CO2 emissions and economic activity : heterogeneity across countries and non stationary series

CO2 emissions and economic ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies Ma ías Piaggio Emilio Padilla 10.09 Facul a d'Economia i Emp esa De p a amen d'Economia A p licada Aques documen pe any al Depa amen d'Economia Aplicada. Da a de publicació : Depa amen d'Economia Aplicada Edi ici B Campus de Bella e a 08193 Bella e a Telè on: (93) 581 1680 Fax:(93) 581 2292 E-mail: [email p o ec ed] h p://www.ecap.uab.es Desemb e 2010 CO2 EMISSIONS AND ECONOMIC ACTIVITY: HETEROGENEITY ACROSS COUNTRIES AND NON STATIONARY SERIES1 Ma ías Piaggio* and Emilio Padilla Depa men o Applied Economics, Uni . Au ónoma de Ba celona, Edi icio B, Campus de Bella e a, 08193, Bella e a, Spain E-mails: ma [email protected]; [email protected] Tel. : +(34) 935814572 Fax: + (34) 935812292 *Co esponding au ho 1 We a e deeply g a e ul o P o . J.L. Raymond, o his ad ice, pa ience, and gene osi y in sha ing his knowledge. M. Piaggio wan s o hank o he Agencia Española de Coope ación In e nacional om he Minis e io de Asun os Ex e io ies o he inancial suppo p o ided. The au ho s also acknowledge suppo om p ojec s ECO2009-10003 (Minis e io de Ciencia e Inno ación), 2009SGR-600 and XREPP (DGR). CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies Abs ac This pape explo es he homogenei y o he unc ional o m, he pa ame e s, and he u ning poin , when app op ia e, o he ela ionship be ween CO2 emissions and economic ac i i y o 31 coun ies (28 OECD, B azil, China, and India) du ing he pe iod 1950 o 2006 using coin eg a ion analysis. Wi h a sample highly o e lapped o e ime be ween coun ies, he esul e eals ha he homogenei y ac oss coun ies is ejec ed, bo h in unc ional o m and in he pa ame e s o long e m ela ionship. This con i ms he ele ance o conside ing he he e ogenei y in explo ing he ela ionship be ween ai pollu ion and economic ac i i y o a oid spu ious pa ame e es ima es and in e a w ong beha io o he unc ional o m, which could lead o induce ha he ela ionship is e e sed when in ac i is di ec . Keywo ds: Bound es ing, coin eg a ion, CO2 emissions, en i onmen al Kuzne s cu e, he e ogenei y. JEL codes: C32, O13, Q53, Q56. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 1 1. In oduc ion The En i onmen al Kuzne s Cu e (EKC) hypo hesis comes om Kuzne s (1955), whe e an in e ed-U shaped ela ionship is supposed be ween income inequali y and income le el. The EKC hypo hesis sugges s he exis ence o an in e ed-U shaped ela ionship be ween en i onmen al deg ada ion and income le el. G ossman and K uege (1991) a gued ha he e a e h ee channels ha explain his pa h. In ea ly s ages o economic g ow h, he g ea e equi emen o na u al esou ces and was e gene a ion inc eases en i onmen al deg ada ion (scale e ec ). This g owing pa h migh lead o changes in he economic s uc u e owa ds less pollu ing ac i i ies (composi ion e ec ), which along wi h he inc ease in he capaci y o highe income coun ies o ace echnological subs i u ion owa ds less pollu ing p ocesses ( echnological e ec ) would lead o a u ning poin in he ela ionship and o he dec easing sec ion o he cu e. The e o e, he ansi ion om he inc easing o he dec easing sec ion o he cu e in he ela ionship be ween en i onmen al deg ada ion and economic ac i i y would a ise when he composi ion and echnological e ec s wo ked in he indica ed di ec ion and o e came he scale e ec 2. 2 The exis ence o composi ion and echnological e ec s do no necessa ily imply a esul as he one sugges ed by he EKC hypo hesis. Fo his o be he case, i is equi ed ha he composi ion e ec in ol es a educ ion o pollu ing sec o s in absolu e and no only in ela i e e ms. As o he echnological change, i migh some imes in ol e new p ocesses wi h new (and some imes unknown) pollu an s o e iciency imp o emen s leading o he inc ease o ex ac i e o o he en i onmen ally damaging ac i i ies (Roca and Padilla, 2003). The e o e, i depends on he ype o echnological and composi ion change ha hese e ec s compensa e o ein o ce he scale e ec o a speci ic pollu an . CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 2 Howe e , an EKC can be d i en by di e en unde lying ac o s, so ha he ela ion behind he hypo hesis can be gene a ed by di e en s uc u al models (Pe man and S e n, 1999). The li e a u e highligh s he dis ibu ion o powe (To as and Boyce, 1998), income-elas ici y o he demand o en i onmen al quali y (McConell, 1997; Dasgup a e al. 2002), en i onmen al egula ion and in e na ional ag eemen s (de B uyn, 1997) o s uc u al ansi ions, like he oil p ice shocks in he 1970s (Moomaw and Un uh, 1997). Also, an EKC can be eached by indi idual coun ies h ough he pollu ion ha en hypo hesis (S e n e al., 1996; Cole e al., 1997). In his way, al hough an in e ed-U ela ionship can be empi ically shown, his can be a s a is ical esul s emming om o he ac o s, which migh imply ha he obse ed ela ionship be ween en i onmen al deg ada ion and economic g ow h is spu ious. Mo eo e , hese ac o s migh a y ac oss coun ies and be di e en o di e en pollu an s. Ea lie wo ks igno ed ha he ela ionship be ween en i onmen al deg ada ion and income can be he e ogeneous ac oss coun ies (o egions), bo h in he unc ional o m as well as he pa ame e s and he u ning poin (G ossman and K uege , 1991 and 1994; Sha ik and Bandyopadhyay, 1992; Selden and Song, 1994; Ca son e al. 1997; Cole e al. 1997 and Vincen , 1997). This issue was i s s udied in he la e 1990s and ea ly 2000s (Pe man and S e n, 1999 and 2003; Lis and Galle , 1999; Dijkg aa and Vollebe gh, 2001; Ma ínez-Za zoso and Bengochea-Mo ancho, 2003 and 2004 and Dijkg aa e al., 2005). Following he same conce ns, a se ies o analyses o he EKC a na ional le el has eme ged, (among hem Vincen , 1997; de B uyn e al., 1998; Moomaw and Un uh, 1998; F iedl and Ge zne , 2003; Lekakis, 2000; Roca e al., 2001; Decon CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 3 and No man, 2004; Egli, 2004; Hung and Shawn, 2004; Shen, 2006; Halicioglu, 2008; Piaggio, 2008; Song e al., 2008; and Wang, 2009). Mo eo e , un il he s udy o Pe man and S e n (1999), he s a is ical p ope ies o he da a employed we e no aken in o conside a ion. The analysis using non- s a iona y se ies has o be ca ied ou aking in o accoun his cha ac e is ic. The adi ional EKC app oach no only igno es ha economies wi h he same le el o ac i i y migh p esen he e ogeneous unc ional o ms wi h espec o he ela ionship be ween income and en i onmen al deg ada ion, bu also assumes pa ame e homogenei y in his ela ionship ac oss coun ies. An EKC es ima ed om c oss-sec ion o panel da a when he se ies a e no o a e ha dly o e lapped o e ime ac oss coun ies can simply e lec he jux aposi ion o a posi i e ela ionship be ween en i onmen al deg ada ion and income in ich coun ies wi h a nega i e one in de eloping coun ies, and no a ela ionship ope a ing o bo h kinds o coun ies (Vincen , 1997). This p oblem can be sol ed i he panel da a se has o e lapped obse a ions o la ge pe iods (Egli, 2004). Howe e , his would no sol e he p oblem o assuming homogenei y in he unc ional o m o he ela ionship be ween en i onmen al deg ada ion and income among coun ies. In ligh o he abo e, he analyses ha assume homogenei y in he unc ional o m and in he pa ame e s ac oss coun ies migh in ac no e lec he beha io o he ela ionship be ween en i onmen al deg ada ion and income o hese a he indi idual le el. So, he conclusions ha , a e ce ain poin , CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 4 en i onmen al deg ada ion dec eases wi h g ea e economic ac i i y o he mo e de eloped coun ies migh be w ong. Consequen ly, mo e a en ion should be paid o indi idual coun ies beha io in o de o assess he possible bene i s o he inc ease in economic ac i i y on en i onmen al quali y o each coun y (de B uyn e al., 1998). To impose a p io i he cons ain o homogenei y be ween coun ies in he unc ional o m and he pa ame e s migh be a s a is ical de ice mo e han a model ha app op ia ely app oxima es eali y. Ca son (2010) a gues ha he analysis should dis inguish be ween a “weak” e sion o he EKC hypo hesis, o a pa icula poli ical ju isdic ion, and a “s ong” one, applying o he di e en poli ical ju isdic ions. The objec i e o his pape is o analyze he homogenei y in he unc ional o m and pa ame e s among coun ies in he long- un ela ionship be ween ca bon dioxide emissions (CO2) and economic ac i i y. The analysis is ca ied ou o 31 coun ies (28 OCDE coun ies, B azil, China and India) o e he pe iod 1950–2006; such a pe iod p esen s a high deg ee o o e lapping ac oss he se ies. Fi s , he unc ional o m homogenei y will be es ed h ough he es ima ion o he ela ion o each indi idual coun y. The ime pe iod conside ed in his pape is longe han he one om p e ious s udies. This is e y impo an , because a longe pe iod inc eases he o e lap among coun ies ha ha e simila economic ac i i y le el bu migh ha e he e ogeneous unc ional o ms. Fo hose coun ies wi h homogeneous unc ional o ms he homogenei y in he pa ame e s o he long un ela ionship would be es ed, allowing a ia ions among hem in bo h sho e m adjus men s and in he a e o con e gence o he long un ela ionship. Also, homogenei y in he u ning CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 5 poin among he coun ies ha p esen s one would be es ed. The use o coin eg a ion echniques would a oid he possibili y o a spu ious ela ionship be ween CO2 emissions and economic ac i i y. In he nex sec ion, he concep ual amewo k o he EKC hypo hesis and he ela ionship be ween economic g ow h and en i onmen al deg ada ion adjus ed o ou analysis is p esen ed. Sec ion 3 p esen s he me hodology and da a used. Sec ion 4 de ails he analysis esul s. Sec ion 5 p esen s he inal ema ks. 2. Concep ual amewo k The EKC hypo hesis a ises om a educed model speci ica ion. The e o e, i can be he esul o one o mo e di e en s uc u al ela ionships, because i is an empi ical phenomenon. So, his is in ac an appa en ela ion analysis be ween en i onmen al deg ada ion and economic ac i i y. In line wi h p e ious wo ks, he educed o m model ela es en i onmen al deg ada ion le el wi h economic ac i i y o each coun y, which can ollow a lineal, quad a ic o cubic unc ional o m: (1) whe e E deno es he indica o o en i onmen al deg ada ion (pe capi a) and Y is income (pe capi a). Subsc ip i=1,…, N indica es subjec s (coun ies), subsc ip = 1, …, T is he ime pe iod indica o , and ε is he e o e m no mally i i i i ii YYYE εβββα ++++= 3 3 2 21 CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 12 pa icula c i e ia (Hall, 1991)5. The e m wi hin b acke s ep esen s he e o co ec ion e m (ECT). The in e p e a ion o i s pa ame e s should be cau ious because when he e m is no malized wi h espec o a iable E, he sign o he o he a iable coe icien s is opposi e o he expec ed one. Besides he imp o emen in he consis ence p o ided by he es ima ion me hod, his speci ica ion, p esen s h ee mo e ad an ages: i) i allows o iden i y he long un ela ionship, he sho un dynamic and he coe icien o adjus men o he equilib ium ela ionship (α), ii) i he se ies in le els a e coin eg a ed, he ECM is a linea combina ion o s a iona y a iables. Then, es ima ions a e obus , and con en ional in e ence p ocedu es can be applied, and iii) his speci ica ion allows es ing di e en es ic ions among indi iduals (Pe man and S e n, 1999 and 2003). Coin eg a ion analysis and he es ima ion o he long un ela ionship by means o he ECM should be ei e a ed o he cubic speci ica ion (equa ions (2) and (3)), quad a ic (when 0 and 0  1…)) and linea (when 0 and 0  1… and 1…). Tha way, he bes unc ional o m o he long un ela ionship be ween CO2 emissions and income le el o each single coun y will be de e mined (i one exis s). Those coun ies ha do no sa is y he BT coin eg a ion es , o ha he model es ima ed is no sa is ac o y o he unc ional o m ha he BT indica es, a uni oo s analysis h ough he Augmen ed Dickey-Fulle es (ADF) and he coin eg a ion analysis h ough Engel-G ange es (1987) should be ca ied ou (Ende s, 2004). Then, when he se ies a e I(1) and a e coin eg a ed he ECM may be es ima ed o each 5 A gene al model o a gi en p, p1, p2 and p3 alue, la ge enough, is speci ied. Then, he lag is educed, in an independen way o each o hem, de e mining he alue o each o hem o he lag o g ea e deg ee s a is ically signi ican . CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 13 speci ica ion. Engle-G ange coin eg a ion es is seen as he mos app op ia e one o he p esen analysis, because a p io i we explo e he exis ence o only one coin eg a ion ela ion. The es p oposed by Johansen and Juselius (1990) and Johansen (1991) becomes complex in he p esence o non linea ans o ma ions o one o he a iables, as i allows o he exis ence o mo e han one coin eg a ion ela ionship. The p esen speci ica ion does no ackle he omission o ele an a iables p oblem. Lis and Galle (1999) a gue ha a educed o m model allows o measu e he di ec and indi ec ela ionship be ween economic ac i i y and en i onmen al deg ada ion, so ha he inclusion o addi ional a iables would dis o he analysis. The e o e, i is no possible o make causali y conclusions based on a educed o m model. So, i is no possible o assess wha causes he ela ionship o exis . This kind o analysis allows he s udy o appa en elas ici ies, no being an analysis o he de e minan s o en i onmen al pollu ion. As i is a uniequa ional speci ica ion, i does nei he sol e he p oblem o a possible eedback be ween he a iables. Howe e , as i is de eloped h ough a coin eg a ion analysis, he es ima ed pa ame e s will be supe consis en , no being a ec ed by he endogenei y bias o he a iables (Vee bek, 2005). The speci ica ion o he ECM o he analysis o his ela ionship is employed by Pe man and S e n (1999 and 2003) o SO2 emissions, and Ma ínez-Za zoso and Bengochea-Mo ancho (2003 and 2004) and Dinda and Condoo (2006) o CO2 emissions, all o hem wo king wi h panel da a. Egli (2004) o di e se kind o con aminan s and Iwa a e al. (2009 and 2010) o CO2 emissions employed CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 14 i o indi idual coun ies, and Haciglou (2008) and Piaggio (2008), who s udy CO2 emissions o indi idual coun ies bu in a mul i equa ion speci ica ion. Once he co ec unc ional o m is speci ied and he long un ela ionship h ough he ECM is es ima ed, he homogenei y o pa ame e s among coun ies wi h equal unc ional o m will be s udied, allowing he sho un coe icien s o be di e en among coun ies, as well as he quan i y o lags in each one o hem. This will be es ed compu ing con idence in e als (CI)6 o he pa ame e s o he long un ela ion, g ouping hose coun ies wi h same unc ional o m he CI o which o e lap. The same exe cise is ca ied ou wi h espec o he coe icien o adjus men o disequilib ia om he long un ela ionship (α). A simila s a egy is ollowed o es ing he u ning poin homogenei y. The u ning poin o coun ies wi h a quad a ic unc ional o m equa ion (3) is gi en by θ   No mal󰇡θ ,Vθ 󰇢7, gi en he dis ibu ion o pa ame e s β1 and β2. F om his, he u ning poin CI will be compu ed o he u ning poin o hose coun ies whose bes adjus men is he quad a ic unc ional o m. A simila p ocedu e migh be de eloped wi h espec o hose wi h cubic unc ional o m. 6 IC: 󰇣󰆹/  √󰇤, whe e   is he s anda d de ia ion associa ed o he es ima ed pa ame e 󰆹, 󰇛1󰇜 is he con idence le el, and  is he sample size. 7 Employing he Del a Me hod, ollowing Hayashi (2000: pp. 93–94) and G eene (2003, p. 70), . 󰇡  󰇢󰇡 󰇢󰇭   󰇮 CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 15 2.2. Da a The analysis akes in o accoun 31 coun ies (28 OECD coun ies8, B azil, China and India) be ween1950–20069. This ime pe iod is longe han he one om p e ious s udies on he homogenei y o he pa ame e s o CO2 emissions, which inc eases he possibili y o aking in o accoun coun ies wi h o e lapped income le els bu he e ogeneous pa hs. Mo eo e , he sample con ains almos all coun ies (excep Iceland and Luxembou g) commi ed o quan i a i e limi s in CO2 emissions h ough Annex B o he Kyo o P o ocol (Uni ed Na ions, 1998). CO2 emission da a is published by he Ca bon Dioxide In o ma ion Analysis Cen e (CDIAC) (Boden e al., 2009). I is consis en wi h he one o he Wo ld Bank (2005) o he pe iod 1960–2005, allowing o ake in o accoun en mo e yea s. CO2 emissions a e measu ed in me ic ons o CO2. Loga i hmic ans o ma ion o emissions pe capi a (co2pc) is employed. Economic ac i i y a na ional le el employed a e es ima ed and ans o med o 1990 Gea y-Khamis dolla s (which co ec s by pu chasing powe pa i y, PPP) by Madison (2003), upda ed o 2005 by he same au ho o 155 coun ies. The 8 Aus alia, Aus ia, Belgium, Canada, o me Czechoslo akia (a e 1992 he alues o Czech Republic and Slo akia a e added), Denma k, Finland, F ance, Ge many ( o he pe iod 1950– 1990 he in o ma ion o he Ge man Fede al Republic and he Ge man Democ a ic Republic a e added), G eece, The Ne he lands, Hunga y, I eland, I aly, Japan, Sou h Ko ea, Mexico, No way, New Zealand, Poland, Po ugal, Spain, Sweden, Swi ze land, Tu key, UK, USA, and o me So ie Union ( om 1992 he alues o Es onia, Geo gia, Kazakhs an, Ky gyzs an, La ia, Li huania, Moldo a, Russia, Tajikis an, Tu kmenis an, Uk aine and Uzbekis an a e added). Two OECD coun ies, Iceland and Luxembou g, a e excluded due o lack o in o ma ion o he en i e pe iod. 9 Excep o Belgium, o which we ook he pe iod 1962–2006, as i p esen ed a ypical alues o he wo i s yea s o he sample. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 16 Na ional Accoun s Sys em was se up in 1950 in a ious coun ies, which allows ha ing eliable in o ma ion. Loga i hmic ans o ma ion o pe capi a g ow h domes ic p oduc o he a iable in le els, and i s quad a ic and cubic ans o ma ion a e used (gdppc, gdppc2, and gdppc3, espec i ely). 4. Resul s In his sec ion, i s he coin eg a ion analysis h ough he BT es is ca ied ou o de e mine he exis ence o a long un ela ion be ween he a iables and he mo e adequa e unc ional o m o each o he coun ies. Second, he analysis o he pa ame e s o he long un ela ion homogenei y be ween coun ies, o he u ning poin and o he ECT coe icien is pe o med h ough con idence in e als cons uc ion. 4.1. Coin eg a ion analysis Following Pesa an e al. (2001) we will ca y ou he con as se e al imes, including up o ou lags, due o he sensi i eness o he analysis o he quan i y o lags included. Though he quan i y o lags seems high when wo king wi h annual da a, he leng h o he se ies allows i . Table I summa izes he esul s o he F-s a is ic o he Wald es o he linea , cubic and quad a ic speci ica ion o equa ion (2). Some coun ies o he sample allow o he exis ence o a long un ela ionship o he a iables o in e es o mo e han one unc ional o m. This migh esul , CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 17 o example, om quad a ic o ms ha ha e no achie ed he maximum, o ha ha e jus su passed i , o cubic o ms wi h iny dec easing sec ions migh be app oached h ough linea models. The e o e, he adequa e unc ional o m o each coun y would be de e mined om he coin eg a ion analysis join ly wi h he es ima ion o he equa ion (3) o each one o he unc ional o ms in he coun ies con i ming he exis ence o a long un ela ionship10. As shown in Table I, BT ejec s he null hypo hesis o no coin eg a ion o Aus alia (linea and quad a ic speci ica ion), Aus ia (quad a ic and cubic), o me Czechoslo akia (linea and quad a ic), Denma k (quad a ic and cubic), Ge many (cubic), G eece (linea and quad a ic), Hunga y (linea and cubic), I eland (linea , quad a ic and cubic), I aly (linea , quad a ic, and cubic), Japan (quad a ic), Sou h Ko ea (linea , quad a ic, and cubic), Poland (linea ), Po ugal (quad a ic and cubic), Swi ze land (linea , quad a ic and cubic), Tu key (quad a ic and cubic), o me So ie Union (quad a ic) and China (linea and cubic). When he BT is inconclusi e, Iwa a el al. (2009 and 2010) a gue ha he non exis ence o a coin eg a ion ela ionship may be ejec ed o no acco ding o he es o signi icance o he pa ame e o adjus men (α) o equa ion (3). The BT is no conclusi e o Belgium (quad a ic), Canada (quad a ic and cubic), o me Czechoslo akia (cubic), Finland (linea , quad a ic and cubic), G eece (cubic), he Ne he lands (quad a ic and cubic), Hunga y (quad a ic), Japan (cubic), 10 Fo he choice o he unc ional o m we employed di e en s a is ical and analy ical ools, such as he -s a is ic signi icance o he pa ame e s, he Schwa z in o ma ion C i e ia, and aking in o accoun i he u ning poin es ima ed is lowe han he maximum le el o income eached by each coun y. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 18 Mexico (linea ), No way (quad a ic and cubic), New Zealand (quad a ic and cubic), Spain (quad a ic), Sweden (linea and cubic), o me So ie Union (linea and Cubic), China (quad a ic) and India (quad a ic). Finally, he es indica es ha he e is no a long un ela ionship o any unc ional o m o F ance, Uni ed Kingdom, USA and B azil. Lags 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 AUS 4.09b7.66** 1.02 0.80 0.72 5.12** 4.93** 1.35 0.72 0.84 2.31 0.51 2.03 1.82 NA AUT 1.50 0.86 1.17 1.25 0.78 4.25* 2.87 3.92b2.15 1.62 3.31b2.10 4.11* 3.86* 2.93b BEL 4.91* 2.96 2.37 1.73 0.89 9.78*** 3.20b2.07 1.65 1.36 9.31*** 1.70 2.17 1.14 0.93 CAN 0.65 0.49 0.78 1.50 2.14 2.57 3.08 1.54 1.88 3.53b2.35 3.40b2.19 1.54 2.81b CZE 8.01*** 3.37 3.12 3.96 5.1* 4.63* 2.87 2.08 1.96 1.99 3.14b1.73 1.39 0.36 0.56 DEN 2.16 2.14 1.75 1.73 1.42 9.66*** 5.53** 5.84** 7.23** 3.27b7.05*** 4.61** 4.98** 5.79*** 3.58b FIN 2.82 2.63 2.81 4.54b3.71 3.22b2.58 2.19 1.61 1.25 2.81b2.46 2.95b1.90 1.32 FRA 1.21 1.49 1.55 2.17 1.69 1.84 1.79 2.07 1.78 1.01 2.22 1.53 2.51 3.03b2.75b GER 1.39 0.36 0.54 0.62 0.75 1.33 1.92 1.67 1.23 1.26 3.89* 2.62 2.26 2.23 0.80 GRE 4.27b5.07* 6.27** 7.58** 7.45** 5.64** 6.15** 6.35** 4.43** 5.30** 2.88b 2.69 2.18 1.37 1.16 HOL 1.23 0.63 0.55 0.85 0.93 3.16 2.45 3.50b2.60 2.07 2.88b2.01 2.92b2.52 2.07 HUN 13.01** * 8.69*** 2.80 4.73b4.43b3.84b3.29b0.63 0.49 0.51 3.15b3.87* 1.10 2.49 1.81 IRE 1.94 2.20 4.73b5.73** 8.32** 7.12*** 3.86b2.75 1.87 3.25b7.80*** 5.14** 3.34b2.08 1.90 ITA 6.50** 2.85 2.75 2.54 1.78 3.67b5.38** 1.75 1.98 2.31 3.10b4.82** 1.72 3.17b3.36b JAP 1.16 3.26 1.95 1.73 1.84 2.01 4.35** 2.65 1.10 1.79 1.54 2.85b1.82 2.36 2.73b KOR 24.53** * 12.57*** 19.36** * 8.56** 3.97** 21.19*** 8.41*** 10.20** * 7.06** 3.39b15.58*** 6.28** 9.75*** 3.50b1.95 MEX 1.09 0.67 0.13 0.49 0.66 1.67 1.50 2.36 1.49 1.52 1.67 1.53 3.48b2.64 3.59b NOR 2.20 1.60 1.51 1.89 2.53 3.54b1.30 1.62 0.82 1.54 3.22b1.86 2.77b2.42 1.41 NZL 2.70 2.50 1.11 2.22 2.23 2.52 2.41 2.68 2.21 3.20b3.12b1.72 1.63 1.80 1.96 POL 5.23** 3.26 1.72 1.54 1.46 2.38 0.82 0.50 0.03 0.06 1.93 1.59 1.06 0.42 0.46 POR 0.04 0.13 0.11 0.06 0.06 6.68*** 3.69b2.50 2.82 3.73b6.95*** 3.75b2.25 3.01b4.59** SPA 0.14 0.09 0.02 0.02 0.12 3.24b1.19 1.62 1.00 1.20 2.25 1.06 1.17 0.91 1.27 SWE 4.41b3.58 2.83 4.43b3.60 1.48 0.74 0.89 0.83 1.23 2.12 2.46 2.55 3.26b2.45 SWI 1.44 0.89 2.76 4.60b6.42** 5.83** 8.05*** 3.82b3.37b2.20 5.67*** 7.02** 1.57 3.14b1.67 TUR 0.64 1.16 2.65 2.31 1.65 7.51*** 6.86*** 3.28b2.17 2.88 5.14** 4.62** 2.38 2.19 3.39b UK 3.97 1.61 1.89 1.70 1.33 2.61 1.65 1.86 1.36 0.83 2.16 1.43 1.70 1.22 0.70 USA 0.53 0.90 1.71 1.74 3.10 1.14 1.11 0.78 0.96 1.17 1.98 1.26 0.54 0.71 0.47 USS 4.53b2.21 3.27 2.33 1.06 4.57* 1.38 2.83 1.76 0.54 3.29b0.92 2.58 2.04 1.76 BRA 3.75 3.48 2.18 3.63 0.81 2.50 1.93 1.61 1.46 1.57 2.65 1.83 1.70 1.65 1.25 CHN 6.70*** 3.88 4.67b7.22** 5.25* 2.64 2.99 2.49 3.46b3.51b3.63b4.39** 2.86b2.78b3.80* IND 0.10 0.58 2.59 1.71 1.66 4.25* 3.82* 1.16 1.14 0.75 2.50 2.39 0.59 0.86 0.37 3 1% CV (4.29;5.61), 5% CV (4.35;3.23) and 10% CV (3.77;2.72) ***, **,* signi ican a 1%, 5% and 10% espec i ely b inconclusi e a 1% Table I - CO 2 emissions and economic ac i i y bound es ing coin eg a ion es Lineal 1 Quad a ic 2 Cubic 3 1 1% CV (6.84;7.84), 5% CV (4.98;5.73) and 10% CV (4.04;4.78) 2 1% CV (5.15;6.36), 5% CV (3.79;4.85) and 10% CV (3.17;4.41) CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 19 F om he analysis abo e, when BT does no ejec he exis ence o a long un ela ionship equa ion (3) is es ima ed. The e o e, he p e e ed unc ional o m o each coun y is de e mined. Table A1 o Annex A summa izes he ECT es ima ion o equa ion (3) o each one o he possible unc ional o ms. The esul s indica e he exis ence o a long un ela ionship be ween CO2 emissions and economic ac i i y, bo h in pe capi a e ms, in a cubic pa h o Sweden, quad a ic o Aus alia, Aus ia, Belgium, Canada, Denma k, Finland, The Ne he lands, I eland, I aly, Japan, No way, Swi ze land, China and India, and lineal o Sou h Ko ea, G eece and B azil. Finally, he e is no long un ela ionship be ween he a iables in ol ed o any unc ional o m o o me Czechoslo akia, Hunga y and he o me So ie Union. F om he 17 coun ies o which a quad a ic speci ica ion is possible, 14 p esen he u ning poin wi hin he sample, which con i ms an in e ed-U pa h. The o he 3 a e e y close o achie ing i . Sweden also p esen s he u ning poin s wi hin he alues o he sample. The unc ional o m speci ica ion o 18 coun ies o he sample has been de e mined, and he ECM o each one o hem has been es ima ed. Mo eo e , o hose coun ies ha BT did no indica e he exis ence o a coin eg a ion ela ion (F ance, Uni ed Kingdom, USA and B azil), and o hose ha BT did no ejec i o one o he speci ica ions bu was no possible o es ima e a sa is ac o y long un ela ionship (Ge many, Mexico, New Zealand, Poland, Po ugal, Spain and Tu key), a uni oo s analysis h ough he ADF s a is ic and a coin eg a ion analysis h ough he Engle-G ange es a e implemen ed. All he se ies o all he coun ies a e I(1), and he exis ence o a long un CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 20 ela ionship is no ejec ed o any o he speci ica ions o F ance and Spain, he linea and quad a ic speci ica ions o Ge many, Mexico, USA and B azil, he cubic o Poland and he linea o Po ugal and Tu key. The analysis ejec s he exis ence o a long un ela ionship o New Zealand11. The e is a long un quad a ic ela ionship o F ance, Ge many and USA, and linea o Mexico, Po ugal, Spain, Tu key and B azil. Poland and Uni ed Kingdom do no p esen any sa is ac o y speci ica ion, as equa ion (3) shows es ima ions o he speci ica ions ha do no ejec he exis ence o a coin eg a ion ela ionship. Table II summa izes he esul s, 25 o he 31 coun ies o he sample do no ejec he exis ence o a long un ela ionship be ween economic ac i i y and CO2 emissions (7 linea , 17 quad a ic and 1 cubic). The esul ob ained con i ms he he e ogenei y among coun ies o beha io pa e ns o simila ac i i y le els. Compa ing hese esul s wi h o he analyses o he same pollu an o indi idual coun ies, hey a e consis en wi h he ones o Iwa a e al. (2009) o F ance ( o he pe iod 1960–2003), and Iwa a e al. (2010) o Finland (1977– 2003) and Japan (1966–2003). The las one es s —and ob ains posi i e e idence o — he exis ence o a quad a ic pa h o Sou h Ko ea (1977–2003) and Spain (1968–2003), in con as wi h he linea model suppo ed by ou esul s. Bo h wo ks quo ed ake in o accoun he sha e o nuclea powe in o al 11 The esul s om he uni oo s and coin eg a ion es s a e a ailable om he au ho s upon eques . CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 21 ene gy gene a ion o each coun y. Howe e , he linea speci ica ion o Spain is consis en wi h Roca and Padilla (2003) o he pe iod 1980–2000, who also included ac o s e e ed o he ene gy sou ces s uc u e. In con as wi h ou esul s, F iedl and Ge zne (2003) ound a cubic ela ionship o Aus ia (1960–1999), in oducing he weigh o impo s and indus y in o al income. Haciloglu (2008) also ound a di e en pa h om ou s o Tu key Model Coun y Decision Me hod BRA EG GRE BT KOR BT MEX EG POR EG SPA EG TUR EG AUS BT AUT BT BEL BT CAN BT CHN BT DEN BT FIN BT FRA EG GER EG HOL BT IND BT IRE BT ITA BT JAP BT NOR BT SWI BT USA EG Cubic SWE BT CZE BT HUN BT NZL EG POL EG UK EG USS BT Table II - Summa y o long e m ela ionship es ima ion Linea Quad a icNo ela ion CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 28 economic ac i i y g ow h in en i onmen al quali y, he s udy should ocus on he analysis o he ela ionship be ween hese ac o s a single coun y le el. The esul s o he p esen esea ch a e consis en wi h Dijkg aa and Vollebe gh (2001) and Dijkg aa e al. (2005) on he p oblema ic assump ion o pa ame e s homogenei y o he long un ela ion be ween CO2 emissions and economic ac i i y le el, bo h pe capi a, employing a longe pe iod sample, which allows a g ea e deg ee o o e lapping o he se ies among coun ies. A he same ime, his g ea e o e lapping ein o ces he esul o ejec ing he homogenei y in he unc ional o m among coun ies (Pe man and S e n, 1999 and 2003; Lis and Galle , 1999; Dijkg aa and Vollebe gh, 2001; Ma ínez-Za zoso and Bengochea-Mo ancho, 2003 and 2004 and Dijkg aa e al., 2005). This is highligh ed by he ac ha he e ogeneous unc ional o ms a e ound o coun ies wi h simila le el o economic ac i i y. The exis ence o a gene al ela ion o all he coun ies be ween CO2 emissions and GDP pe capi a is clea ly pu in o ques ion. Following Ca son (2010), his esul ejec s he op imis ic iew o he EKC, whe e de eloping coun ies migh igno e en i onmen al p oblems un il hey become de eloped. De eloped coun ies can and ha e o conside his p oblem, since no hing gua an ees a pa h as he one o he EKC o all coun ies (and nei he he exis ence o a common pa h o hem) (Dasgup a e al., 2002). Finally, he u ning poin homogenei y is ejec ed o he whole sample o coun ies. Howe e , he e a e g oups o coun ies ha p esen pa ame e CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 29 he e ogeneous pa hs bu o which u ning poin s homogenei y is no ejec ed. Al hough his is no s ong e idence in a o o he op imis ic iew o he EKC, i sugges s ha i would be in e es ing o analyze he de e minan s o hese coun ies. Mo eo e , he impo ance o conside ing he en i onmen al deg ada ion le el om which he damage u ns ex emely high o i e e sible mus be highligh ed, as while di e en coun ies migh p esen a u ning poin o a simila le el o ac i i y, he en i onmen al deg ada ion achie ed in his poin migh be di e en . 6. Re e ences A ow, K., Bolin, B., Cos anza, R., Dasgup a, P., Folke, C., Helling, C. S., Jansson, B.-O., Le in, S., Maile , K.-G., Pe ings, C., Pimen el, D. (1995) “Economic g ow h, ca ying capaci y, and he en i onmen ”, Science, Vol. 268, pp. 520–521. Boden, T.A., Ma land, G., And es, R.J. (2009) Global, Regional, and Na ional Fossil-Fuel CO2 Emissions, Ca bon Dioxide In o ma ion Analysis Cen e , Oak Ridge Na ional Labo a o y, U.S. Depa men o Ene gy, Oak Ridge, Tenn., U.S.A. doi 10.3334/CDIAC/00001 Ca son, R. (2010) “The En i onmen al Kuzne s Cu e: seeking empi ical egula i y and heo e ical s uc u e”, Re iew o En i onmen al Economics and Policy, Vol. 4:1, pp. 3–23. Ca son, R, Jeon, Y., Mccubbin, D. (1997) “The ela ionship be ween ai pollu ion emissions and income: US Da a”, En i onmen and De elopmen Economics, Nº 2, pp. 433–450. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 30 Cole, M.A, Rayne , A.J., Ba es, J.M. (1997) “The en i onmen al Kuzne s cu e: an empi ical analysis”, En i onmen and De elopmen Economics, Nº 2, pp. 401–416. Dasugp a, S., Laplan e, B., Wang, H., Wheele , D. (2002) “Con on ing he en i onmen al Kuzne s cu e”, Jou nal o Economic Pe spec i es, Vol. 16, pp. 147–168. de B uyn, S.M., an den Be gh, J.C., Opschoo , J.B. (1998) “Economic g ow h and emissions: econside ing he empi ical basis o en i onmen al Kuzne s cu es”, Ecological Economics, Vol. 25, pp 161–75. Dijkg aa , E., Melenbe g, B., Vollebe gh,H. (2005) “En i onmen al Kuzne s Cu es o CO2: he e ogenei y e sus homogenei y”, En i onmen al and Resou ce Economics, Vol. 32, pp. 229–239. Dijkg aa , E., Vollebe gh, H. (2005), “A Tes on pa ame e he e ogenei y in CO2 panel EKC es ima ions”, En i onmen al and Resou ce Economics, Vol. 32, pp. 229–239 Dinda, S., Coondoo, D. (2006) “Income and emission: A panel da a-based coin eg a ion analysis”, Ecological Economics, Vol. 57, pp. 167–181. Egli, H. (2004) “The En i onmen al Kuzne s Cu e – E idence om Time Se ies Da a o Ge many”, ETH Zu ich, H p://ideas. epec.o g/p/e h/wpswi /03- 28.h ml. Ende s, W. (2004) Applied Econome ic Time Se ies, Wiley, 2nd Edi ion. Engle, R. F., G ange , C. W. J. (1987) “Co-in eg a ion and E o Co ec ion: Rep esen a ion, Es ima ion, and Tes ing”, Econome ica, Vol. 55, pp. 251– 276. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 31 F iedl, B., Ge zne (2003) “De e minan s o CO2 emissions in a small open economy”, Ecological Economics, Vol. 45, pp. 133–148. G ange , C.W., Hallman, J. (1991) “Nonlinea ans o ma ions o in eg a ed ime se ies”, Jou nal o Time Se ies Analysis, Vol. 12, pp. 207–224. G eene, W. (2003) Econome ic Analysis, P en ice Hall. G ossman, G., K uege , A. (1991) “En i onmen al impac s o a No h Ame ican ee ade ag eemen ”, NBER wo king pape , Nº 3914. Halicioglu, F. (2009) “An econome ic s udy o CO2 emissions, ene gy consump ion, income and o eign ade in Tu key”, Ene gy Policy, Vol. 37, pp. 1156–1164. Hall, A. (1991) “Tes ing o a Uni Roo in Time Se ies wi h P e es Da a Based Model Selec ion”, Depa men o Economics, No h Ca olina S a e Uni e si y. Mimeo. Hayashi, F. (2000) Econome ics, P ince on Uni e si y P ess. Hung, M.F., Shaw, D. (2004) “Economic G ow h and he En i onmen al Kuzne s Cu e in Taiwan: a simul anei y model analysis”, en in Bold in, M., Chen, B.L. y Wang, P. (eds.), Human Capi al, T ade and Public Policy in Rapidly G owing Economies: F om Theo y o Empi ics, pp. 269–290. Iwa a, H., Okada, K., Sam e h, S. (2010) “Empi ical s udy on he de e minan s o CO2 emissions: e idence om OECD coun ies”, Munich Pe sonal RePEc A chi e, pape nº 21520. Iwa a, H., Okada, K., Sam e h, S. (2009) “Empi ical s udy on he En i onmen al Kuzne s Cu e o CO2 in F ance: he ole o nuclea ene gy”, Munich Pe sonal RePEc A chi e, pape nº 18997. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 32 Johansen, S. (1991) “Es ima ion and Hypo hesis Tes ing o Coin eg a ion Vec o s in Gaussian Vec o Au o eg essi e Models”, Econome ica, Vol. 59, pp. 1551–1580. Johansen, S., Joselius, K. (1990) “Maximum Likelihood Es ima ion and In e ences on Coin eg a ion— wi h applica ions o he demand o money”, Ox o d Bulle in o Economics and S a is ics, Vol. 52, pp. 169–210. Kuzne s, S., (1955) Economic g ow h and income inequali y, AER. Lee, C., Lee, J. (2009) “Income and CO2 emissions: e idence om panel uni oo and coin eg a ion es s”, Ene gy Policy, Vol. 37, pp. 413–423. Lekakis, J. (2000) “En i onmen and De elopmen in a sou he n Eu opean coun y: which En i onmen al Kuzne s Cu es?”, Jou nal o En i onmen al Planning and Managemen , Vol. 43, pp. 139–153. Lis , J., Galle , C. (1999) “The En i onmen al Kuzne s Cu e: does one i s all?”, Ecological Economics, Vol. 31, pp. 409–423. McConnell, K. (1997) “Income and demand o en i onmen al quali y”, En i onmen and De elopmen Economics, Vol. 2, pp. 383–399. Maddison, A. (2003) The Wo ld Economy: His o ical S a is ics, OECD. Ma ínez-Za zoso, I., Bengochea-Mo ancho, A. (2004) “Pooled mean g oup es ima ion o an en i onmen al Kuzne s cu e o CO2”, Economics Le e s, Vol. 82, pp. 121–126. Ma inez-Za zoso, I., Bengochea-Mo ancho, A. (2003) “Tes ing o an En i onmen al Kuzne s Cu e in la in-ame ican coun ies”, Re is a de Análisis Económico, Vol. 18, pp. 3–26. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 33 Moomaw, W., Un uh, G. (1997) “A e en i onmen al Kuzne s cu es misleading us? The case o CO2 emissions”, En i onmen and De elopmen Economics, Vol. 2, pp. 451–463. Pe man, R., S e n, D. (2003) “E idence om panel uni oo and coin eg a ion es ha he En i onmen al Kuzne s Cu es dows no exis ”, The Aus alian Jou nal o A icul u al and Resou ce Economics, Vol. 47, pp. 325–347. Pe man, R, S e n, D. (1999) “The En i onmen al Kuzne s Cu e: implica ions o non-s a iona i y”, The Aus alian Na ional Uni e si y, Cen e o Resou ce and En i onmen al S udies, Wo king Pape in Ecological Economics Nº 9901. Pesa an, M.H., Shin, Y. Smi h, R. (2001) “Bounds Tes ing app oaches o he analysis o le el ela ionships”, Jou nal o Applied Econome ics, Vol. 16, pp. 289–326. Piaggio, M. (2008) “Relación en e la Con aminación A mos é ica y la Calidad del Ai e con el C ecimien o Económico y o os De e minan es: U uguay a lo la go del Siglo XX”, Quan um, Vol. III, Nº 1, pp. 35–54, h p://www.quan um.edu.uy/nume o2/no a%203.pd . Roca, J., Padilla, E. (2003) “Emisiones a mos é icas y c ecimien o económico en España”, Economía Indus ial, Vol. 351, pp. 73–86. Roca, J., Padilla, E., Fa é, M., Galle o, V. (2001) “Economic g ow h and a mosphe ic pollu ion in Spain: discussing he en i onmen al Kuzne s cu e hypo hesis”, Ecological Economics, Vol. 39, pp. 85–99. Selden, T., Song, D. (1994) “En i onmen al quali y and de elopmen : Is he e a Kuzne s cu e o ai pollu ion emissions?”, Jou nal o En i onmen al Economics and Managemen , Vol. 27, pp. 147–162. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 34 Sha ik, N., Bandyopadhyay, S. (1992) “Economic g ow h and en i onmen al quali y: ime-se ies and c oss-coun y e idence”, Policy esea ch wo king pape s, Backg ound pape o he Wo ld De elopmen Repo . Shen, J. (2006) “A simul aneous es ima ion o En i onmen al Kuzne s Cu e: E idence om China”, China Economic Re iew, Vol. 17, pp. 383–394. Song, T., Zheng, T., Tong, L. (2008) “An empi ical es o he en i onmen al Kuzne s cu e in China: A panel coin eg a ion app oach”, China Economic Re iew, Vol. 19, pp. 381–392. S e n, D., Common, M., Ba bie , E. (1996) “Economic g ow h and en i onmen al deg ada ion: he en i onmen al Kuzne s cu e and sus ainable de elopmen ”, Wo ld De elopmen , Vol. 24, pp. 1151–1160. To as, M., Boyce, K. (1998) “Income, inequali y, and pollu ion: a eassessmen o he en i onmen al Kuzne s Cu e”, Ecological Economics, Vol. 25, pp. 147–160. Uni ed Na ions (1998) Kyo o P o ocol o he F amewo k Con en ion o he Uni ed Na ion on Clima e Change, Uni ed Na ions. Vee beek, M. (2004) A Guide o Mode n Econome ics, Wiley. Vincen , J. (1997) “Tes ing o en i onmen al Kuzne s cu es wi hin a de eloping coun y”, En i onmen and De elopmen Economics, Vol. 2, pp. 417–431. Wagne , M. (2008) “The Ca bon Kuzne s Cu e: a cloudy pic u e emi ed by bad econome ics?”, Resou ce and Ene gy Economics, Vol. 30, pp. 388– 408 CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 35 Wang, Y. (2009) “Tes ing he EKC hypo hesis: some econome ic pi alls and solu ions”, School o Economics, A nd Building 25a, The Aus alian Na ional Uni e si y. Wo ld Bank (2005) “Wo ld de elopmen indica o s”, Washing on, D.C. CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 36 Annex A Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea ADRL (0,0,0,0) (0,0,0) (1,1) (0,2,2,2) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,3) (1,0,0,0) (0,0,0) (1,1) (0,0,0,0) (0,0,0) (0,1) al a -0.45 -0.37 -0.07 -0.30 -0.32 -0.04 -0.19 -0.21 -0.13 -0.16 -0.18 -0.09 0.01 -0.02 0.01 -0.53 -0.55 -0.11 -s a is ic -3.47 -3.38 -1.92 -2.73 -3.32 -1.29 -2.59 -3.52 -1.89 -2.10 -2.30 -1.73 0.27 -0.79 0.42 -4.52 -4.96 -1.84 C -9.40 -3.13 -9.43 -9.89 -6.08 0.37 -6.48 -6.75 -9.19 21.29 -3.60 -8.52 11.21 -8.60 0.01 5.14 0.69 -8.48 -s a is ic -1.95 -5.01 2.04 -3.11 -14.14 1.31 -6.04 -9.22 -39.02 0.99 -1.92 -19.24 2.99 -0.88 0.00 0.63 0.81 -9.23 GDPPC(-1) 3.26 -4.18 0.00 1.84 -1.68 0.00 -2.59 -2.08 -0.06 -33.57 -4.35 -0.32 -7.11 -5.03 -5.60 -12.75 -7.40 -0.19 -s a is ic 0.57 -8.43 0.08 0.45 -4.37 0.00 -1.82 -3.49 -0.65 -1.34 -2.91 -1.90 -1.11 -0.59 -0.47 -1.29 -10.84 -0.55 GDPPC^2(-1) -2.25 0.66 -0.72 0.24 0.72 0.43 12.07 0.77 5.81 2.26 3.47 1.36 -s a is ic -1.01 6.77 -0.41 2.88 1.01 3.40 1.26 2.64 1.63 0.90 0.89 10.18 GDPPC^3(-1) 0.37 0.07 -0.05 -1.44 -1.34 -0.27 -s a is ic 1.31 0.27 -0.43 -1.19 -2.07 -0.54 In e en ions S ep 1970 Impulse Tu ning Poin 3.07 3.19 5.36 3.46 2.39 2.42 2.61 2.83 0.88 1.11 2.70 2.72 0.95 1.68 7.15 2.97 2.01 5.75 Schwa z iC -3.84 -4.03 -3.86 -2.68 -3.10 -3.86 -3.14 -3.21 -2.97 -3.43 -3.46 -3.49 -3.51 -3.81 -3.71 -2.15 -2.21 -1.86 JB 1.08 0.83 0.58 0.45 0.75 0.75 0.10 0.32 0.72 2.07 1.97 4.72 4.78 5.44 7.56 1.16 0.85 1.13 p- alue 0.58 0.66 0.75 0.80 0.69 0.69 0.95 0.85 0.70 0.36 0.37 0.09 0.09 0.07 0.02 0.56 0.66 0.57 BG (4 lags) 0.03 0.18 0.18 1.82 2.02 2.08 1.08 0.76 0.38 0.34 0.90 1.07 0.89 2.07 1.95 2.44 2.09 2.34 p- alue 1.00 0.95 0.95 0.14 0.11 0.10 0.38 0.56 0.82 0.85 0.47 0.39 0.48 0.10 0.12 0.06 0.10 0.07 TP in he sample YES NO YES YES YES YES Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea ADRL (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (3,3) al a -0.33 -0.33 -0.09 -0.31 -0.20 -0.06 -0.26 -0.11 -0.04 -0.26 -0.19 -0.14 -0.20 -0.23 -0.02 -0.16 -0.11 0.01 -s a is ic -3.02 -3.02 -1.68 -2.71 -2.47 -1.60 -2.49 -1.75 -1.16 -2.44 -2.41 -1.68 -2.43 -3.10 -0.48 -2.10 -1.90 0.26 C 2.92 -0.90 -7.44 4.10 -3.90 -7.46 -1.64 -2.88 -9.80 -5.28 -4.69 -5.78 -20.04 -1.36 -10.55 21.29 -8.24 6.09 -s a is ic 0.60 -0.93 -6.48 0.72 -2.92 -7.07 -0.20 -1.01 -7.94 -6.94 -12.62 -15.05 -1.86 -1.05 -1.59 0.99 -3.43 0.12 GDPPC(-1) -11.44 -5.84 -0.64 -13.61 -4.34 -0.26 -6.55 -5.14 0.36 -1.73 -2.72 -1.35 14.98 -5.94 0.54 -33.57 -1.10 -9.03 -s a is ic -1.70 -6.94 -1.44 -1.73 -3.81 -0.86 -0.67 -2.44 0.67 -1.05 -4.68 -9.18 1.16 -5.48 0.21 -1.34 -0.42 -0.28 GDPPC^2(-1) 3.65 1.01 4.64 0.91 1.37 1.03 -0.14 0.39 -6.40 1.10 12.07 0.44 -s a is ic 1.21 5.57 1.32 3.86 0.35 2.60 -0.14 2.13 -1.26 4.96 1.26 0.60 GDPPC^3(-1) -0.40 -0.53 0.00 0.09 0.87 -1.44 -s a is ic -0.92 -1.05 0.00 0.49 1.32 -1.19 In e en ions S ep 1982 1982 Impulse 1956 Tu ning Poin imag. 2.90 imag. 2.37 2.40 2.51 3.11 3.45 3.00 2.70 2.61 1.26 imag. imag. 1147.68 -2.08 1.92 2.97 Schwa z iC -1.61 -1.67 -1.59 -1.57 -2.83 -3.02 -3.80 -3.85 -3.91 -2.82 -2.92 -2.81 -2.82 -3.09 -2.77 -3.43 -2.98 -2.84 JB 2.01 1.25 1.13 3.36 1.56 0.34 1.47 0.26 0.48 0.48 1.91 0.45 4.96 7.87 9.06 2.07 7.00 0.26 p- alue 0.37 0.53 0.57 0.19 0.46 0.84 0.48 0.88 0.79 0.78 0.39 0.80 0.08 0.02 0.01 0.36 0.03 0.88 BG (4 lags) 0.29 0.07 2.34 2.11 0.49 2.37 1.46 1.26 1.37 1.09 2.19 0.36 1.22 0.87 0.88 0.34 2.59 0.28 p- alue 0.88 0.99 0.07 0.09 0.74 0.07 0.23 0.30 0.26 0.37 0.08 0.84 0.32 0.49 0.48 0.85 0.05 0.89 TP in he sample YES YES YES NO YES YES Table A1 - E o Co ec ion Te m - ECM cubic, quad a ic and linea model AUS AUT BEL CAN CZE DEN FIN FRA GER GRE HOL HUN CO2 Emissions and Economic Ac i i y: he e ogenei y ac oss coun ies and non s a iona y se ies 37 Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea ADRL (0,1,1,0) (0,0,0) (3,2) (0,0,0,0) (1,0,0) (0,0) (4,3,3,3) (0,1,0) (1,1,0) (3,3,0,0) (0,0,0) (2,2) (2,1,1,1) (2,0,0) (2,0) (0,2,2,2) (0,0,0) (0,1) al a -0.70 -0.57 -0.04 -0.15 -0.25 -0.04 -0.37 -0.25 -0.09 0.02 -0.10 -0.08 -0.24 -0.17 -0.12 -0.48 -0.34 -0.11 -s a is ic -5.56 -4.58 -0.50 -1.96 -3.68 -1.32 -4.33 -3.64 -2.48 0.32 -2.54 -3.59 -3.30 -2.83 -2.14 -4.34 -3.19 -1.64 C -2.84 -5.94 -11.10 0.00 -2.45 -8.63 -6.96 -5.61 -5.58 7.50 -7.18 -7.44 -12.31 -3.87 -5.80 3.11 -3.49 -7.32 -s a is ic -3.51 -28.65 -1.52 0.00 -5.51 -3.29 -7.82 -24.90 -10.14 0.19 -11.31 -17.48 -4.73 -3.28 -20.82 0.51 -3.14 -11.98 GDPPC(-1) -6.52 -2.06 1.44 -9.33 -4.56 -0.04 -1.21 -2.12 -1.13 -23.64 -1.45 -0.54 13.45 -4.34 -1.34 -12.86 -3.65 -0.66 -s a is ic -5.52 -10.62 0.34 -1.43 -11.70 -0.04 -0.87 -8.50 -7.29 -0.38 -3.16 -2.80 2.35 -2.51 -8.18 -1.82 -4.03 -3.09 GDPPC^2(-1) 2.29 0.31 3.30 0.80 0.20 0.31 13.01 0.24 -11.06 1.05 4.78 0.60 -s a is ic 4.18 6.79 1.04 9.10 0.29 4.91 0.37 1.93 -2.72 1.77 1.76 3.32 GDPPC^3(-1) -0.27 -0.40 -0.02 -2.48 2.67 -0.62 -s a is ic -3.35 -0.81 -0.14 -0.36 2.84 -1.80 In e en ions S ep Impulse 2006 Tu ning Poin imag. 3.36 imag. 2.85 imag. 3.38 imag. 3.05 1.85 2.07 imag. 3.06 imag. imag. imag. imag. 0.91 imag. Schwa z iC -2.44 -2.36 -2.15 -3.52 -3.65 -3.66 -3.10 -3.12 -3.59 -2.75 -2.59 -3.16 -2.83 -2.89 -2.98 -1.62 -1.72 -1.69 JB 1.95 4.18 0.18 2.72 1.37 2.05 0.74 0.39 7.21 0.13 1.94 1.09 5.53 6.38 4.27 0.95 7.80 0.59 p- alue 0.38 0.12 0.92 0.26 0.50 0.36 0.69 0.82 0.03 0.94 0.38 0.58 0.06 0.04 0.12 0.62 0.02 0.75 BG (4 lags) 0.85 0.76 1.63 0.60 0.53 0.63 3.11 2.32 1.88 1.31 0.89 0.53 1.79 0.49 1.21 0.88 2.69 2.30 p- alue 0.50 0.56 0.19 0.67 0.71 0.64 0.03 0.07 0.13 0.28 0.48 0.71 0.15 0.74 0.32 0.48 0.04 0.07 TP in he sample NO YES YES NO NO YES Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea Cubic Quad. Linea ADRL (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0,0) (1,0) (0,0,0,0) (0,0,0) (0,0) (0,0,0,0) (0,0) (0,0) (1,1,1,1) (0,0,0) (0,1) al a -0.34 -0.26 -0.25 0.03 0.03 0.00 -0.57 -0.55 -0.31 -0.29 -0.31 -0.16 -0.20 -0.14 -0.06 -0.38 -0.36 -0.04 -s a is ic -3.40 -2.66 -2.80 0.78 0.66 0.03 -3.82 -3.79 -3.21 -3.48 -3.93 -2.15 -2.62 -2.00 -1.57 -2.51 -3.80 -0.74 C -64.77 -7.07 -6.50 -2.18 -0.31 93.13 -5.58 -5.82 -5.58 -5.54 -5.94 -6.42 44.63 -0.58 -11.72 41.20 14.38 -8.87 -s a is ic -2.64 -2.20 -20.50 -0.44 -0.04 0.02 -14.23 -46.03 -61.00 -6.34 -26.10 -34.58 2.00 -0.14 -4.90 0.85 5.29 -1.72 GDPPC(-1) 70.12 -0.43 -0.87 -6.77 -8.76 -66.08 -1.37 -0.90 -1.20 -2.39 -1.63 -0.95 -61.52 -7.17 1.07 -42.33 -15.80 0.17 -s a is ic 2.33 -0.16 -6.91 -1.41 -1.14 -0.03 -1.77 -5.18 -30.44 -1.48 -6.08 -12.61 -2.26 -2.30 1.18 -0.81 -8.00 0.09 GDPPC^2(-1) -28.66 -0.09 0.39 1.63 2.19 0.21 -0.08 0.61 0.19 23.04 1.50 11.32 2.70 -s a is ic -2.34 -0.17 1.82 1.23 1.14 0.45 -1.65 0.69 2.63 2.11 2.48 0.61 7.56 GDPPC^3(-1) 3.83 0.33 -0.06 -0.07 -2.82 -0.92 -s a is ic 2.32 0.60 -0.64 -0.48 -1.96 -0.42 In e en ions S ep 1981, 1990 1981, 1990 Impulse Tu ning Poin 2.83 -2.44 1.44 2.00 imag. imag. 4.27 2.33 2.39 2.87 2.92 2.16 -4.72 imag. imag. 3.12 5.36 Schwa z iC -2.83 -2.81 -2.87 -3.72 -3.80 -3.47 -2.66 -2.72 -2.84 -2.74 -2.81 -2.80 -2.23 -2.23 -2.26 -2.61 -2.74 -2.45 JB 3.94 1.55 0.95 4.60 3.76 1.54 1.30 1.67 0.43 2.40 2.22 0.41 0.72 1.01 0.69 0.64 0.67 0.64 p- alue 0.14 0.46 0.62 0.10 0.15 0.46 0.52 0.43 0.81 0.30 0.33 0.81 0.70 0.60 0.71 0.73 0.72 0.73 BG (4 lags) 1.08 0.50 1.04 1.80 1.99 2.17 1.38 1.22 0.49 1.76 1.86 1.01 0.93 0.86 0.82 1.55 3.28 2.26 p- alue 0.38 0.74 0.40 0.15 0.11 0.09 0.25 0.31 0.74 0.15 0.13 0.41 0.45 0.50 0.52 0.21 0.02 0.08 TP in he sample YES YES YES NO YES YES Table A1 - E o Co ec ion Te m - ECM cubic, quad a ic and linea model IRE ITA JAP KOR MEX NOR NZL POL POR SPA SWE SWI