€because€of€a€cable€systemððs€finite€channel€capacity,€carryingÐ ¸'#& Ða€channel€with€negative€or€zero€implicit€marginal€value€precludes€another,€perhaps€moreÐ )à$( Ðattractive€channel€from€being€offered€to€subscribers.€€Additionally,€associated€with€most€cableÐ h+¸&* Ðchannel€offerings€is€a€programming€fee€paid€by€a€cable€system€that€could€be€avoided€if€theÐ ° Ðchannels€not€valued€by€subscribers€or€those€with€a€negative€implicit€marginal€valuation€were€notÐ ˆØ Ðcarried€by€the€cable€system.€€On€the€negative€side€is€the€potential€loss€in€revenue€to€cableÐ `° Ðsystems.€€Whether€this€would€materialize,€however,€would€be€a€function€of€the€price€ofÐ 8 ˆ Ðindividual€channel€offering,€the€aggregate€number€of€subscribers€to€the€various€channels€offeredÐ  ` Ðby€the€cable€system,€an€so€on.€€There€are€also€additional€costs€that€the€cable€system€would€incurÐ è 8  Ðto€upgrade€its€system€to€provide€channels€individually€to€subscribers.€€In€fact,€this€is€the€mainÐ À  Ðargument€that€cable€systems€are€currently€using€against€adopting€an€òòa€la€carteóó€approach€toÐ ˜è  Ðpricing€analog€and€digital€cable€service€(see,€e.g.,€òòÔ_ÔCableFAXÔ_Ô€Dailyóó,€September€2,€2003]).Ð pÀ ÐòòConclusionóóÐ H˜ Ðà  àThe€objective€has€been€to€estimate€the€implicit€marginal€value€of€channel€offerings€byÐ  p Ðcable€systems€in€the€United€States.€€Cable€service€is€an€example€of€a€bundled€commodity€and,€asÐ øH Ðsuch,€presents€some€unique€issues€that€must€be€addressed€when€endeavoring€to€estimate€theÐ Ð  Ðimplicit€marginal€valuation€of€individual€channels.€€Hedonic€analysis€presents€a€convenientÐ ¨ø Ðframework€in€which€to€explore€the€relationship€between€the€price€of€cable€service€and€theÐ €Ð Ðchannels€offered€by€a€cable€system.€€There€are€a€number€of€estimation€concerns€that€must€beÐ X ¨ Ðtaken€into€account€in€order€to€model€adequately€the€price€of€cable€service€including€theÐ 0"€  ÐÔ_ÔendogeneityÔ_Ô€of€both€quantities€and€implicit€marginal€prices€and€the€presence€of€market€power€onÐ $X" Ðthe€part€of€cable€service€providers.€€Data€from€the€òò2002€FCC€Annual€Cable€Price€Surveyóó€are€usedÐ à%0!$ Ðin€the€estimation.€€These€data€include€the€complete€channel€line€up€for€both€analog€and€digitalÐ ¸'#& Ðcable€service€for€all€of€the€cable€systems€surveyed.€€Ð )à$( Ðà  àA€number€of€estimation€issues€are€investigated€including€the€presence€ofÐ h+¸&* ÐÔ_ÔheteroscedasticityÔ_Ô,€the€appropriate€functional€specification,€and€the€existence€of€outliers€in€theÐ ° Ðdata.€€Ô_ÔHeteroscedasticityÔ_Ô€is€present€and€corrected€for€in€the€estimation.€€A€Ô_ÔsemilogarithmicÔ_ÔÐ ˆØ Ðspecification€is€used€and€corrections€are€made€for€the€data€outliers.Ð `° Ðà  àTwo€separate€hedonic€equations€are€estimated€„€one€for€the€price€of€analog€cable€serviceÐ 8 ˆ Ðand€one€for€the€price€of€digital€cable€service.€€With€regard€to€the€price€of€analog€cable€service,€theÐ  ` Ðimplicit€marginal€valuation€is€statistically€significantly€different€from€zero€at€the€5€percent€levelÐ è 8  Ðand€positive€for€just€fifteen€of€106€channel€offerings€and€negative€for€seven€channels.€€When€theÐ À  Ðlevel€of€statistical€significance€is€raised€to€10€percent,€an€additional€seven€channels€have€aÐ ˜è  Ðpositive€implicit€marginal€valuation€and€an€additional€three€have€a€negative€implicit€marginalÐ pÀ Ðvaluation.€€For€the€price€of€digital€cable€service,€the€implicit€marginal€valuation€is€statisticallyÐ H˜ Ðsignificantly€different€from€zero€at€the€5€percent€level€and€positive€for€just€eleven€of€105€channelÐ  p Ðofferings€and€negative€for€five€channels.€€When€the€level€of€statistical€significance€is€raised€to€10Ð øH Ðpercent,€an€additional€four€channels€have€a€positive€implicit€marginal€valuation€and€an€additionalÐ Ð  Ðthree€have€a€negative€implicit€marginal€valuation.€€Ð ¨ø Ðà  àThe€implications€of€the€results€are€clear.€€Subscribersðð€utility€can€be€enhanced€if€aÐ €Ð Ðdifferent€pricing€structure€were€adopted€for€cable€service.€€Adoption€of€an€òòa€la€carteóó€approach€toÐ X ¨ Ðpricing€both€analog€and€digital€cable€service€would€improve€subscribersðð€well„being€by€havingÐ 0"€  Ðthem€pay€for€just€those€channels€on€which€they€place€a€positive€implicit€marginal€value.€€ThisÐ $X" Ðapproach€has€some€potential€benefits€and€well€as€pitfalls€for€cable€systems.€€On€the€benefit€side,Ð à%0!$ Ðbecause€of€a€cable€systemððs€finite€channel€capacity,€carrying€a€channel€with€negative€or€zeroÐ ¸'#& Ðimplicit€marginal€value€precludes€another,€perhaps€more€attractive€channel€from€being€offered€toÐ )à$( Ðsubscribers.€€Additionally,€associated€with€most€cable€channel€offerings€is€a€programming€feeÐ h+¸&* Ðpaid€by€cable€systems€that€could€be€avoided€if€the€channels€with€negative€or€zero€implicitÐ ° Ðmarginal€valuation€were€not€carried€by€the€cable€system.€€On€the€negative€side€is€the€potentialÐ ˆØ Ðloss€in€revenue€to€cable€systems.€€There€are€also€additional€costs€that€the€cable€system€wouldÐ `° Ðincur€to€upgrade€its€system€to€provide€channels€individually€to€subscribers.€Ð 8 ˆ ÐÌÐ  è 8  ÐÓ  ÓòòÓÓReferencesóóÐ ° ÐÓà Óà  àà ` àà ¸ àà  àà h àà À àà  àà p àà È àÌAdams,€W.,€and€J.€Ô_ÔYellenÔ_Ô,€ð ðCommodity€Bundling€and€the€Burden€of€Monopoly,ðð€òòQuarterlyÐ ˆØ ÐJournal€of€Economicsóó,€Vol.€90€(1976),€pp.€475„498.Ð tÄ ÐÌÔ_ÔAnstineÔ_Ô,€D.,€ð ðHow€Much€Will€Consumers€Pay?€A€Hedonic€Analysis€of€the€Cable€TelevisionÐ L œ ÐIndustry,ðð€òòReview€of€Industrial€Organizationóó,€Vol.€19€(2001),€pp.€129„147.Ð 8 ˆ ÐÌÔ_ÔBartikÔ_Ô,€T.,€ð ðThe€Estimation€of€Demand€Parameters€in€Hedonic€Price€Models,ðð€òòJournal€ofÐ  ` ÐÔ_ÔPolicticalÔ_Ô€Economyóó,€Vol.€95€(1987),€pp.€81„88.à  àÐ ü L  ÐÌBeard,€R.,€R.€Ô_ÔEkelundÔ_Ô,€G.€Ford,€and€R.€Ô_ÔSabaÔ_Ô,€ð ðPrice„Quality€Tradeoffs€and€Welfare€Effects€inÐ Ô$  ÐCable€Television€Markets,ðð€òòJournal€of€Regulatory€Economicsóó,€Vol.€20€(2001),€pp.€107„123.Ð À  ÐÌÔ_ÔBelsleyÔ_Ô,€D.,€E.€Ô_ÔKuhÔ_Ô,€and€R.€Ô_ÔWelschÔ_Ô,€òòRegression€Diagnostics:€Identifying€Influential€Data€andÐ ˜è  ÐSources€of€Ô_ÔCollinearityÔ_Ôóó,€John€Wiley€and€Sons,€Inc.,€New€York,€NY,€1980.Ð „Ô  ÐÌBrown,€K.,€and€N.€Uri,€òòModeling€the€Market€for€Multichannel€Video€Programming€SubscriptionÐ \¬ ÐServiceóó,€Media€Bureau,€Federal€Communications€Commission,€Washington,€DC,€2003.Ð H˜ ÐÌCabral,€L.,€òòIntroduction€to€Industrial€Organizationóó,€The€MIT€Press,€Cambridge,€2000.à Ð àÐ  p ÐÌCourt,€A.,€ð ðHedonic€Price€Indexes€with€Automotive€Examples,ðð€in€òòThe€Dynamics€of€AutomobileÐ øH ÐDemandóó,€General€Motors€Corporation,€1939,€pp.€99„117.Ð ä4 ÐÌCowling,€K.,€and€J.€Ô_ÔCubbinÔ_Ô,€ð ðHedonic€Price€Indices€for€United€Kingdom€Cars,ðð€òòEconomicÐ ¼  ÐJournalóó,€Vol.€82€(1972),€pp.€963„978.Ð ¨ø ÐÌCrandall,€R.,€and€H.€Ô_ÔFurchtgottÔ_Ô„Roth,€òòCable€TV:€Regulation€or€Competition?óó,€BrookingsÐ €Ð ÐInstitution,€Washington,€DC,€1996.Ð l¼ ÐÌDavidson,€R.,€and€J.€MacKinnon,€ð ðSeveral€Tests€for€Model€Specification€in€the€Presence€ofÐ D!” ÐAlternative€Hypotheses,ðð€òòÔ_ÔEconometricaÔ_Ôóó,€Vol.€49€(1981),€pp.€781„793.Ð 0"€  ÐÌDavidson,€R.,€and€J.€MacKinnon,€òòEstimation€and€Inference€in€Econometricsóó,€Oxford€UniversityÐ $X" ÐPress,€Oxford,€1993.Ð ô$D # ÐÌÔ_ÔEmmonsÔ_Ô,€W.,€and€R.€Ô_ÔPragerÔ_Ô,€ð ðThe€Effects€of€Market€Structure€and€Ownership€on€Prices€andÐ Ì&"% ÐService€Offerings€in€the€Cable€Television€Industry,ðð€òòRand€Journal€of€Economicsóó,€Vol.€28€(1997),Ð ¸'#& Ðpp.€732„750.Ð ¤(ô#' ÐÌÔ_ÔEppleÔ_Ô,€D.,€ð ðHedonic€Prices€and€Implicit€Markets:€Estimating€Demand€and€Supply€FunctionsÐ |*Ì%) Ðform€Differentiated€Products,ðð€òòJournal€of€Political€Economyóó,€Vol.€95€(1987),€pp.€59„80.à Ð àÐ h+¸&* ЇFederal€Communications€Commission,€òòNinth€Annual€Reportóó,€MM€Docket€02„145,€FederalÐ ° ÐCommunications€Commission,€Washington,€DC,€December€31,€2002.Ð œì ÐÌFederal€Communications€Commission,€òòReport€on€Cable€Industry€Pricesóó,€MM€Docket€92„266,Ð tÄ ÐFederal€Communications€Commission,€Washington,€DC,€July€8,€2003.Ð `° ÐÌÔ_ÔFeenstraÔ_Ô,€R.,€Exact€Hedonic€Price€Indices,ðð€òòReview€of€Economics€and€Statisticsóó,€Vol.€72€(1995),Ð 8 ˆ Ðpp.€634„653.Ð $ t ÐÌFord,€G.,€and€J.€Jackson,€ð ðHorizontal€Concentration€and€Vertical€Integration€in€the€CableÐ ü L  ÐTelevision€Industry,ðð€òòReview€of€Industrial€Organizationóó,€Vol.€12€(1997),€pp.€501„518.Ð è 8  ÐÌFreeman,€A.,€òòThe€Benefits€of€Environmental€Improvement:€Theory€and€Practiceóó,€Johns€HopkinsÐ À  ÐUniversity€Press,€Baltimore,€1979.Ð ¬ü  ÐÌFreeman,€A.,€òòThe€Measurement€of€Environmental€and€Resource€Valuesóó,€Resources€for€theÐ „Ô  ÐFuture,€Washington,€1993.Ð pÀ ÐÌGeneral€Accounting€Office,€òòThe€Changing€Status€of€Competition€to€Cable€Televisionóó,Ð H˜ ÐGAO/Ô_ÔRCEDÔ_Ô€99„158,€Resources€and€Community€Division,€U.S.€General€Accounting€Office,Ð 4„ ÐWashington,€DC,€1999.Ð  p ÐÌÔ_ÔGoldfeldÔ_Ô,€S.,€and€R.€Ô_ÔQuandtÔ_Ô,€ð ðSome€Tests€for€Ô_ÔHomoscedasticityÔ_Ô,ðð€òòJournal€of€the€AmericanÐ øH ÐStatistical€Associationóó,€Vol.€60€(1965),€pp.€539„547.Ð ä4 ÐÌGreene,€W.,€òòEconometric€Analysisóó,€Third€Edition,€Prentice€Hall,€Upper€Saddle€River,€NJ,€1997.Ð ¼  ÐÌÔ_ÔGrilichesÔ_Ô,€Z.,€òòPrice€Indices€and€Quality€Changeóó,€Harvard€University€Press,€Cambridge,€1971.Ð ”ä ÐÌÔ_ÔHinderyÔ_Ô,€L.,€ð ðTestimony€Before€the€Senate€Commerce€Committee€on€Commerce,€Science,€andÐ l¼ ÐTransportation,ðð€U.S.€Senate,€Washington,€DC,€July€28,€1998.Ð X ¨ ÐÌHolt,€C.,€and€R.€Sherman,€ð ðQuality€Uncertainty€and€Bundling,ðð€in€P.€Ô_ÔIppolitoÔ_Ô€and€D.€Ô_ÔScheffmanÔ_ÔÐ 0"€  Ð(Ô_ÔedsÔ_Ô.),€òòEmpirical€Approaches€to€Consumer€Protectionóó,€Federal€Trade€Commission,€Washington,Ð #l! Ð1986,€pp.€221„250.Ð $X" ÐÌMartin,€S.,€òòAdvanced€Industrial€Economicsóó,€Blackwell€Publishers,€Oxford,€1993.Ð à%0!$ ÐÌMayo,€J.,€and€Y.€Ô_ÔOtsukaÔ_Ô,€ð ðDemand,€Pricing,€and€Regulation:€Evidence€from€the€Cable€TVÐ ¸'#& ÐIndustry,ðð€òòRand€Journal€of€Economicsóó,€Vol.€22€(1991),€pp.€396„410.à  àÐ ¤(ô#' ÐÌÔ‡X÷äXXX÷äÔNational€Cable€and€Television€Association,€òòBroadband€Servicesóó,€Ô_ÔNCTAÔ_Ô,€Washington,€DC,€2003.Ô#†X÷äXXX÷ä!&#ÔÐ |*Ì%) ÐÐ h+¸&* ÐNational€Cable€and€Telecommunications€Association,€òòCable€Developmentsóó,€CommunicationsÐ ° Ðand€Public€Affairs€Department,€Washington,€DC,€2003.Ð œì ÐÌOwen,€B.,€and€S.€Ô_ÔWildmanÔ_Ô,€ð ðProgram€Competition,€Diversity,€and€Multichannel€Bundling€in€theÐ tÄ ÐNew€Video€Industry,ðð€in€E.€Noam€(ed.),€òòVideo€Media€Competition:€Regulation,€Economics,€andÐ `° ÐTechnologyóó,€Columbia€University€Press,€New€York,€1985.à p àà È àÐ L œ ÐÌPark€Center€for€Investment€Research,€òòClear€Channel€Communicationsóó,€Johnson€Graduate€SchoolÐ $ t Ðof€Management,€Cornell€University,€Ithaca,€NY,€2003.Ð  ` ÐÌRosen,€S.,€ð ðHedonic€Prices€and€Implicit€Markets:€Product€Differentiation€in€Pure€Competition,ððÐ è 8  ÐòòJournal€of€Political€Economyóó,€Vol.€82€(1974),€pp.€34„55Ð Ô$  ÐÌÔ_ÔRubinovitzÔ_Ô,€R.,€ð ðMarket€Power€and€Price€Increases€for€Basic€Cable€Service€since€Deregulation,ððÐ ¬ü  ÐòòRand€Journal€of€Economicsóó,€Vol.€24€(1993),€pp.€1„18.Ð ˜è  ÐÌÔ_ÔSchmalenseeÔ_Ô,€R.,€ð ðÔ_ÔGaussianÔ_Ô€Demand€and€Commodity€Bundling,ðð€òòJournal€of€Businessóó,€Vol.€57Ð pÀ Ð(1984),€pp.€S211„S230.Ð \¬ ÐÌUri,€N.,€ð ðA€Note€on€the€Estimation€of€the€Demand€for€Sugar€in€the€Presence€of€MeasurementÐ 4„ ÐError€in€the€Data,ðð€òòApplied€Economicsóó,€Vol.€27€(1995),€pp.€83„94.Ð  p ÐÌVarian,€H.,€ð ðPrice€Discrimination,ðð€in€R.€Ô_ÔSchmalenseeÔ_Ô€and€R.€Ô_ÔWilligÔ_Ô€(Ô_ÔedsÔ_Ô.),€òòHandbook€ofÐ øH ÐIndustrial€Organizationóó,€North„Holland€Publishing€Company,€Amsterdam,€1989.Ð ä4 ÐÌÔ_ÔVeronisÔ_Ô€Ô_ÔSuhlerÔ_Ô,€òòCommunications€Industry€Forecastóó,€Ô_ÔVeronisÔ_Ô€Ô_ÔSuhlerÔ_Ô€Media€Merchant€Bank,€NewÐ ¼  ÐYork,€NY,€July€2001.Ð ¨ø ÐÌWarren€Publishing,€òòTelevision€and€Cable€Ô_ÔFactbookÔ_Ô:€Stations,€Cable€and€Servicesóó,€WarrenÐ €Ð ÐPublishing,€Washington,€DC,€annual.Ð l¼ ÐÌWaugh,€F.,€Quality€Factors€Influencing€Vegetable€Prices,ðð€òòJournal€of€Farm€Economicsóó,€Vol.€10Ð D!” Ð(1928),€pp.€185„196.Ð 0"€  ÐÌWhite,€H.,€ð ðA€Ô_ÔHeteroskedasticityÔ_Ô„Consistent€Covariance€Matrix€Estimator€and€a€Direct€Test€forÐ $X" ÐÔ_ÔHeteroskedasticityÔ_Ô,ðð€òòÔ_ÔEconometricaÔ_Ôóó,€Vol.€48€(1980),€pp.€817„838.Ð ô$D # ÐÌÌÌÌÐ  )à$( ÐÓ  ÓAppendixÐ ° ÐÌDefinitions€and€Data€SourcesÐ ˆØ ÐÌÓ  ÓThe€analog€cable€price€is€the€sum€of€the€Ô_ÔBSTÔ_Ô€and€Ô_ÔCPSTÔ_Ô€cable€price.€€The€data€come€from€theÐ `° Ðòò2002€FCC€Annual€Cable€Price€Surveyóó.€€Ð L œ ÐÌThe€digital€cable€price€is€the€price€of€the€most€highly€subscribed€digital€tier.€€The€data€come€fromÐ $ t Ðthe€òò2002€FCC€Annual€Cable€Price€Surveyóó.Ð  ` ÐÌÔ_ÔDSMEDÔ_Ô,€Ô_ÔDSLGÔ_Ô,€and€Ô_ÔDSSUPÔ_Ô€correspond€to€the€definitions€of€medium,€large,€and€very€largeÐ è 8  Ðnoncompetitive€cable€systems€found€in€the€text.€€The€value€of€the€variable€is€defined€to€equal€oneÐ Ô$  Ðif€the€cable€operator€meets€the€definition€criterion€and€zero€otherwise.Ð À  ÐÌÔ_ÔDSLECÔ_Ô,€Ô_ÔDSOBÔ_Ô,€Ô_ÔDSMUNIÔ_Ô,€Ô_ÔDSLPÔ_Ô€correspond€to€the€definitions€of€LEC,€Overbuild,€Municipal,Ð ˜è  Ðand€LP€competitive€cable€systems€found€in€the€text.€€The€value€of€the€variable€is€defined€to€equalÐ „Ô  Ðone€if€the€cable€operator€meets€the€definition€criterion€and€zero€otherwise.Ð pÀ ÐÌDREG€denotes€whether€the€basic€service€tier€price€is€subject€to€local€regulation€for€the€franchiseÐ H˜ Ðarea.€€It€is€defined€to€equal€one€if€there€is€local€regulation€and€zero€otherwise.€€The€data€comeÐ 4„ Ðfrom€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€Ð  p ÐÌÔ_ÔDCOMPÔ_Ô€denotes€whether€the€FCC€has€made€a€finding€of€effective€competition€within€theÐ øH Ðcommunity.€€It€is€defined€to€equal€one€if€there€is€effective€competition€and€zero€otherwise.€€TheÐ ä4 Ðdata€come€from€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€Ð Ð  ÐÌÔ_ÔDCLUSTÔ_Ô€denotes€whether€the€cable€system€is€part€of€a€Ô_ÔMSOÔ_Ô€cluster€of€two€or€more€systems.€€ItÐ ¨ø Ðis€defined€to€equal€one€if€the€cable€system€is€part€of€a€cluster€and€zero€otherwise.€€The€data€comeÐ ”ä Ðfrom€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€Ð €Ð ÐÌDMSO€denotes€whether€the€cable€system€is€affiliated€with€a€multiple€systems€operator€(Ô_ÔMSOÔ_Ô).€€ItÐ X ¨ Ðis€defined€to€equal€one€if€the€cable€system€is€affiliated€with€a€Ô_ÔMSOÔ_Ô€and€zero€otherwise.€€The€dataÐ D!” Ðcome€from€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€Ð 0"€  ÐÌCAPACITY€is€the€cable€system€capacity€as€of€July€1,€2002€measured€in€terms€of€megahertz.€€TheÐ $X" Ðdata€come€from€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.€€Ð ô$D # ÐÌÔ_ÔMEDINCÔ_Ô€is€the€median€household€income€corresponding€to€a€specific€cable€franchise€area€keyedÐ Ì&"% Ðby€Zip€code.€€The€data€come€from€the€2000€Census€of€Population.Ð ¸'#& ÐÌÔ_ÔUNEPCTÔ_Ô€is€the€rate€of€unemployment€corresponding€to€a€specific€cable€franchise€area€keyed€byÐ )à$( ÐZip€code.€€The€data€come€from€the€2000€Census€of€Population.Ð |*Ì%) ÐÐ h+¸&* ÐPOV€is€the€percent€of€the€population€with€a€household€income€below€the€official€poverty€levelÐ ° Ðcorresponding€to€a€specific€cable€franchise€area€keyed€by€Zip€code.€€The€data€come€from€theÐ œì Ð2000€Census€of€Population.Ð ˆØ ÐÌÔ_ÔRETIAMTÔ_Ô€is€the€monthly€amount€of€retirement€income€received€by€residents€in€the€franchiseÐ `° Ðarea.€€The€data€come€from€the€2000€Census€of€Population.Ð L œ ÐÌÔ_ÔPCDBSÔ_Ô€is€the€ratio€of€the€total€number€of€multichannel€video€programming€subscribers€whoÐ $ t Ðsubscribe€to€DBS€to€the€number€of€subscribers€to€all€multichannel€video€subscription€service.€Ð  ` ÐThe€data€come€from€the€òò2002€FCC€Annual€Cable€Price€Surveyóó.Ð ü L  ÐÌÔ_ÔDALOGÔ_Ô€is€a€qualitative€(dummy)€variable€introduced€to€mark€those€observations€that€are€judgedÐ Ô$  Ðto€be€outliers€for€the€analog€cable€price€equation€as€observations€of€interest.Ð À  Ðà  àà ` àà ¸ àà  àà h àà À àà  àà p àà È àà  àÌÔ_ÔDDLOGÔ_Ô€is€a€qualitative€(dummy)€variable€introduced€to€mark€those€observations€that€are€judgedÐ ˜è  Ðto€be€outliers€for€the€digital€cable€price€equation€as€observations€of€interest.€€