2010୍ֻ9௹技术进步a教育收益与收入不平等*徐 舒 ଽಸิေ:本文建立了一个基于技能偏向型技术进步的一般均衡模型,表明教育回报率的变化是引起我国劳动者收入不平等扩大的重要原因b基于该模型的进一步推导分解出了教育的要素回报效应与要素结构效应,前者拉大了收入不平等,后者降低了收入不平等,但总效应仍然是收入不平等的扩大b利用基于RIF回归的分解方法,本文验证了上述理论预测与实际数据的高度一致性bܱՍ:收入不平等 技能偏向型技术进步 RIF回归分解၂aႄა໓ངሸඍ୍࣍ট,ݓসᆀ൬ೆҵए҂؎ঔն֥གྷའ൳֞ᆟکa࿐ඌࢸၛࠣഠ߶նᇙ֥ۚ؇ܱᇿ,္ᄀটᄀӮູૂุބಕᇙษં֥ࢊׄbऌൗࢸၿྛܙෘ,ݓ2001୍֥ࠎୄ༢ඔჿູ0 447b၇ᅶ2006୍2ᄅ14ರ∀ܻૼರБ#֥ႄඍ,ᇏݓഠ॓ჽ൬ೆٳज़ีቆ۴ऌӬཔೆטҰඔऌ࠹ෘԛ֥ݓ2002୍֥ࠎୄ༢ඔჿູ0 454b2008୍2ᄅ,∀ຬ#ᇛ़ᄛ໓ᆷԛ,ݓ֥ࠎୄ༢ඔۚႿधն؟ඔؿᅚᇏݓࡅ,ѩԩႿݓൎ֥ቋۚׄbՎު,ܱႿ۲ᇕޙਈ൬ೆ҂֥֩ᆷѓ,Ⴍః൞ࠎୄ༢ඔ֥࠹ෘٚمࠣఃᄝݓ֥ൡႨྟ֥ᆚંჟࡆࠗਛbෙಖ۲ᇕޙਈ൬ೆٳ҂֥֩ᆷѓᄝݓ֥ൡႨྟಯႵրഅ,ڿ۪ष٢ၛটݓ൬ೆҵएঔնಏ൞҂ᆚ֥൙ൌbࣜ࠶࿐ࢸؓ൬ೆٳ҂֩໙ี֥ܱᇿႮটၘࣲ,ݓ࿐ᆀ္ࣉྛਔܼٗطധೆ֥࣮b༱౩ࢭ(֩2007)০Ⴈᇏݓࡅ๖൬ೆטҰཛଢ(CHIP)ඔऌ,ٳ༅ਔݓ1988∃2002୍Ӭᆓ֥э߄൝ބଆൔb࣮іૼᄝဢЧൈ௹ଽ,ݓӬᆓं֥धؓཁᇷࡨഒ,൬ೆٳ҂֩ࡆख;ྷགྷབaݚ(ۖ2008)ษંਔݓԚՑٳᇏ֥ਆࠞٳ߄ࠣఃӮၹ,іૼਆࠞٳ߄ᇶေ൞Ⴎস܊ངᆃ۱ٳѓሙᄝӁြࡗ֥ҵၳᄯӮ֥;ޠྖࡹaࣁӄ(2007)ᄝؓ൬ೆਆࠞٳ߄ҩ؇ٚمࣉྛඍ֥ࠎԤഈ,ᄎႨࠫ۱ᇶေᆷѓൌᆣٳ༅ਔᇏݓं֥൬ೆਆࠞٳ߄,ඪૼᇏݓӬཔਆࠞٳ߄aӬᆓࠣ୪ըଽ֥҆ਆࠞٳ߄aခݚაଽ֥ਆࠞٳ߄ၛࠣྛြਆࠞٳ߄ն؟ӯགྷഈശ൝bؓ൬ೆ҂࣮֥֩ਸ਼၂۱ࠃᄁਵთ൞۽ሧ൬ೆ҂֥֩ٳࢳbᆃ҆ٳ֥࣮؟ҐႨ࠹ਈࣜ࠶࿐ٚم,ࡼਆ۱҂ಕ(ุࠇཌྷಕุᄝ҂ൈ௹)֥۽ሧ൬ೆҵၳٳࢳູস৯۱ุหᆘ(ೂྟљa୍ਭa൳࢝ტӱ؇)֥֩႕ཙ,ࢹՎฐ෬҂ഠ߶ಕุᄝ۽ሧ൬ೆഈթᄝҵљ֥ჰၹbӈҐႨ֥ٚمႵ:ࠎႿཌྟ݂֥߭Blinder OaxacaٳࢳaJuhnetal(1993)֥ٚم,ၛࠣMachado&Mata(2005)ิԛ֥ࠎႿٳ໊ඔ݂֥߭ٳࢳٚمb!২ೂ,ိ༵ݓaཫ(2007)ࣼ০Ⴈquantile JMPٚمՖሹุഈ࣮ਔ۽ሧ൬ೆ҂֩ഈശ֥ࢲܒིႋაࡎི۬ႋbऎุ࣮֞ଽಸ,ഈඍ࣮ၘࣜട֥ࠣ֞ٚ૫Ⴕ۽ሧ൬ೆ֥ྟљҵၳ(ૅဇ,2005a;Ӿະ,2008)aྛြҵၳ(અԣਊaൌ,2007)ၛࠣੀದ१* ྷൽ,༆ଲҍࣜն࿐ࣜ࠶ა࣮ܵჽ,Ⴏᆟщ:610074,ሰྐད:sxu@Ч໓൳༆ଲҍࣜն࿐ 211۽ӱ 3௹ࡹഡཛଢሧᇹbۋ྆ཐۖଲն࿐້ཟת࢝൱ᄝ໓ᅣൌᆣٚمഈ֥ᆷ֝ၛࠣٓӵᄶ࢝൱֥ಣ౦Ϻᇹbۋ྆ਆ໊େബ۠ದ֥Џၩ,ૌؓ໓ᅣ֥ڿࣉބປఏ֞ਔ٤ӈᇗေ֥ቔႨb֒ಖ,໓ᄳሱڵb! བྷ֬౷ޚ(2007);ਸ਼ຓ,ѯ֩(2007)ؓགྷႵ֥൬ೆҵएঔնӮၹ֥ٳࢳٚمࣉྛਔࢠູಆ૫֥߭ܤބඍb79
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩აЧֹದ१֥۽ሧ൬ೆҵၳ֩(ૅဇ,2005b;֬౷ޚ,2007)bაഈඍ໓ང҂,Ч໓০ႨᇏݓࡲूაႏအטҰඔऌ(CHNS)࣮ਔ1991୍֞2006୍ݓ۽ሧ൬ೆ҂֩ӱ؇ᄝൈࡗົ؇ഈ֥ሺཟэ߄หᆘb໓ᅣࡹ৫ਔ၂۱ࠎႿିொཟඌࣉ҄(skill biasedtechnicalchange)֥၂Ϯनޙଆ,ѩ๙ݖൌᆣіૼھଆაགྷൌඔऌ֥ۚ؇໖ކ∃∃∃࢝ტ֥ေࢲܒིႋ(ᆷۚ֩࢝ტসᆀбᇗ֥ᄹࡆ)ࢆ֮ਔ۽ሧ൬ೆ҂֩ӱ؇,࢝ტ֥ေБԏིႋ(ᆷ࢝ტ߭Бੱ֥ഈശ)ᄝ۷նӱ؇ഈঔնਔ۽ሧ൬ೆ҂֩ӱ؇,ఃሹིݔಯಖ൞൬ೆ҂֩ӱ؇ഈശ,ѩᆃᇕིႋؓӬᆓসᆀ۷ࡆཁᇷbЧ໓֥ᇶေ܊ང,ၧࠧა࠻Ⴕ໓ང֥љᇶေႵਆ۱ٚ૫:၂൞๙ݖંଆ֥,֝ٳࢳԛ࢝ტ֥ေࢲܒིႋࠣေБԏིႋ,ѩყҩਔਆᇕིႋؓ۽ሧ൬ೆ҂֥֩҂ቔႨٚཟbؽ൞ᄝൌᆣٚمഈҐႨFirpoetal(2007)ؿᅚԛ֥ࠎႿᄜᇏྏ߄႕ཙݦඔ݂߭(recenteredinfluencefunctionregression,༯ӫRIF݂߭)֥ٳࢳٚم,ၛّ႘൬ೆ҂֥֩҂࠹ᆷѓٳљख़߂ԛ࢝ტ֥ေࢲܒིႋაေБԏིႋؓ൬ೆٳ҂֩ӱ؇֥႕ཙbؽaݓ֥۽ሧ൬ೆ҂֩ა࢝ტ߭Бੱ:1991∃2006୍ೂభ໓෮ඍ,Ч໓֥ൌᆣ࣮ᇶေࠎႿᇏݓࡲूაႏအטҰඔऌ(CHNS)b!۴ऌ࣮ଢ֥,ૌ࿊౼ਔ1991୍a1993୍a2004୍ࠣ2006୍֥ඔऌ,ѩࡼ1991୍ބ1993୍֥ဢЧކѩ,קၬູࠎ௹(t0ൈ௹);ࡼ2004୍ა2006୍֥ဢЧކѩ,קၬູБۡ௹(t1ൈ௹)b%ބགྷႵ֥ݓଽຓ໓ངোර,໓ᇏ֥সᆀဢЧᆷ֥൞୍ਭᄝ18∃65ෟaѩႵᆞӈ۽ቔ֥সᆀ;൬ೆᆷ֥൞সᆀ֥ᄅ۽ሧ൬,ೆЇও۽ሧaࢂࣁބఃҀ์൬ೆb҂୍ٺ֥൬ೆ๙ݖCPIטᆜູ2006୍֥ၬ൬ೆ,ᄝൌ࠽ٳ༅ᇏ۴ऌܸ২ؓᄅ൬ೆ౼ሱಖؓඔb(၂)۽ሧ൬ೆ҂֥֩э߄൝ູਔّ႘۽ሧ൬ೆٳ҂֩ᄝൈࡗഈ֥ሺཟэ߄౦ঃ,ࡼt0ൈ௹ބt1ൈ௹ဢЧᇏসᆀ֥൬ೆٳ҃ࡆၛؓб,ೂ1෮ൕb1ቐཁൕ֥൞ਆ۱ൈ௹ؓඔ۽ሧ֥ނૡ؇ܙ࠹bॖၛुԛ:აt0ൈ௹ཌྷб,t1ൈ௹ؓඔ۽ሧ֥ૡ؇ݦඔૼཁቐொ,ቐແ֥ި؇ᄹࡆ,ٳ֥҃ӱ؇ᄹࡆ,ٚҵഈശb1Ⴗཁൕ֥൞ਆ۱ൈ௹ؓඔ۽ሧ֥ٳ໊ඔ(ቐᇠ)ࠣٳ໊ඔᆭҵ(Ⴗᇠ)bႷᇏႵਆ۱หᆘ٤ӈૼཁ:ֻ၂,t1ൈ௹৵ࢤ۲ٳ໊ඔོ֥ཌࢠt0ൈ௹۷ࡆొ,ᄝቐແٳ໊ඔࢠູࢤ֥࣍౦ঃ༯Ⴗແٳ໊ඔӁളऍնҵၳ,ᆃඪૼაt0ൈ௹ཌྷб,t1ൈ௹൬ೆٳ֥҃ӱ؇ഈശ;ֻؽ,ਆ۱ൈ௹֥ٳ໊ඔᆭҵ൞၂่ཟഈౠོ֥౷ཌ,္ඪૼ൬ೆٳ֥҃ӱ؇ᄝ҂؎ঔնbႮՎॖ,Ֆഈൗࡀ90୍սԚ֞2006,୍ݓসᆀ֥൬ೆٳ҂֩ಒႵཁᇷഈശb(ؽ)࢝ტ߭Бੱ֥э߄൝۴ऌЧ໓֥࣮ଢ֥,ࣉ၂҄ؓਆ۱ဢЧൈ௹֥࢝ტ߭Бੱࣉྛбࢠ,і1b表1高等教育劳动者比例与教育回报率变化样本数大学生比重非大学生对大学生对数平均工资数平均工资工资比对数工资方差t0时期58630 065 615 761 030 45t1时期46230 156 517 181 100 90!ႮႿଢభݓଽႨCHNSඔऌࣉྛ֥࣮ޓ؟,Ч໓҂ؓඔऌࣉྛབྷ༥ࢺകbۋྖ౿֥ᆀॖၛ๙ݖCHNSຩᅟ(http: projects china)ਔࢳ۷؟ྐ༏b%Ч໓ીႵ࿊౼1989୍֥ඔऌ,၂ٚ૫ၹ୍ູ֒ඔऌᇏݣႵࢠ؟ಌാაၳӈᆴ;ਸ਼၂ٚ૫ݓຓཌྷܱ໓ང֥࣮ࢲં,௴ђಪູ൬ೆҵएঔն֥གྷའႿഈൗࡀ90୍սԚb80
2010୍ֻ9௹1 ൬ೆٳ҃э߄൝ і1ᇏႵਆ۱൝ᆴ֤ᇿၩ:၂൞ᄝt1ൈ௹,ࢤ൳ۚ֩࢝ტ֥সᆀбᇗࢠt0ൈ௹Ⴕਔૼཁᄹࡆ,ఃбᇗႮ6%նږഈശᇀ15%;ؽ൞ᄝࢤ൳ۚ֩࢝ტসᆀбᇗᄹࡆ֥ൈ,ۚ֩࢝ტসᆀ֥ཌྷؓ۽ሧ္Ⴕૼཁഈശbᄝt0ൈ௹,ն࿐ളა٤ն࿐ള֥ؓඔ۽ሧбູ1 03,ਆᆀࠎЧӻ,ఃधؓඣԩႿؓඔ۽ሧٳ֥҃50%∃60%ٳ໊ඔᆭࡗbᄝt1ൈ௹,ն࿐ള֥ؓඔन۽ሧູ7 18,ԩႿؓඔ۽ሧٳ҃75%∃80%֥ٳ໊ඔᆭࡗ;ط٤ն࿐ള֥ؓඔन۽ሧູ6 51,ԩႿؓඔ۽ሧٳ҃35%∃40%֥ٳ໊ඔᆭࡗ,ਆᆀ֥ҵएૼཁᄹնbᆴ֤ᇿၩ֥൞,ᆃ֥࠹ෘીႵ॥ᇅ႕ཙ൬ೆ֥ఃၹ,ೂ۽ቔࣜဒb๙ӈಪູીႵն࿐࿐໊֥সᆀႚႵ۷Ӊ֥۽ቔࣜဒ,ೂݔࡼ۽ቔࣜဒჍၛ॥ᇅ,ᄵۚ֩࢝ტ֥൬ೆၮࡎӱ؇߶۷ۚbՎຓ,t1ൈ௹ؓඔ۽ሧ֥ٚҵູ0 9,ࠫެ൞t0ൈ௹֥ਆП,ᄜՑဒᆣਔᄝဢЧ௹ࡗଽݓসᆀ൬ೆҵएཁᇷঔնᆃ၂൙ൌb๙ݖ၂۱ࡥ߄֥২ሰ,ૌॖၛࡼభ໓૭ඍ֥ݓস৯൧ӆᄝ1991୍ᇀ2006୍ࡗӁള֥ਆնэ߄∃∃∃(1)ࢤ൳ۚ֩࢝ტসᆀ֥бᇗնږᄹࡆ,(2)ਆোসᆀ֥ཌྷؓ൬ೆҵए҂؎ঔն,აᆜ۱স৯൧ӆ֥൬ೆ҂֩ӱ؇(൬ೆ֥ٚҵ)ഈശ৳༢ఏটbࡌഡ൧ӆഈᆺႵۚඌ۽ದ(ն࿐ള)ა֮ඌ۽ದ(٤ն࿐ള)ਆোসᆀ,෮ᅝбᇗٳљູ ࠣ(1- );ն࿐ളა٤ն࿐ള֥۽ሧٳљູcyაncy,ఃᇏcncy>ybᄝՎࡌഡ༯,স৯൧ӆഈ֥न۽ሧ:ູE(y)= cy+(1- nc)y(1) সᆀ֥൬ೆٳ҂֩ӱ؇Ⴎ൬ೆ֥ٚҵޙਈ,ࡥֆ࠹ෘॖ֤:nc2nc2cnc2var(y)=E2(y)-[E2(y)]= (1- c)(y-y)=(y) (1- )(y y-1)(2)Ֆ(2)ൔॖၛ֤֞:nc2c var(y) =(1-2 )(y)(y nc2y-1)>0 (֒ <0 5)(3)cnc var(y) 2 (y ncy)=2(y) c(1- )(y ncy-1)>0(4) ഈඍ(3)ൔა(4)ൔ֥ࢲંᆞ൞ඔऌ෮іགྷԛ֥หᆘ:ۚඌ۽ದ֥бᇗഈശ(бᇗಯཬႿ50%)ၛࠣਆোসᆀ֥ཌྷؓ൬ೆҵएঔն,Ֆطᆜ۱সಕุ֥൬ೆ҂֩ӱ؇ᄹնb!!֒ಖ,ᄝᆃဢࡥ߄֥ࡌഡ༯,yc yncሱദࣼ൞ؓ൬ೆ҂֩ӱ؇֥ҩ؇b81
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩aିொཟඌࣉ҄ა൬ೆ҂֥֩ંଆିொཟඌࣉ҄ં֥ิԛႿݓຓ࿐ᆀؓૅݓস৯൧ӆࢲܒྟэ߄֥࣮bሱഈൗࡀ60୍սఏ,ૅݓস৯൧ӆԛགྷਔ൬ೆ҂֩ঔնaۚିস৯֥ཌྷؓ܂۳ᄹࡆၛࠣཌྷϴෛ֥࢝ტ߭Бੱഈശ֩ࢲܒྟэ߄,ିொཟඌࣉ҄ંᆞ൞ᄝᆃဢ֥Мࣟ༯ႋᄎطള,ѩູࢳᆃུэ߄ิ܂ਔંᆦӻbAcemoglu(1996,1998,2002a,2002b)֥၂༢ਙ໓ᅣؓඌࣉ҄ႄఏ֥൬ೆ҂֩ࠣఃቔႨࠏᇅࣉྛਔಆ૫ӂඍbAutoretal(2003)ࠎႿھં࣮ਔ࠹ෘࠏ֥௴ࠣؓ҂োস৯ླ֥႕ཙ,࣮іૼ࠹ෘࠏ֥௴ࠣބෛުࡎ֥۬༯ࢆᄹࡆਔؓ࠹ෘࠏ҂ॖูս֥স৯,Ⴍః൞ۚ࢝ტস৯֥ླbAutoretal(2006)֥࣮ࣉ၂҄іૼ,ඌࣉ֥҄ିொཟაۚିস৯ཌྷؓ܂۳ᄹ֥༯ࢆିޓݺࢳૅݓস৯൧ӆ֥ࠞ߄གྷའၛࠣ҂؎ഈശ֥൬ೆ҂֩bෙಖCard&DiNado(2002)བྷ༥ษંѩᆷԛਔିொཟඌࣉ҄ંაൌᆣࢲݔթᄝ၂ུ؛,ھଆಯಖ൞ࢳഈඍགྷའቋႵ৯֥ં۽ऎᆭ၂bЧ໓ᄝᆃ၂҆ٳࡹ৫၂۱ࠎႿିொཟඌࣉ֥҄၂Ϯनޙଆ,ѩၛՎࢣൕ໓ᅣֻؽ҆ٳ૭ඍ֥ݓস৯൧ӆ֥э߄หᆘb(၂)ଆഡק൮༵,ࡌഡࣜ࠶ᇏӌഅ֥ളӁݦඔູCESྙൔ:Y=[(Al( )L)+(Ah( 1 )H )] (0< &1)(5)ఃᇏ,Y൞Ӂԛ,L൞֮ିসೆ,H൞ۚିস,ೆAl( )ބAh( )ٳљսі֮ିসބۚିস֥ֆ໊ඌ߭Бੱ, սіຓള֥ඌඣ,ડቀ Al( ) >0, Ah( ) >0b!∋1 (1- ),ࠧۚିসა֮ିসࡗู֥սྟbӌഅ֥ളӁડቀቋႪ߄֥၂ࢨ่ࡱ,ႮՎ֤֞ਆোস৯֥ֆ໊۽ሧٳљ:ູw -1l=Al( )( L( -1[(Al( )L)+(Ah( )H )](6)w h=Ah( )(H -1([(Al( )L)+(Ah( )H -1)](7) ఃՑ,ࡌഡসᆀሹ൞ॖၛิ܂1ֆ໊֥֮ඌস,ิ܂֥ۚඌসඔਈ∀i౼थႿھসᆀ֥൳࢝ტӱ؇ၛࠣ࢝ტӮЧ,ࠧડቀ∀i=[f(edui)-ci])[0,1],ఃᇏfіൕหק൳࢝ტӱ؇༯ิ܂ۚඌস֥ି৯,Ⴕ f edui>0,ciູসᆀ۱ุ֥ၳᇉྟ൳࢝ტӮЧbসᆀᄝਆᇕস৯܂۳ٚൔࡗ࿊ᄴb֒∀i∗wl whൈ,সᆀ࿊ᄴิ܂∀iֆ໊֥ۚඌস;֒∀i<wl whൈ,࿊ᄴิ܂1ֆ໊֥֮ඌসb!ႮՎ,ਆᇕস৯֥ሹ܂۳ٳљ:ູsH=+∀i(I(∀i∗swl wh), L=+I(∀i<wl wh)(8)iiఃᇏ,I(()ູൕྟݦඔ,֒ওݼᇏ่ࡱડቀൈ౼1,ڎᄵ౼0brw∋wl wh,֮ࠧඌসაۚඌস֥ཌྷؓ۽ሧbᄝൔᇏ,ؓၩi,ཁಖႵ Hs s ∀i>0, L ∀si<0,ၛࠣ Hs rw<0, L rw>0bቋު,স৯൧ӆԛౢေHs=HsބL=LൈӮ৫bഈඍ၂Ϯनޙଆᇏቋᇗေ֥ࡌഡ൞:ຓള֥ඌࣉ҄ऎႵିொཟ,ྟࠧඌࣉ҄ؓۚඌস৯۷ࡆႵ০b%ᆃ၂ࡌഡᄝഈඍଆᇏॖၛіඍູ: [Al( ) Ah( )] <0(9) (ؽ)ඌࣉ҄a൬ೆ҂֩აস৯ཌྷؓ܂۳ିொཟඌࣉ҄ࡆखস৯൧ӆ൬ೆ҂֥֩ቔႨࠏᇅॖၛႮೂ༯֝ݖӱ֤:֞!ཁಖ,၂۱ႅݣ֥ࡌഡ൞wl<wh,ڎᄵࣜ࠶ᇏࡼ҂߶Ⴕসᆀิ܂ۚିসb%ު໓ࡼՖඔऌᇏဒᆣھࡌഡ֥ކྟb82
2010୍ֻ9௹rw∋wl wh=[Al( ) Ah( )](H 1-L)(10)(ؓ10)ൔਆш౼ؓඔ:ln(rw)∋ln(wl wh)= ln[Al( ) Ah( )]+(1- )ln(H L)(11)֩ൔ(11)ਆшൈؓ ֝: sln(rw) r(ww( Hs rww = Ah (Al Ah)Al +(1- r)( L rw))(12) =1r H-Lഈඍ֝Ⴈ֞ਔH=Hss,L=Lᆃ۱ࢲંbࣉ၂҄ᆜ֤֞:+ssHL-(1- )L H rwh)+(1- )H L(13)rw rw (rw =HL Ah (Al AAl +-++-ཁಖႵ rw <0bᆃіૼෛሢඌࣉ҄( ഈശ),֮ඌ۽ದ֥ཌྷؓ൬ೆࡨഒ(rw༯ࢆ),স৯൧ӆ֥҂֩ӱ؇ഈശbᄝՎࠎԤഈॖၛࣉ၂֤҄֞ିொཟඌࣉ҄ؓਆোস৯ཌྷؓ܂۳֥႕ཙ(Hs=Hs,L=L):--+-s (H L) =[( Hs rw)( rw )L-( L rw)( rw )H2] L>0(14) ᇀՎ,Ч໓Ֆଆᇏ֤֞ਔඔऌّ႘ԛ֥ਆ۱หᆘ∃∃∃ֻ၂,ۚඌস৯֥ཌྷؓ۽ሧഈശ,൬ೆҵएঔն;ֻؽ,აՎൈ,ۚඌস৯ཌྷؓ܂۳ᄹࡆb!ഈඍ֝ࢣൕਔস৯൧ӆ֥טᆜݖӱbିொཟඌࣉ҄ۚඌ۽ದ֥ཌྷؓ۽ሧഈശ,ᆃႄఏۚඌ۽ದ֥ཌྷؓ܂۳ᄹࡆ;ۚඌ۽ದཌྷؓ܂۳֥ᄹࡆࢆ֮ਔۚඌ۽ದ֥ཌྷؓ۽ሧ,ᄝ၂קӱ؇ഈߏࢳਔস৯൧ӆ֥൬ೆ҂֩b%൞,ᆃ۱טᆜݖӱ֥ቋᇔࢲݔಯಖ൞൬ೆٳ҂֩ӱ؇֥ഈശ,ჰၹᄝႿۚඌ۽ದཌྷؓ܂۳֥ᄹࡆ҂ିປಆҀႮႿඌࣉ҄ջট֥൬ೆҵएঔնb()࢝ტ߭Бა൬ೆ҂֩ഈ໓֥ٳ༅ॿࡏູႨ࢝ტ୍ཋቔູসᆀඌඣ֥ޙਈѓሙิ܂ਔᆦӻbᄝ໓ᅣ֥ଆᇏ,ႮႿ֮ඌসᆀ֥ቋႪ࿊ᄴ൞҂ࣉྛ࢝ტೆ(ࠇᆀඪ,ᄝགྷൌᇏᆺࣉྛࠎԤඣ֥࢝ტ)ೆ,طᆺႵۚඌসᆀࣉྛ࢝ტೆbၹՎ,൳࢝ტӱ؇൞সᆀඌඣ֥֩ࡎҩ؇,ؽᆀ֥ྐ༏൞ᇗކ֥bᄝᆃ۱ၩၬഈ,ॖၛϜ࢝ტඣुቔ൞ ৵࿃ ,ّ֥႘সᆀඌඣ֥ٳቆᆷѓb҂࢝ტඣসᆀ֥൬ೆҵए,ࣼ൞҂ඌඣসᆀ֥൬ೆҵएbᆃࢳथਔمՖඔऌᇏᆰࢤ്љۚିসᆀބ֮ିসᆀ֥໙ีbऎุط,ᄝഈඍଆॿࡏ༯,ૌॖၛࣉ၂҄ࡼ൬ೆҵएঔնაসᆀ֥൳࢝ტӱ؇৳༢ఏট,ٳࢳԛসᆀ֥൳࢝ტӱ؇ؓ൬ೆҵएঔն֥ਆᇕ҂ིႋb۴ऌభ໓ࡌഡ,ഠ߶֥न൳࢝ტӱ؇ॖၛႮೂ༯֩ൔޙਈ,:!∀∋+edui(I(∀i∗wl w-1h)=+f(∀i+ci)I(∀i∗rw)(15)iiᇿၩ֩ൔ(15)Ӯ৫֥ჰၹ൞:ႮႿসᆀሹ൞ॖၛิ܂1ֆ໊֥֮ඌস,ၹՎؓႿ∀i<wl wh֥সᆀط,ఃቋႪ࿊ᄴ൞҂ࣉྛ࢝ტೆ(edui=0)طᆰࢤՖ൙֮ඌসb֩ൔ(15)ਆҧؓrw֝,֤:֞!ᄝস৯൧ӆሹุ܂۳ඣ҂э֥౦ঃ༯,ᆃၩሢۚඌস܂۳ഈശ,֮ඌস܂۳༯ࢆbᆃ۱ࢲં္ॖၛႮHsaLsᆰࢤؓ ֤֝֞b%֩ൔ(11)ਆшؓln(H L)֝ॖ ֤lnrw ln(H L)=1- >0b,҂ാ၂Ϯྟ,ࡌഡࣜ࠶ᇏ֥সᆀሹਈູ1b83
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩ !∀ rw=+-1f∀i+ci I∀i∗rw rw<0(16)iၹՎႵ lnrw !∀=( rw !∀) rw<0bᆃඪૼᄝିொཟඌࣉ่҄ࡱ༯,൳࢝ტӱ؇֥ሹุഈശ( !∀ = !∀ rw( rw >0)ࡆखਔসᆀ֥൬ೆ҂֩bᆃ൞࢝ტၹؓ൬ೆ҂֩႕ཙ֥ሹིႋbൈ,൳࢝ტӱ؇֥ഈശၩሢ൧ӆഈۚඌসཌྷؓ܂۳֥ᄹࡆ,ೂభ໓෮ඍ,ᆃᄝ၂קӱ؇ഈߏބਔ൬ೆ҂֩: (Hs sL) !∀=[( Hs ss !∀)L-( L !∀)Hs] s2(L)+-s=[( Hs rw)( rw !∀s)L-( L rw)( rw !∀)Hs] s2(L)>0(17) Ч໓Ϝᆃ҆ٳིႋקၬູ࢝ტ֥ေࢲܒིႋbᆃဢ,ॖၛࡼຓള֥ିொཟඌࣉ҄ႄఏসᆀ൬ೆ҂֥֩ሹིႋٳࢳӮೂ༯ਆ۱აসᆀ൳࢝ტӱ؇ཌྷܱ֥҆ٳ:ሹིႋ(+)=࢝ტ֥ေࢲܒིႋ(-)+࢝ტ֥ေ߭Бིႋ(+)(18)ఃᇏ,ᆞݼіൕھིႋঔնਔসᆀ൬ೆ҂,֩ڵݼіൕھིႋࡨཬਔসᆀ֥൬ೆ҂֩b࢝ტ֥ေ߭Бིႋקၬູሹིႋა࢝ტࢲܒིႋᆭҵ,ّ႘֥൞Ⴎିொཟඌࣉ҄ջট֥ш࠽࢝ტ߭Бੱ֥ഈശb(ඹ)ିொཟඌࣉ֥҄թᄝྟႮႿЧ໓֥ં֝ၛࠣު૫֥ൌᆣ࣮ࡹ৫ᄝඌࣉ҄ऎႵିொཟྟᆃ၂ܱࡌഡഈ,ၹՎဒᆣᆃ၂ࡌഡ,ࠧ(9)ൔӮ৫,ऎႵٳᇗေ֥ၩၬbᄝࣉೆЧ໓֥ൌᆣ҆ٳభ,Վԩ၇ᅶAuthoretal(1998)ၛࠣAcemoglu(2002)֥ٚم,ᆣૼିொཟඌࣉ҄ᆃ၂่ࡱᄝݓ൞Ӯ৫֥bࢲކ(6)ൔა(7)ൔ,֤֞ೂ༯֩ൔ:(!-1) !Sh=whH wlL=(Ah Al)(H (!-1) !L)(19)ఃᇏ,!∋1 (1- ),Sh൞ۚඌস৯֥ཌྷؓ۽ሧٺحb๙ݖࡥֆэߐॖၛ֤֞:! (!-1)Ah Al=Sh (H L)(20)ႮႿཌྷؓ۽ሧٺحShაস৯֥ཌྷؓ܂۳H LॖၛՖඔऌᇏ֤,֞ၹՎॖၛ๙ݖ֩ൔ(20)࠹ෘԛAh AlᄝဢЧࡗଽ֥э߄౦ঃ,ೂ2෮ൕb2۳ԛਔ1991∃2006୍ՖCHNSඔऌᇏ࠹ෘԛ֥Ah Al֥ඔᆴbࢹAcemoglu(2002)֥ቓم,ٳљ౼!=1 4ၛࠣ!=2bॖၛुԛ,ᄝᆜ۱ဢЧࡗଽ,Ah Alӯགྷഈശ൝,ѩᄝ20ൗࡀ90୍սᇏ௹ၛުഈശ؇ૼཁࡆॹbႮႿඌࣉ҄აൈࡗ൞၂ᇁ,֥ᆃࣼဒᆣਔ (Ah Al) >0,ࠧ (Al Ah) <0,္ࣼ൞ିொཟ2 Ah AᄝဢЧ௹ࡗ֥э߄lඌࣉ҄ᄝݓ൞ಒൌթᄝ֥bᆃླေᇿၩ֥൞,ഈඍࢲં֥Ӯ৫ࠎႿᆃဢ၂۱ႅݣࡌഡ:ඌඣ ෛሢൈࡗ၍൞҂؎ഈശ,֥ࠧ t>0b!Ч໓ᇏ֥ඌඣ ॖၛࢳູসህႵඌ(labor specific)bෙಖ҂!٤ӈۋ྆େബ۠ದᆷԛᄝ໓ᅣᇏླေؓھႅݣࡌഡ֥ކྟࣉྛษંb84
2010୍ֻ9௹ିᆰࢤՖඔऌᇏဒᆣ t>0ᆃ၂่ࡱ,၂Ϯط,সህႵඌඣაಆေളӁੱ(TFP)൞ૡ్ཌྷܱ,֥ؽᆀӯᆞཌྷܱܱ༢bၹՎ,აݓಆေളӁੱཌྷܱ֥࣮ࢲંॖၛՖҧ૫ቑᆣھࡌഡ֥ކྟbݒ౩ຫaࡉम࿒(2005),ڰႧaϢ(2009),້༯ݚ(2009)ၛࠣਾѥ৶aౢћ(2009)ٳљҐႨಆݓaസࠩބ൧ࠩඔऌ,ႋႨ҂ٚم࣮ਔݓ҂ൈ௹֥TFPэ߄౦ঃ,ః࣮ࢲંनಪູݓ֥TFPෛൈࡗႵ҂ӱ؇֥ᄹӉbࣉ၂҄ॉ੮֞1991∃2006୍ࡗྐ༏ඌ֥ؿᅚބ࠹ෘࠏ֥௴ࠣؓসᆀসളӁੱ֥႕ཙ,ཁಖ t>0൞၂۱ކ֥ࡌഡbඹaࠎႿRIF݂߭ٳࢳ֥ൌᆣ࣮Ч໓֥ᆃ၂҆ٳࡼ๙ݖൌᆣٚمՖඔऌᇏဒᆣ(18)ൔbೂ໓ᅣֻ၂҆ٳ෮,֒భᄝ൬ೆ҂֩໙ีഈ֥ൌᆣ࣮,ఃᇏ၂۱ᇗေ֥ਵთ൞๙ݖ࠹ਈࣜ࠶࿐ٚمࡼ൬ೆ҂֩ٳࢳӮসᆀ۱ุหᆘ֥ေ܊ངbطЧ໓֥࣮ᇶေܱᇿ൳࢝ტӱ؇ᆃ၂۱ุหᆘؓসᆀ൬ೆ҂֥֩႕ཙbऎุটඪ,ླေᄝ॥ᇅఃၹ֥౦ঃ༯ဒᆣ:(1)সᆀन൳࢝ტඣ֥ิۚ෪ཬਔ൬ೆҵए;(2)࢝ტ߭Бੱ֥ิۚᄝ۷նӱ؇ഈࡆखਔ൬ೆҵए;(3)সᆀ൳࢝ტӱ؇ᆃ၂۱ุหᆘ֥э߄ᄝሹุഈࡆखਔ൬ೆ҂֩b(၂)Blinder OaxacaٳࢳაRIF݂߭ٳࢳBlinder Oaxacaٳࢳٚم൞ࠎႿཌྟ݂֥߭नᆴٳࢳbܱᇿэਈyູসᆀ֥ؓඔᄅन൬,ೆᄝཌྟ݂֥߭ࡌഡ༯Ⴕ:yt=xt−#+ut, E(ut|xt)=0, t={1,0}(21) Ч໓ܱᇿ൬ೆyᄝt0ൈ௹(1991∃1993୍)ၛࠣt1ൈ௹(2004∃2006୍)֥नҵၳDy=E(y1)-E(y0)ೂޅ൳֞ఃၹX֥႕ཙbս(ೆ21)ൔॖၛ֤:֞Dy=E(y1)-E(y0)=E(x1−)#1-E(x0−)#0=E(x1−)(#1-#*)+[E(x−)#*-E(x0−)#0]=E(x1−)(#1-#*)+[E(x−)-E(x0−)]#0+E(x1−)(#*-#0)(22)ఃᇏ,#*ູनޙ֥ေ߭Бੱ;∃S∋E(x1−)(#1-#*)ӫູေ֥Бԏིႋ,ّ႘֥൞ཌྷေࢲܒ༯Ⴎေ߭Бੱ҂ႄఏ֥൬ೆҵљ;∃X∋[E(x1−)-E(x0−)]#0ӫູေ֥ࢲܒིႋ!,ّ႘֥൞Ⴎေٳ҃ࢲܒ҂ႄఏ֥൬ೆҵљ;R=E(x1−)(#*-#0)൞҂ॖࢳ֥҆ٳbBlinder Oaxacaٳࢳ֥अཋྟᄝႿॉҳ֥ࣇࣇ൞൬ೆy֥नᆴᄝ҂ൈ௹֥ҵၳb۷མ֥ٚمႋھՖy֥ш࠽ٳ҃ݦඔԛؿ,࣮ေXࠣః߭Бೂޅ႕ཙyᄝ҂ൈ௹֥ٳ҃bFirpoetal(1996)षԷྟֹิԛॖၛ๙ݖܒᄯّ൙ൌሑ֥ٳ҃ݦඔ֥ٚمট࣮҂ေؓ൬ೆٳ֥҃႕ཙb෮໌ّ൙ൌሑ֥ٳ҃ݦඔ,ࠧt0ൈ௹ᄝऎႵაt1ൈ௹ཌྷ֥ေٳ҃X1ൈy0෮ऎႵ֥ٳ҃b࠺ّ൙ൌሑ֥ٳ҃ݦඔູF(yC),ॖၛुቔ൞၂۱ ܄ ֥ٳ҃,๙ݖбࢠF(y0)aF(y1)აF(yC)֥ҵљ,ॖၛ֤֞۲ေX֥э߄ؓyٳ҃э߄֥႕ཙb%Firpoetal(2007)ڿࣉਔః1996୍֥࣮Ӯݔ,ิԛਔ၂ᇕࠎႿRIF݂֥߭ྍٳࢳٚمbھٚم๙ݖࡹ৫RIFა۲ᇕ࠹ਈࡗ֥ؓႋܱ༢,ॖၛࡼၩ֥࠹ਈіൕӮఃэਈ֥ཌྟ႕,ࣉطႋႨোරႿ!ݓຓ໓ངᇏӫ∃Sູstructuraleffect,ӫ∃XູcompositioneffectbູਔьႿაభ໓၂,ѩ۷์࣍གྷൌݣၬ,Ч໓Ґ౼ਔ҂֥ଁٚൔb%ೂޅܒᄯّ൙ൌሑ֥ٳ҃ݦඔ,ၛࠣೂޅ֤֞۲ေX֥э߄ؓyٳ֥҃႕ཙ֥ऎุٚمࡼᄝ༯໓ӂඍbѯ֩(2007)္ؓھٚمࣉྛਔࢠູབྷ༥֥ࢺകbMachado&Mata(2005)০Ⴈٳ໊ඔ݂߭(quantileregression)֥ٚم༵֤֞y۳קX่֥ࡱٳ҃,ᄜ๙ݖbootstrapٚمؓXࣉྛԎဢটܒᄯy֥ш࠽ٳ҃,ՖطӮۿࡼေٳ҃ࠣః߭Бაy֥ш࠽ٳ҃৳༢ఏটbᆃӮູਸ਼၂۱ФܼٗႨ֥ٚمb85
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩Blinder Oaxaca֥ٳࢳٚم,ٳࢳԛ۲ࢳэਈؓھ࠹ਈ֥႕ཙbv=v(F)൞קၬᄝၩٳ҃ݦඔFഈ֥ٗݦbЧ໓ᇏvหᆷख़߂ٳ҃F(y)֥۲ᇕ࠹ਈ,Їওनᆴaٚҵaٳ໊ඔၛࠣࠎୄ༢ඔ֩by1აy0֥ٳ҃ҵၳॖၛႨ۲ᇕ࠹ਈটޙਈ,ࠧ:D=v(F1)-v(F0)=[v(F1)-v(FC)]+[v(FC)-v(F0)](23)ఃᇏFC(()൞ّ൙ൌሑ֥ٳ҃ݦඔbཌྷႋֹ,v(FC)൞קၬᄝFCഈ֥࠹ਈbၩٳ҃F(y)֥ᄜᇏྏ߄႕ཙݦඔRIFקၬູ:v(FR%)-v(F)IF(y;v)∋v(F)+IF(y;v)=v(F)+lim%.0%(24) Firpoetal.(2007)ᆣૼႵೂ༯֩ൔӮ৫:/RIF(y;v)(dF(y)=Ey(RIF(y;v))=v(F)(25) ๙ݖםս௹ຬჰ,ॖၛࡼ࠹ਈv(F)іൕӮఃэਈX֥ݦඔ,ࠧ:v(F)=Ey(RIF(y;v))=Ex[Ey(RIF(y;v)|X)]=E[mv(X)](26)ఃᇏEy(RIF(y;v)|X)∋mv(X)bmvL(X)=X(&v൞mv(X)ᄝXഈ֥ཌྟ႕,ཁಖႵ:&v=E(XXT-1)(E(XTmv(X))(27)Ⴎཌྟ႕֥ྟᇉॖၛ֤֞E[mvL(X)]=E[mv(X)]bࢲކൔ(23)a(26)ބ(27)ॖၛ֤֞:D=[E(mv1(X1))-E(mvC(X1))]+[E(mvC(X1))-E(mv0(X0))]=[E(XT1)((&v1-&vC)]+[E(XT1)(&vC-E(XT0)(&v0]TTTT=[E(X1)((&v1-&vC)]+[E(X1)-E(X0)](&v0+E(X1)(&vC-&v0)(28)(28)ൔა(22)ൔཌྷؓႋbၹՎ,ᄝ֤֞&vl(l=1,0,c)֥၂ᇁܙ࠹,ުॖၛႋႨაBlinder Oaxacaཌྷර֥ٚمটؓ҂ིႋࣉྛٳࢳbሹֹটඪ,ࠆ֤(28)ൔ֥ٳࢳࢲݔॖၛٳਆ҄ࣉྛb൮༵൞֤֞ଢѓ࠹ਈ(ٳ໊ඔaٚҵࠇࠎୄ༢ඔ)RIF֥၂ᇁܙ࠹!;֤֞RIF֥၂ᇁܙ࠹ު,&vl(l=1,0,c)֥၂ᇁܙ࠹ॖၛࡥֆֹႮࡆಃቋཬؽӰم֤,֞ॉ੮֞RIFܙ࠹֥༂ҵ,Firpoetal.(2007)ࡹၰႨWhiteၳٚҵ၂ᇁܙ࠹b(ؽ)ࠎႿRIF݂֥߭ٳࢳࢲݔЧ໓ᇏႨّ֥֞႘൬ೆyٳ֥҃࠹ਈЇওٚҵaࠎୄ༢ඔၛࠣٳ໊ඔbٚҵაࠎୄ༢ඔିᄝሹุഈّ႘ਆ۱ൈ௹൬ೆ҂֥֩э߄౦ঃ,ط҂ٳ໊ඔିᄝᆜ۱ٳ҃ഈख़߂y֥э߄౦ঃ,Ֆط֤֞۷ࡆ༥ᇁaࣚಒ֥ࢲݔb҂࠹ਈ֥RIF၂ᇁܙ࠹ॖၛ๙ݖ(2007)ࢺക֥ٚم֤֞bᄝֻؽࢨ؍֥RIF݂߭ᇏ,сྶ॥ᇅ႕ཙ൬ೆٳ֥ఃၹট֤֞൳࢝ტӱ؇ؓ൬ೆٳ֥႕ཙbऎุط,RIF݂߭ٚӱ֥ഡקೂ༯:v(Fl)∋Ey(RIF(yl;v))=mvL(X)+%=&0l+educ(&*l+Z&vl+% (l=0,1,c)ఃᇏeduc൞সᆀ֥൳࢝ტӱ؇,ҐႨ൳࢝ტ୍ཋটޙਈ%;Z൞႕ཙ൬ೆ֥ఃၹ,Їওসᆀ֥ྟљ(female,1ູ୯,ྟ0ູଳྟ)aӬཔࠡ(rural,1ູ୪ը१,0ູӬ൧१)aྟљა൳࢝ტӱ؇֥ࢌཛaࠡა൳࢝ტӱ؇֥ࢌཛa୍ਭࠣఃٚཛ(age,agesq)a܄ඳྟᇉྴэਈၛֹࠣთ(സٺ)ྴэਈ;%൞݂֥߭җҵb!Firpoetal.(2007)ࢺകਔ҂࠹ਈRIF֥ܙ࠹ٚم,ѩᆣૼਔRIFܙ࠹֥ࡶࣉᆞྟᇉbఃᇏၛٳ໊ඔູ࠹ਈ֥RIF္݂߭Ӯ่ູࡱٳ໊ඔ݂߭(unconditionalquantileregression,Firpoetal.(2009))bႮႿRIF(yl;v)֥ܙ࠹ݖӱࢠູگᄖ,ෛ࠹ਈ֥҂ط҂,Վԩ҂ؓఃܙ࠹ٚمࣉྛऎุษ,ંႵྖ౿֥ᆀॖၛҕॉFirpoetal.(2007)b%ڸAਙԛਔҐႨ߄࢝ტҩ؇֥RIFٳ໊ඔ݂߭ࠣٳࢳࢲݔ,აᆞ໓ᇏҐႨ৵࿃࢝ტҩ؇෮֤֥֞ࢲݔ൞၂ᇁ֥bۋ྆େബ۠ದ֥ᆃ၂ࡹၰb86
2010୍ֻ9௹1 эਈ૭ඍ࠹і2۳ԛਔt0(1991∃1993)୍ބt1(2004∃2006)୍ਆ۱ൈ௹ᇶေэਈ֥૭ඍ࠹ࢲݔbఃᇏႵਆ۱ԛ֥หᆘ:(1)সᆀ֥۽ሧ൬ೆႵਔૼཁഈശ,ఃؓඔ۽ሧनᆴՖt0ൈ௹֥5 62ഈശ֞t1ൈ௹֥6 61,ഈശږ؇ղ֞17 6%;ൈ,۽ሧ൬ೆ֥҂֩ӱ؇ཁᇷঔն,ఃؓඔ۽ሧ൬ೆ֥ѓሙҵՖt0ൈ௹֥0 67ഈശ֞t1ൈ௹֥0 95,ഈശږ؇ູ41 8%b(2)ਆ۱ൈ௹সᆀ֥۱ุหᆘܒӮႵਔૼཁэ߄bཌྷؓႿt0ൈ௹,t1ൈ௹ဢЧᇏ୪ըসᆀ֥бᇗ۷ն,Ֆ49%ഈശ֞57%;ൈসᆀ֥न୍ਭഈശਔ5 37ෟ,طन൳࢝ტ୍ཋഈശਔჿ1 5୍bഈඍসᆀ൬ೆඣa൬ೆҵएၛࠣ۱ุหᆘᄝ҂ဢЧൈ௹֥ҵၳູଆҕඔ֥ሙಒܙ࠹ิ܂ਔቀܔ֥эၳ(ྟvariation),္Ֆҧ૫ّ႘ਔݓস৯൧ӆ֥ऍնэ߄b表2主要变量描述统计变量名样本量均值标准差最小值最大值变量名样本量均值标准差最小值最大值1991∃1993年2004∃2006年lnwage58635 620 670 708 74lnwage46236 610 951 4010 06rural58630 490 500 001 00rural46230 570 500 001 00female58630 390 490 001 00female46230 400 490 001 00educ58638 963 540 0018 00educ462310 453 780 0018 00age586335 6910 5418 0264 95age462341 0610 5018 0364 89 2 RIFٳ໊ඔ݂߭ࢲݔႮႿ۲ٳ໊ඔିಆ૫ख़߂൬ೆٳ҃ሑঃ,ѩၛ൬ೆٳ֥҃ٳ໊ඔູФࢳэਈ֥݂߭ିᆰܴुԛ۲ေᄝ҂൬ೆඣ֥ш࠽߭Бੱ,ၹՎЧ໓༵۳ԛᄝ҂ൈ௹RIFٳ໊ඔ݂֥߭ࢲݔ,ऎุі3b表3RIF分位数回归结果时期1991∃1993年2004∃2006年分位数10%50%90%10%50%90%rural-0 280**-0 304**-0 171-0 481-0 572**-0 057(0 081)(0 046)(0 089)(0 335)(0 077)(0 104)female-0 256**-0 201**-0 396**-1 638**-0 122*-0 023(0 084)(0 044)(0 082)(0 297)(0 058)(0 082)educ0 0040 004-0 0050 0250 027**0 058**(0 007)(0 004)(0 009)(0 021)(0 005)(0 009)rural0educ0 0140 013**0 0030 0140 034**-0 019(0 008)(0 005)(0 009)(0 027)(0 007)(0 010)female0educ0 016*0 010*0 021**0 105**-0 005-0 011(0 008)(0 005)(0 008)(0 024)(0 005)(0 008)age0 089**0 029**0 062**0 177**0 050**0 014(0 010)(0 005)(0 008)(0 030)(0 007)(0 010)agesq-0 001**-0 000**-0 001**-0 002**-0 001**-0 000(0 000)(0 000)(0 000)(0 000)(0 000)(0 000)Constant3 091**4 955**5 044**3 426**6 288**7 271**(0 378)(0 185)(0 399)(0 673)(0 228)(0 457)公司虚拟变量有有有有有有地域虚拟变量有有有有有有样本个数586358635863462346234623R20 1000 1310 1180 2790 2750 095 ᇿ:ওݼᇏູWhite໗ࡲѓሙҵ**,ა*ٳљіൕᄝ1%ა5%֥ඣഈཁᇷ(ၛ༯ڸіA1)b87
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩Ֆі3ॖၛुԛ,ᄝt0ൈ௹,࢝ტؓসᆀ֥൬ೆᄝ࠹ഈીႵཁᇷ႕ཙ;ᄝt1ൈ௹,ᆃᇕ႕ཙэ֤ٳཁᇷ,ѩෛሢ൬ೆٳ໊ඔ֥ഈശھቔႨჟࡆૼཁbਸ਼၂ٚ૫,Ֆनඣഈु(50%ٳ໊ඔ),୪ըंթᄝ۷֥ۚ࢝ტၮࡎ(rural0educ֥༢ඔཁᇷູᆞ),ᄝt0ൈ௹֥नၮࡎӱ؇ჿູ0 013,ᄝt1ൈ௹֥नၮࡎӱ؇ჿູ0 034,ჿູt0ൈ௹֥2 5ПbഈඍࢲݔᄜՑّ႘ԛ࢝ტ߭БੱᄝဢЧ௹ࡗଽ֥ཁᇷഈശbՎຓ,і3֥ࢲݔߎіૼં൞t0ൈ௹ߎ൞t1ൈ௹,թᄝሢӬཔ൬ೆҵएၛࠣྟљ൬ೆҵए,ѩӬཔҵएᄝ൬ೆ֥ᇏ໊ඔඣഈႿঔն,t0ൈ௹༢ඔູ-0 304,t1ൈ௹༢ඔູ-0 572;ྟљҵएᄝ෪ཬ,t0ൈ௹ູ-0 201,t1ൈ௹ູ-0 122b3 RIFٳ໊ඔ݂߭༢ඔູਔ۷ಆ૫ਔࢳ࢝ტ߭Бੱᄝ҂ൈ௹ؓসᆀ൬ೆ֥႕ཙ,3߂ԛਔeducაrural0educਆ۱эਈᄝ҂ൈ௹҂ٳ໊ඔഈ֥RIF݂߭༢ඔb!Ֆቐॖၛुԛ,ᄝt0ൈ௹,࢝ტေԢਔᄝ֮ٳ໊ඔഈႵཁᇷ֥ᆞཟቔႨຓ,ؓ൬ೆࠫެીႵહ႕ཙ,݂߭༢ඔᄝ0ഈ༯ޓཬ֥ࡗଽѯ;ᄝt1ൈ௹,࢝ტႵૼཁ֥ᆞིႋ,ѩھၮࡎིႋෛሢ൬ೆٳ໊ඔ֥ഈശطᄹࡆb୪ըದ१֥حຓ࢝ტ൬ၭᄝt1ൈ௹္іགྷ֤ٳૼཁ(Ⴗ),ᄝۚٳ໊ඔഈෙಖ݂߭༢ඔູڵ,ᄝ࠹ഈѩ҂ཁᇷbᆃॖၛࢳູᆃᇕحຓ֥࢝ტ൬ၭ҂թᄝႿ୪ը֥ۚ൬ೆದಕ,ࠧ൬ೆᄝ85%ٳ໊ඔၛഈ֥ದಕb3 ࠎႿRIF݂֥߭ေིႋٳࢳࢲݔෙಖՖRIFٳ໊ඔ݂߭ॖၛुԛ҂ൈ௹࢝ტ൬ၭੱ֥э߄౦ঃ,ࠎႿ(28)ൔ֥RIF݂߭ٳࢳҌିّ႘۲ေؓ൬ೆ҂֥֩႕ཙ,ൈॖၛဒᆣ໓ᅣֻ҆ٳ֥ંყҩbі4ࣉ၂҄۳ԛਔܱႿ҂൬ೆ҂֩ҩ؇֥ࠎႿRIF݂֥߭ٳࢳࢲݔbֻ၂҆ٳਙԛ֥൞ࠎႿ෮Ⴕ݂߭эਈX=(educ,Z)֥ٳࢳࢲݔ,ֻؽ҆ٳਙԛ֥൞ა࢝ტཌྷܱ֥эਈ֥ٳࢳࢲݔbՖі4ᇏॖၛ֤֞ၛ༯ඹׄࢲં:ֻ၂,۲ᇕ൬ೆ҂֥֩ҩ؇∃∃∃ં൞ࠎୄ༢ඔaٚҵߎ൞ٳ໊ඔᆭҵ,ّ႘ԛݓসᆀ൬ೆҵएᄝ҂؎ঔնbਸ਼၂ٚ૫,Ֆ۲ေ֥ᆜุ႕ཙট,ुေБԏིႋनູᆞ,طေࢲܒིႋູڵ,ᆃඪૼ҂ေБԏੱ֥ഈശ൞ᄯӮ൬ೆҵएঔն֥ᇶေၹbбࢠ҂ٳ໊ඔᆭҵߎॖၛࣉ၂҄ؿགྷ,൬ೆҵएঔնᇶေ൞Ⴎഈҧٳ໊ඔაᇏ໊ඔ֥ҵएঘնᄯӮ֥b္ࣼ൞ඪ,ۚ൬ೆಕุაః൬ೆඣসᆀᆭࡗ൬ೆҵए֥ঔն൞ᄯӮᆜ۱সಕุ൬ೆҵएঔն֥ᇶေ!ႮႿэਈfemale0educᄝն҆ٳٳ໊ඔഈ҂ཁᇷ,Ч໓ᄝٳ༅ᇏᄠൈޭھࢌཛ֥႕ཙb88
2010୍ֻ9௹ჰၹbᆃဢ֥ؿགྷა൙ൌཌྷژ,္ّ႘ԛRIF݂߭ٳࢳ֥Ⴊ൝෮ᄝ,ࠧॖၛਲࠃֹܒᄯ҂֥࠹ਈ,ؓ໙ีࣉྛ۷ࡆധೆ֥૭ඍbֻؽ,ؓ൳࢝ტඣ֥RIF݂߭ٳࢳ֤֞ਔაЧ໓ֻ҆ٳ֥ંყҩۚ؇၂ᇁ֥ࢲݔ(֩ൔ(18))b൮༵,ؓႿ҂֥҂֩ҩ؇,࢝ტ֥ေБԏིႋಆູ҆ᆞ,ࠧঔնਔস৯ಕุ֥൬ೆ҂;֩ఃՑ,࢝ტ֥ေࢲܒིႋಆູ҆ڵ,ᆃඪૼসᆀन൳࢝ტӱ؇֥ิۚିᄝ၂קӱ؇ഈࢆ֮൬ೆҵए;ቋު,࢝ტ֥ሹིႋ္ູᆞ,ᆃّ႘ԛିொཟඌࣉ҄ۚିসᆀࠆၭ,ิۚਔۚିসᆀ֥ཌྷؓ۽ሧ,Ֆطঘնਔ൬ೆ҂֩bֻ,ࡼі4ֻؽ҆ٳֻ 表4RIF回归分解结果1ྛაֻ2ྛ֥ඔᆴٳљཌྷࡆ基尼系数方差Q50-Q10Q90-Q50Q90-Q10,֤֞൳࢝ტඣؓ୪ըসᆀ总效应0 0110 4300 0610 5900 650൬ೆҵए֥႕ཙbሹุ 要素报酬效应0 0150 4240 1380 6610 800ط 要素结构效应-0 0040 006-0 077-0 072-0 150,൳࢝ტඣႄఏ֥൬ೆҵएঔնؓ୪ը总效应সᆀ֥႕ཙࢠཬbeduc0 0420 8210 0650 5830 648ؓႿ൬ೆᇏ໊ඔၛ༯֥rural0educ-0 008-0 1370 106-0 226-0 120୪ըসᆀط,൳࢝female0educ-0 014-0 386-0 435-0 062-0 497ტӱ؇֥҂ّطิۚ要素报酬效应ਔૌ֥൬ೆ҂֩ӱeduc0 0570 9860 2510 7831 034؇rural0educ0 0110 1520 0940 0470 140(rural0educᄝ50%ٳ໊ඔა10%ٳ໊ඔfemale0educ-0 030-0 591-0 534-0 354-0 888ᆭҵഈ֥ሹིႋູ要素结构效应educ-0 043-0 650-0 240-0 376-0 6160 106)bֻඹrural0educ-0 019-0 2900 013-0 273-0 260,ॉ੮࢝ტ֥ྟљҵၳ,ࡼіfemale0educ0 0160 2040 0990 2920 3914ֻؽ҆ٳֻ1ྛაֻ3ྛ֥ ᇿ:Qx-Qyіൕx%ٳ໊ඔაy%ٳ໊ඔᆭҵ(ၛ༯ڸіA2)bඔᆴٳљཌྷࡆ,ॖၛु֞࢝ტඣؓ୯ྟসᆀ൬ೆ҂֥֩႕ཙေჹჹ֮Ⴟؓଳྟসᆀ֥႕ཙ,ѩᄝ50%ٳ໊ඔა10%ٳ໊ඔᆭҵഈࢆ֮ਔ୯ྟসᆀ֥൬ೆ҂֩ӱ؇,ఃඔᆴູ0 065-0 435=-0 37bູਔ۷ಆ૫ֹख़߂࢝ტေؓਆ۱ൈ௹൬ೆ҂֩э߄֥႕ཙ,ࡼ൳࢝ტӱ؇ᄝ҂ٳ໊ඔഈRIF݂߭ٳࢳ֥ࢲݔ૭ᄝ၂ᅦഈ,ೂ4෮ൕb൮༵൞࢝ტ֥ေࢲܒིႋbॖၛुԛ,Ԣਔᄝਆ؊֥ࠞᆴׄၛຓ,࢝ტ֥ေࢲܒིႋᄝ55%ٳ໊ඔၛ༯֥ቔႨູᆞ,ᆃіૼაt0ൈ௹ཌྷб,ေࢲܒིႋᄝt1ൈ௹ิശਔ༯ҧٳ໊ඔbაՎཌྷ,ّᄝ55%ٳ໊ඔၛഈ,࢝ტ֥ေࢲܒིႋູڵ,ᆃіૼაt0ൈ௹ཌྷб,࢝ტ֥ေࢲܒིႋᄝt1ൈ௹ࢆ֮ਔഈҧٳ໊ඔbሹ֥ࢲݔ൞ഈ༯ҧٳ໊ඔ֥एࡨཬ,൬ೆٳ҃ཟनᆴࠢᇏ,Ֆطӱ؇ࡨཬ,ٚҵэཬ,ࠧ҂֩ӱ؇༯ࢆbఃՑ൞࢝ტ֥ေБԏིႋbა࢝ტ֥ေࢲܒིႋႵປಆཌྷّ֥ቔႨٚཟ:ᄝ༯ҧٳ໊ඔི֥ႋູڵ,ᄝഈҧٳ໊ඔི֥ႋູᆞbᆃဢ,࢝ტ֥ေБԏིႋᄝശഈҧٳ໊ඔ֥ൈঘ֮ਔ༯ҧٳ໊ඔ,൬ೆٳ҃э֤۷ࡆ,ࡆնਔ൬ೆ҂֩ӱ؇bਆᇕིႋཌྷןࡆ,֤֞࢝ტؓ൬ೆٳ֥҃ሹིႋbॖၛुԛԢࠞཬᆴׄຓ,࢝ტ֥ሹིႋᄝ༯ҧٳ໊ඔഈ֥ቔႨູᆞնཬࠎЧЌӻ҂э;ᄝഈҧٳ໊ඔഈ֥ቔႨᇯࡶ־ᄹ,൬ೆٳ֥҃Ⴗҧӱ؇ࡆն,Ֆطঔնਔ89
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩൬ೆ҂֩bᆃ္აі4֥ࢲં၂ᇁ,ࠧ࢝ტေᇶေ൞๙ݖঔնۚ൬ೆಕุ֥აః൬ೆඣসᆀ֥൬ೆҵए,ࡆखਔস৯ಕุ֥ᆜุ൬ೆ҂֩ӱ؇bཁಖ,4ࣉ၂҄ဒᆣਔ֩ൔ֥ંყҩ,ѩіૼᄝᆜ۱൬ೆٳ҃ഈ,࢝ტ֥ေࢲܒིႋaေБԏིႋၛࠣሹིႋაંყҩཌྷژba࣮ࢲંЧ໓ᄝିொཟඌࣉ֥҄ॿࡏ༯,๙ݖંٳ༅іૼିொཟ4 ࢝ტᄝ҂ٳ໊ඔഈ֥ေٳࢳඌࣉ҄ିޓݺֹࢳඔऌᇏіགྷԛ֥ۚିসᆀб২ᄹࡆაۚିসᆀཌྷؓ൬ೆഈശൈթᄝ֥གྷའbࣉ၂҄,ֹ๙ݖႋႨFirpoetal.(2007)ؿᅚԛ֥ࠎႿRIF݂֥߭ٳࢳٚم,ဒਔቔّູ႘সᆀି֥൳࢝ტӱ؇ؓসᆀ൬ೆ҂֩э߄֥႕ཙb໓ᅣ֥ൌᆣٳ༅֤֞ਔބંყҩۚ؇၂ᇁ֥ࢲݔ:সᆀन൳࢝ტӱ؇֥ิۚࢆ֮ਔ൬ೆ҂,֩ିொཟඌࣉ҄ಏᄝ۷նӱ؇ഈิۚਔ࢝ტ֥ш࠽൬ၭ,ੱቋᇔ֥ࢲݔಯಖ൞൬ೆ҂֩ഈശb࣮ࢲݔߎіૼ,Ֆ1991୍֞2006,୍সᆀ൬ೆ҂֥֩ঔնऎႵ҂ؓӫྟb၂ٚ૫іགྷᄝ൬ೆ҂֥֩ഈശᇶေ൞ႮႿԩᄝ൬ೆनඣၛഈ֥ۚ൬ೆসᆀაᄝ൬ೆनඣၛ༯֥֮൬ೆসᆀᆭࡗ൬ೆҵएঔնႄఏ֥,აՎൈ֮൬ೆসᆀᆭࡗ֥൬ೆҵएીႵૼཁ֥э߄bਸ਼၂ٚ૫ุགྷᄝӬཔҵљഈ,֮൬ೆ୪ըসᆀ֥൬ೆ҂֩ӱ؇൳֞࢝ტေ֥႕ཙбཌྷ൬ೆඣ֥Ӭ൧সᆀ۷ն,طۚ൬ೆ୪ըসᆀ֥൬ೆ҂֩ӱ؇ࢠ֩൬ೆඣ֥Ӭ൧সᆀ൳࢝ტေ֥႕ཙ۷ཬbЧ໓֥၂۱ᆟҦݣၬ൞,Ⴟ1999୍֥ۚঔᅱॖିᄝ၂קӱ؇ഈࢆ֮ਔসᆀ֥൬ೆ҂֩ඣbႮႿۚঔᅱࢆ֮ਔࢤ൳ۚ֩࢝ტ֥ӮЧ,ൈအਔ۷؟ିՖ൙ۚିস֥সᆀ,Ֆط఼߄ਔ࢝ტေ֥ࢲܒིႋ,൬ೆ҂֩ඣ༯ࢆb֒ಖ,႕ཙ൬ೆ҂֩э߄֥ၹޓ؟,Ч໓ᄝൌᆣݖӱᇏؿགྷ,ྟљ൬ೆҵएaসᆀ୍֥ਭࢲܒ֩ؓ൬ೆ҂֥֩э߄Ⴕᇗေ֥႕ཙ,ః႕ཙӱ؇ა࢝ტေ֥ඔਈࠩٳࢤ࣍bၹՎ,ೂޅᄝ၂۱၂֥ંॿࡏ༯۷ࡆಆ૫ֹࢳᆃུགྷའ൞၂۱ᇗေ֥࣮ٚཟbҕॉ໓ངӾະ,2008:∀ᇏݓӬ൧ྟљ൬ೆҵए࣮#,∀࠹࣮#ֻ8௹b֬౷ޚ,2007:∀Ӭᆓंაੀದ१֥൬ೆҵၳ∃∃∃ࠎႿOaxaca BlinderބQuantileٚم֥ٳࢳ#,∀ᇏݓದ१॓࿐#ֻ2௹bڰႧaϢ,2009,∀ᇏݓڿ۪ष٢ၛটಆေളӁੱэࠣఃٳࢳ(1978∃2006୍)#,∀ࣁವ࣮#ֻ7௹bݒ౩ຫaࡉम࿒,2005,∀ᇏݓಆေളӉੱ֥ܙෘ:1979∃2004#,∀ࣜ࠶࣮#ֻ6௹bޠྖࡹaࣁӄ,2007:∀ਆࠞٳ߄ҩ؇ٚمඍაᇏݓं൬ೆਆࠞٳ߄#,∀ࣜ࠶࣮#ֻ11௹bਾѥ৶aౢћ,2009,∀ᇏݓӬ൧ಆေളӁੱ֥ൌᆣٳ༅:1990∃2006#,∀ଲषࣜ࠶࣮#ֻ3௹bઅԣਊaൌ,2007:∀ದ৯ሧЧaྛြหᆘა൬ೆҵए∃∃∃ࠎႿֻ၂Ցಆݓࣜ࠶௴Ұሧਘ֥ࣜဒ࣮#,∀ܵൗࢸ#ֻ10௹bѯaӾະaჅ౬઼,2007:∀၂ᇕྍ֥൬ೆҵए࣮֥࠹ਈٚم∃∃∃ࠎႿٳ҃ݦඔ֥϶ҕඔ߄ܙ࠹#,∀ඔਈࣜ࠶აඌࣜ࠶࣮#ֻ8௹b90
2010୍ֻ9௹ૅဇ,2005a:∀ᇏݓӬ൧স৯൧ӆഈ֥ྟљ۽ሧҵၳ#,∀ࣜ࠶࣮#ֻ12௹bૅဇ,2005b:∀Ӭ൧স৯൧ӆഈ֥ࣼြࠏ߶ა۽ሧҵၳ∃∃∃ຓটস৯ࣼြაБԏ࣮#,∀ᇏݓഠ߶॓࿐#ֻ5௹b້༯ݚ,2009,∀સၞष٢aದ৯ሧЧაᇏݓಆေളӁ#ੱ,∀ඔਈࣜ࠶აඌࣜ࠶࣮#ֻ7௹b༱౩ࢭaසৡଫ,2007:∀ᇏݓӬᆓ֥э߄൝ބଆൔ:1988∃2002#,∀ࣜ࠶࣮#ֻ9௹bྷགྷབaݚۖ,2008:∀ݓԚՑٳᇏ֥ਆࠞٳ߄ࠣӮၹ#,∀ࣜ࠶࣮#ֻ2௹bိ༵ݓaཫ,2007:∀۽ሧ҂֥֩ഈശ:ࢲܒིႋაࡎི۬ႋ#,∀ᇏݓದ१॓࿐#ֻ1௹bAcemoglu,D.,1996, AMicrofoundationforSocialIncreasingReturnsinHumanCapitalAccumulation ,QuarterlyJournalofEconomics,111,p779∃,D.,1998, WhydoNewTechnologiesComplementSkills?DirectedTechnicalChangeandWageInequality ,QuarterlyJournalofEconomics,113,p1055∃,D.,2002a, TechnicalChange,InequalityandtheLaborMarket ,JournalofEconomicLiterature,40,p7∃,D.,2002b, DirectedTechnicalChange ,ReviewofEconomicStudies,69,p781∃,D.,Katz,L,.Krueger,A.,1998, ComputingInequality:HaveComputersChangedtheLaborMarket? ,QuarterlyJournalofEconomics,113(4),p1169∃1213Autor,D.,Levy,F.,Murnane,R.,2003, TheSkillContentofRecentTechnologicalChange:AnEmpiricalExploration ,QuarterlyJournalofEconomics,118(4),p1279∃,D.,Katz,L.,Kearney,M.,2006, ,AmericanEconomicReview,PapersandProceedings,96(2),p189∃.,Dinardo,J.,2002, Skill BiasedTechnologicalChangeandRisingWageInequality:SomeProblemsandPuzzles ,JournalofLaborEconomics,20(4),p733∃,J.,Fortin,N.,Lemieux,T.,1996, LaborMarketInstitutionsandtheDistributionofWages,1973∃1992:ASemiparametricApproach ,Econometrica,64,1001∃,S.,Fortin,N.,Lemieux,T.,2007, DecomposingWageDistributionUsingRecenteredInfluenceFunctionRegression ,,S.,Fortin,N.,Lemieux,T.,2009, UnconditionalQuantileRegression ,Econometrica,77(3),953∃,Chinhui,,1993, WageInequalityandtheRiseinReturnstoSkill ,JournalofPoliticalEconomy,101(3),410∃,F.,A.,Mata,J.,2005, CounterfactualDecompositionofChangesinWageDistributionsUsingQuantileRegression ,JournalofAppliedEconometrics,20,p445∃465.ڸڸіA1ބA2ਙԛਔҐႨ߄࢝ტҩ؇֥RIFٳ໊ඔ݂߭ၛࠣٳࢳࢲݔbοᅶݓଽຓཌྷܱ໓ང֥ԩܸ২,ૌࡼসᆀ֥൳࢝ტӱ؇߄,ഡᇂ ն࿐ a ۚᇏބᇏህ a Ԛᇏ ၛࠣ ཬ࿐ࠣၛ༯ ඹ۱ྴэਈbসᆀ֥҂൳࢝ტӱ؇ّ႘ਔః҂֥ඌඣbࡼ ཬ࿐ࠣၛ༯ ቔູࠎሙቆ(֮ඌসᆀ),ఃྴэਈ֥༢ඔࠣRIF݂߭ٳࢳࢲݔࣼିᆰֹّܴ႘ԛ(҂ඌඣ֥)ۚඌ۽ದބ֮ඌ۽ದ֥ཌྷؓ۽ሧҵएbՖіA1ބA2֥ൌᆣࢲݔॖၛुԛ:(1)ა৵࿃࢝ტҩ؇֥ࢲݔཌྷ,RIF่ࡱٳ໊ඔ݂߭ࢲݔཁൕ,ཌྷؓႿཬ࿐ࠣၛ༯൳࢝ტቆط,ն࿐ބۚᇏቆ֥সᆀᄝt0ބt1ൈ௹नႵૼཁ֥࢝ტၮࡎ,ѩ࢝ტ߭Бᄝt1ൈ௹҂ٳ໊ׄഈбt0ൈ௹Ⴕૼཁഈശb൞,ᆃဢི֥ݔᄝ Ԛᇏ ቆᇏѩ҂ཁᇷ,іૼཌྷؓႿཬ࿐ࠣၛ༯൳࢝ტӱ؇֥সᆀط,ऎႵԚᇏ൳࢝ტӱ؇֥সᆀીႵૼཁ֥࢝ტၮࡎbᆃဢ֥ࢲݔ္൞აᆰतཌྷژ֥b(2)ᄝ҂֥҂֩ҩ؇༯(ࠎୄ༢ඔaٚҵaٳ໊ඔҵ),҂൳࢝ტӱ؇(္ࣼ൞҂ඌඣ)֥সᆀࠎႿRIF֥ٳࢳࢲݔაંყҩ၂ᇁ∃∃∃࢝ტ֥ေࡎི۬ႋູᆞ,ေࢲܒིႋູڵ(Ԛᇏቆ֥ေࢲܒིႋᄝ50%∃10%ٳ໊ඔഈູᆞ,൞၂۱২ຓ),ѩሹིႋູᆞbᆃაჰ໓ᇏႨ৵࿃࢝ტҩ؇ࣉྛٳࢳ֥ࢲݔ൞၂ᇁ֥b(3)ՖRIFٳࢳࢲݔ֥ඔᆴնཬഈुbն࿐࢝ტაۚᇏ࢝ტᄝေࡎི۬ႋഈ֥նཬնᇁཌྷ֒,ؽᆀनཁᇷۚႿԚᇏ࢝ტbᄝ50%ٳ໊ඔა10%ٳ໊ඔᆭҵഈ,ն࿐࢝ტ֥ေࡎི۬ႋေૼཁնႿۚᇏ࢝ტbᄝ࢝ტ֥ေࢲܒིႋഈ,ն࿐࢝ტ֥ေࢲܒིႋᄝधؓඔᆴഈေնႿۚᇏ࢝ტބԚᇏ࢝ტb91
ྷ ൽ:ඌࣉ҄a࢝ტ൬ၭა൬ೆ҂֩表A1离散教育测度的RIF分位数回归结果1991∃1993年2004∃2006年分位数10%50%90%10%50%90%rural-0 156**-0 182**-0 145**-0 264**-0 185**-0 257**(0 031)(0 018)(0 036)(0 080)(0 027)(0 042)female-0 114**-0 112**-0 209**-0 588**-0 189**-0 167**(0 031)(0 017)(0 030)(0 081)(0 023)(0 032)college0 220**0 166**0 0600 218*0 357**0 613**(0 046)(0 040)(0 070)(0 097)(0 038)(0 071)highschool0 093**0 0270 0130 249**0 136**0 116**(0 034)(0 020)(0 039)(0 090)(0 029)(0 038)juniorschool-0 104-0 1130 084-0 8410 062-0 067(0 162)(0 082)(0 180)(0 920)(0 144)(0 180)age0 088**0 029**0 061**0 168**0 052**0 019*(0 010)(0 005)(0 008)(0 030)(0 007)(0 010)agesq-0 001**-0 000**-0 001**-0 002**-0 001**-0 000(0 000)(0 000)(0 000)(0 000)(0 000)(0 000)Constant3 131**4 981**5 022**3 932**6 433**7 464**(0 374)(0 181)(0 393)(0 620)(0 218)(0 445)公司虚拟变量有有有有有有地域虚拟变量有有有有有有样本个数586358635863462346234623R20 0990 1270 1170 2710 2720 105 ᇿ:collegeahighschoolބjuniorschoolٳљ൞ն࿐aۚᇏࠣᆯህބԚᇏ൳࢝ტӱ؇ྴэਈ(ၛ༯іA2)b表A2离散教育测度的RIF回归分解结果基尼系数方差Q50-Q10Q90-Q50Q90-Q10总效应0 01080 39430 04460 52320 5403 要素报酬效应0 01270 00910 01580 49220 1801 要素结构效应-0 00190 38520 02880 03100 3602总效应college0 00600 09760 09080 08260 1733highschool0 00800 10870 05450 09920 1899juniorschool0 00010 00130 00520 00210 0074要素报酬效应college0 00710 11210 09590 09270 1885highschool0 00860 12840 05940 10020 1957juniorschool0 00010 00130 00520 00290 0081要素结构效应college-0 0011-0 0135-0 0051-0 0101-0 0152highschool-0 0006-0 0097-0 0049-0 0010-0 0059juniorschool-0 0000-0 00000 0000-0 0008-0 0007(༯ሇֻ108်)92
ေaᅵ ಅ:ۚঔᅱაն࿐иြളࣼြHigherEducationExpansionandEmploymentofUniversityGraduatesWuYaowuandZhaoQuan(InstituteofPopulationandLborEconomics,CASS;SchoolofEconomics,BeijingTechnologyandBusinessUniversity)Abstract:UsingDID(DiffinDiffs)methodanditsextendedform(DiffinDiffinDiffs)undera treatment∃control framework,thispaperevaluatestheimpactsofhighereducationexpansionpolicy,introducedin1999,ontheperformanceofcollege universitygraduates,increasedtheirunemploymentrateandbroughtdowntheirwagerate,,:NaturalExperiment;DiffinDiffinDiffs;LaborForceParticipation;Unemployment;WageJELClassification:I21,J64(责任编辑:松 木)(校对:昱 莹)(ഈࢤֻ92်)TechnicalProgress,ReturntoEducationandIncomeInequalityinChinaXuShu(ResearchInstituteofEconomicsandManagement,SouthwesternUniversityofFinanceandEconomics)Abstract:ThispaperestablishesageneralequilibriummodelbasedontheframeworkofSkill ,,,ourempiricalresultsindicatethatthemodel :IncomeInequality;Skill BiasedTechnicalChange;RIFRegressionJELClassification:C14,D33,I21,J31(责任编辑:松 木)(校对:昱 莹)108
