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