中国地区收入差距_全要素生产率及其收敛分析

中国地区收入差距全要素生产率及其收敛分析*彭国华(510275):本文在测算和分析19822002年省区全要素生产率(TFP)的基础上进行TFP的收敛检验,并与收入(劳均GDP)的收敛模式做了对比分析分析表明TFP解释了我国省区收入差距的主要部分;我们应用OLSPanelData固定效应估计方法和DynamicPanelData(一阶差分GMM)方法进行收敛检验,得出的结论是:TFP与收入的收敛模式具有很大的相似性,全国范围内没有绝对收敛,只有条件收敛,但是TFP的收敛速度明显高于收入的收敛速度三大地区中只有东部地区存在俱乐部收敛现象而且,我国TFP与收入的收敛模式与世界范围的TFP和收入的收敛模式具有很大的相似性:全要素生产率收入差距收敛*19702002!,JonathanTempleSteveDowrick,,,,,(TFP),ChowandLin(2002)(2003),TFPTFP,TFPYoung(1995)TFPEasterlyandLevine(2001),TFPPrescott(1998),TFPTFPBernardandJones(1996)OECDTFP;MillerandUpadhyay(2002)TFP,,TFP,,,TFPTFP1982200228TFP,TFP,TFPOLSPanelDataDynamicPanelData(GMM)TFP,(GDP):,TFP1920059,,;TFP,TFP13%169%,48%73%,∀TFP(),;,TFP:TFP,TFP,TFP()TFP,HallandJones(1999)BilsandKlenow(2000),CobbDouglas,Yi=Ki(AiHi)1-(1)iYi,KiHiAiAiTFP,Ei,Hi:Hi=e(Ei)Li(2)#(Ei),#(Ei)f(0)=0,E=0,H=e(0)L=L,y=YL,h=HL,HallandJones(1999),:yi=AiKY1-ihi(3)TFPTFPGDP,:28,,,,∃19822002GDP19821998!,19992002:19962002!,GDP,1978,1978,(2004)19821998!,19992002:19962002!1995:19521995!,199620:∀∃1982(),(),()2002!(2004)21970a04,,19821998!,19992002!!!198219902000,19871993199515198319861988198919911992,,()%&,1563335,(2),,,GeorgePsacharopoulos,Psacharopoulos(1994)Psacharopoulosetal(2004),018,0134,0151,,(E),∀06018,6120134,120151,,,,,,()TFP(3)2819822002TFP,TFP59%1984,123%;19921994,10%19891990TFP,∃,1982200282%,19841992199411%,1989199025%,TFP,TFPTFP,,,2002,10%12002TFP2002,2002TFP,,TFP,TFP,HallandJones(1999),(3)20022120059∀∃(2003)TFP,125,lnh=018∋(6)+0134∋(6)+0151∋(05)=19612002TFPTFP1,,1,1,,TFP,,,18%,65%,TFP7%,TFP,,76%,200223%,122,80%,TFP,24%,,,TFP(Counterfactual):,TFP,95%;TFP,,8,KY22(KY100),,TFP表12002年(部分)省区收入差距的分解YL(KY)1-HLA北京050212110021上海100100100100江苏036129075037湖北021101070029广东030101083036贵州006118065007平均值023122078024标准差0191033601110186与YL的相关系数1000010707430885:1,;21,TFP,0885,0743,0107,TFPTFP,TFPKlenowandRodriguezClare(1997),X:Xi=KY1-ihi(4)(3),,lnyi=lnAi+lnXi(5)lnylnA,,22:var(lny)var(lny)=cov(lny,lnA)var(lny)+cov(lny,lnX)var(lny)=1(6)lnXlnAlny,1,XTFP表2部分年份的方差分解198219861990199419982002平均cov(lny,lnA)var(lny)088077068070073080075cov(lny,lnX)var(lny)012023032030027020025:X(4),198220022,TFP,199068%198288%,75%,25%:TFP,TFP,?TFP?,TFP():()!!,!!,!∀,gclny0g=c+lny0+∀,lny0,,,,,,,,()(),,2%,,,,2%35,2%,35()2320059∀Quah(1996)70(Romer,2001,p25)TFPPanelData,MillerandUpadhyay(2002)∀,,,x,(),g=c+lny0+#x+∀,x,PanelData,,,xx,FDIDynamicPanelDataGMMSystemGMM,Casellietal(1996)Bondetal(2001),,PanelData,DynamicsPanelDataGMM,,PanelData,,Islam(1995)5,RiveraandCurrais(2004)4,5,419821985,19861989,19901993,19941997,19982001,lnyitlnAit,i=1,2,(,28,t=1,2,(,5,lnyi2i19861989lny,lnATFP()!2(a)lny2(b)lnATFP!2(a),!,19912000,!,,2(b)TFP,TFP19821991!,lnA1990,,24:∀MillerandUpadhyay(2002)PanelData,2(a)2(b),TFP,,!,,,!,,TFP,:(lny5-lny1)16=+ylny1+∀(7)(lnA5-lnA1)16=+AlnA1+∀(8)t=5t=11998200119821985,16,16∃Mankiw,RomerandWeil(1992):%=16=-(1-e-∃%)%(9)表3收入和TFP的绝对收敛检验,OLS回归全国东部中部西部y0010**-0011*-00140006(0004)(0005)(0012)(0016)R2001702000036-0102see0007001000100013Implied∃-000900120015-0006A0005-0015*∀00050013(0006)(0008)(0015)(0017)R2-00210051-0153-0081see0016001900120013Implied∃-00040018-0005-0012:1(7)(8)OLS,,∃(9),%=16;2*10%**5%***1%;3∀TFPP0103,10%3(7)(8)OLS,,,,09%TFP,,10%,12%,TFP,10%(P0103)!,,TFP1%,,():PanelDatad(lnyt)=lnyt-lnyt-1=+ylnyt-1+∀t(10)d(lnAt)=lnAt-lnAt-1=+AlnAt-1+∀t(11)t=1,2(5,∃:2520059=-(1-e-∃%)(12),,,(Wooldridge,2002),Islam(1995),MillerandUpadhyay(2002),,PanelData,PanelData,,MillerandUpadhyay(2002)表4收入和TFP的条件收敛检验,PanelData固定效应回归全国东部中部西部y-0174***-0363**-0179**-0322***(0062)(0149)(0080)(0099)R20699077708640580see0061005000560053Implied∃0048011300490097A-0491***-0509***-0744***-0483***(0083)(0093)(0245)(0168)R20547075506440387see0093008000870075Implied∃0169017803410165:1(10)(11)PanelData,;2∃(12),%=4(4);3*10%**5%***1%4,,5%,TFP,TFP48%,146,TFP169%,41():DynamicPanelDataGMM(2Step)PanelData,,Mankiw,RomerandWeil(1992)Islam(1995)Casellietal(1996)Bondetal(2001),:d(lnyt)=lnyt-lnyt-1=ylnyt-1+lnst+#ln(n+g+d)t+∀t(13)st()(n+g+d)t(++),st(+)GDP,nt,g+d005,st(n+g+d)t19821998stnt!,19992002!lnAt(13)lnytTFP:d(lnAt)=lnAt-lnAt-1=AlnAt-1+lnst+#ln(n+g+d)t+∀t(14),st(n+g+d)t(13)Yi=Ki(AiHi)1-,(KH)(KHL),Mankiw,RomerandWeil(1992)ISolow,26:IIAugmentedSolow,(13)(14)I(7),,SolowKHAugmentedSolowKHL,,,TFP,,DynamicPanelDataGMMBondetal(2001),(13)(14),,(13)lnyt-2,lnst-2,ln(n+g+d)t-2,(14)lnAt-2,lnst-2,ln(n+g+d)t-25表5收入和TFP条件收敛检验,DynamicPanelData一阶差分GMM(2Step)回归全国东部中部西部y-0253***-0147*∀-0184**-0288***(0077)(00870)(0069)(0098)lnst1334***0779**1308*0968**(0320)(0370)(0758)(0367)ln(n+g+d)t-0483**-0450-0409-0476**(0195)(0297)(0369)(0200)see0186019000950121SarganStatistic193634637047PValue5888964950092540Implied∃0073004000510085A-0406***-0227*-0298*-0485***(0126)(0122)(0171)(0099)lnst1818***1138**2067***0909(0480)(0467)(0549)(0559)ln(n+g+d)t-0222-0490-0195-0401*(0372)(0396)(0452)(0208)see0247026602300145SarganStatistic526750549207PValue153857513905600Implied∃0130006400890166:1(13)(14)DynamicPanelDataGMM(2Step),,(Eviews50);2(13)lnyt-2,lnst-2,ln(n+g+d)t-2,(14)lnAt-2,lnst-2,ln(n+g+d)t-2;3∃(12),%=4(4);4*10%**5%***1%;5SarganNullHypothesis;6∀TFP