Mao, Qianqian; Ren, Yanjun; Loy, Jens-Pe e
A icle
Nonlinea p ice ansmission and asynch onous p ice
bubbles: Empi ical e idence om China's ag icul u al
u u es and spo ma ke s
Jou nal o Applied Economics
P o ided in Coope a ion wi h:
Uni e si y o CEMA, Buenos Ai es
Sugges ed Ci a ion: Mao, Qianqian; Ren, Yanjun; Loy, Jens-Pe e (2024) : Nonlinea p ice ansmission
and asynch onous p ice bubbles: Empi ical e idence om China's ag icul u al u u es and spo
ma ke s, Jou nal o Applied Economics, ISSN 1667-6726, Taylo & F ancis, Abingdon, Vol. 27, Iss. 1,
pp. 1-27,
h ps://doi.o g/10.1080/15140326.2024.2369441
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/314280
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc/4.0/
Jou nal o Applied Economics
ISSN: (P in ) (Online) Jou nal homepage: www. and online.com/jou nals/ ecs20
Nonlinea p ice ansmission and asynch onous p ice
bubbles: empi ical e idence om China’s ag icul u al
u u es and spo ma ke s
Qianqian Mao, Yanjun Ren & Jens-Pe e Loy
To ci e his a icle: Qianqian Mao, Yanjun Ren & Jens-Pe e Loy (2024) Nonlinea
p ice ansmission and asynch onous p ice bubbles: empi ical e idence om China’s
ag icul u al u u es and spo ma ke s, Jou nal o Applied Economics, 27:1, 2369441, DOI:
10.1080/15140326.2024.2369441
To link o his a icle: h ps://doi.o g/10.1080/15140326.2024.2369441
© 2024 The Au ho (s). Published by In o ma
UK Limi ed, ading as Taylo & F ancis
G oup.
Published online: 19 Jun 2024.
Submi you a icle o his jou nal
A icle iews: 395
View ela ed a icles
View C ossma k da a
Ci ing a icles: 2 View ci ing a icles
Full Te ms & Condi ions o access and use can be ound a
h ps://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode= ecs20
RESEARCH ARTICLE
Nonlinea p ice ansmission and asynch onous p ice
bubbles: empi ical e idence om China’s ag icul u al u u es
and spo ma ke s
Qianqian Mao
a,b
, Yanjun Ren
c
and Jens-Pe e Loy
d
a
School o Economics, Zhejiang Uni e si y o Technology, Hangzhou, China;
b
Ins i u e o Indus ial Sys em
Mode niza ion, Zhejiang Uni e si y o Technology, Zhejiang, China;
c
College o Economics & Managemen ,
No hwes A&F Uni e si y, Xianyang, China;
d
Depa men o Ag icul u al Economics, Uni e si y o Kiel, Kiel,
Ge many
ABSTRACT
P e ious s udies on commodi y p ice bubbles mainly ocused on
u u es ma ke s and igno ed he pe o mance o spo ma ke s.
Using he p ice da a o co n and soybeans in China, his s udy
iden i ies he exac bubble da es o he u u es and spo ma ke s,
and inds asynch onous p ice bubbles be ween hese wo ma ke s.
Bubbles a e mo e equen o commodi y spo p ices, while he
co esponding u u es p ices s ill domina e he p ocess o p ice
disco e y. Fu he analysis e eals ha , he lack o (immedia e)
linea ansmission be ween he coin eg a ed p ices may ha e
inhibi ed bubble synch oniza ion, and caused mo e spo p ice
bubbles. The nonlinea ansmission e ec s be ween he u u es
and spo p ices sugges he exis ence o specula i e s o age and
ma ke powe . This may u he explain why spo p ice bubbles
canno be a bi aged away.
ARTICLE HISTORY
Recei ed 15 May 2023
Accep ed 12 June 2024
KEYWORDS
P ice bubbles; ag icul u al
commodi ies; u u es
ma ke ; spo ma ke
1. In oduc ion
The con o e sy on p ice bubbles in commodi y u u es ma ke s is long las ing
(Gu ie ez, 2013). P ice bubbles associa ed wi h apid and pe sis en p ice inc eases
could dis o ma ke ades since p ices a e he mos impo an signals o ade s
(Phillips e al., 2012). Meanwhile, p ice bubbles in ag icul u al commodi y ma ke s
could gene a e de as a ing consequences. Fo ins ance, in 2007–2008, he nominal p ices
o almos all ood commodi ies inc eased by mo e han 50% and 130 million people in
de eloping coun ies ell in o ex eme po e y (Wo ld Bank, 2008). The impac s o ood
p ice bubbles mainly hu he poo , who spend la ge sha es o hei income on s aple
oods (Tadesse e al., 2014). The public and some schola s end o hink ha ag icul u al
p ice bubbles a e caused by agg essi e inancializa ion o commodi y u u es ma ke s
(Basak & Pa lo a, 2016; Mas e , 2008, 2009; Tang & Xiong, 2012). They a gue ha oo
many ins i u ional unds ha e aken long posi ions in ag icul u al u u es ma ke s
CONTACT Jens-Pe e Loy [email p o ec ed] Depa men o Ag icul u al Economics, Uni e si y o Kiel, Kiel,
Ge many
JOURNAL OF APPLIED ECONOMICS
2024, VOL. 27, NO. 1, 2369441
h ps://doi.o g/10.1080/15140326.2024.2369441
© 2024 The Au ho (s). Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-NonComme cial License (h p://
c ea i ecommons.o g/licenses/by-nc/4.0/), which pe mi s un es ic ed non-comme cial use, dis ibu ion, and ep oduc ion in any medium,
p o ided he o iginal wo k is p ope ly ci ed. The e ms on which his a icle has been published allow he pos ing o he Accep ed
Manusc ip in a eposi o y by he au ho (s) o wi h hei consen .
wi hou conside ing he undamen al alue o ag icul u al commodi ies, and d i e
ag icul u al commodi y p ices up; his u he dis o s p ice expec a ion by comme cial
ade s in u u es ma ke s, who aim o hedge agains p ice isks in spo ma ke s.
Limi ed e idence has been ound o suppo he misp icing e ec s on u u es ma ke s
caused by o e - inancializa ion (Bohl e al., 2021; Boyd e al., 2018); howe e , policymake s
s ill end o es ain he specula i e ades in commodi y u u es ma ke s, when commodi y
p ices inc ease apidly (Mao e al., 2021). Meanwhile, much o he ela ed li e a u e o e looks
he p ice bubbles aking place in spo ma ke s, and p o ides no basis o he synch oniza ion
(o asynch oniza ion) o bubbles be ween ag icul u al u u es and spo ma ke s. A e he e
synch onous p ice bubbles be ween ag icul u al u u es and spo ma ke s? I no , wha may
ha e caused p ice bubbles in ag icul u al spo ma ke s? This s udy seeks o close he esea ch
gap by iden i ying he exac p ice bubble da es i s , and hen in es iga ing he in e ac ion
mode be ween ag icul u al u u es and spo p ices du ing hei bubble pe iods. We ex end he
exis ing empi ical li e a u e on ag icul u al commodi y p ice bubbles by highligh ing he ole
o nonlinea ansmission e ec s ac oss ag icul u al u u es and spo ma ke s. Mo e impo -
an ly, om he es ima es o nonlinea p ice ansmission e ec s, esea che s o en make
in e ences as o he exis ence o ma ke powe o specula i e s o age con olled by some
ma ke pa icipan s (such as e aile s, wholesale s o p oduce s) (Asse a e al., 2017; Loy e al.,
2018; Nakamu a & Ze om, 2010; Sex on, 2013; Sex on & Zhang, 2001). Thus, o iden i y and
analyze commodi y p ice bubbles, mo e a en ion should be paid o he p icing beha iou
and s uc u e o spo ma ke s.
Fu u es ma ke s se e impo an unc ions in p ice disco e y and hedging o ag icul u al
commodi ies. The heo y o s o age p edic s ha he u u es and spo p ices should be
coin eg a ed wi h each o he (Pindyck, 1992; Telse , 1958; Wo king, 1948), which has been
e i ied by nume ous empi ical s udies (C ain & Lee, 1996; Ga bade & Silbe , 1983;
He nandez & To e o, 2010; Ma os & Ga cia, 2004). Thus, when i comes o commodi y
p ice bubbles, a seemingly plausible deduc ion is ha bubbles synch onize be ween he
u u es and spo ma ke s.
Ne e heless, he non-linea i y o p ice ansmission wi hin and ac oss ma ke s calls he
synch oniza ion o bubbles be ween he u u es and spo p ices in o ques ion. Coin eg a ion
ela ionship indica es a common s ochas ic end be ween p ice se ies (Engle & G ange ,
1987). The igh connec ion be ween commodi y u u es and spo p ices is based on he
hypo hesis o linea ansmission be ween hem. Howe e , he immedia e and linea ans-
mission be ween coin eg a ed p ices has long been challenged in eal ma ke s (Loy e al.,
2018).
Some s udies ha e heo e ically p o en ha a coin eg a ion ela ionship be ween
p ices emains e en o bubbles ha occu wi hin one o he coin eg a ed p ice se ies
(Engs ed, 2006; Magdalinos & Phillips, 2009; Nielsen, 2010). In he case o nonlinea
con e gence be ween coin eg a ed p ice se ies, p ices may e en expe ience an explo-
si e beha iou wi hin a band o inac ion (Fan & Wei, 2006). Alexakis e al. (2017)
also doub he di ec ansmission o p ice bubbles wi hin he con ex o he hog
supply chain (hog, co n and soybeans). Based on he coin eg a ion esiduals among
hese h ee commodi y p ices, hey ind ha bubbles in eed p ices and lack o
associa ed bubbles in hog p ices do no a ec he long- un coin eg a ion ela ionship
and ha he hog p ices will e en d ag he o he explosi e p ice episodes back o
no mal. Espos i and Lis o i (2013) conside p ice bubbles as exogenous s uc u al
2Q. MAO ET AL.
b eaks, inding ha p ice bubbles only ha e e y limi ed e ec s on he coin eg a ed
in e na ional and I aly domes ic g ain p ices. Adämme and Bohl (2015) and
Gu ie ez (2013) ind ha specula i e p ice bubbles can a ec he ela ionship
be ween spo and u u es p ices. Li and Xiong (2019) e en ind ha he pe o mance
o p ice disco e y in commodi y u u es ma ke is be e du ing bubble pe iods,
compa ed wi h ha o non-bubble pe iods.
Abo e all, we aim o p o ide new insigh s in o he o ma ion o p ice bubbles in ag icul-
u al commodi y ma ke s by highligh ing he nonlinea ansmission e ec s. We i s de ec
he bubble da es and measu e he deg ee o he bubble synch oniza ion ac oss he u u es and
spo ma ke s o co n and soybeans in China. Then, we use he uni oo es based on
momen um- h eshold au o eg essi e model (M-TAR), he h eshold Vec o E o -
Co ec ion Model (VECM) and he ime- a ying pa ially coin eg a ed Vec o E o -
Co ec ion Model (TV-PC-VECM) o es ima e he non-linea ansmission e ec s and
iden i y he in e ac ion mode be ween he u u es and spo p ices ha ela es wi h equen
bubbles.
Ou s udy is closely ela ed o he li e a u e on he ela ionship be ween inancializa-
ion o commodi ies and p ice bubbles (E ienne e al., 2015, 2017; I win & Sande s, 2012;
Mao e al., 2020; Sande s e al., 2010). In con as , we ex end he s udy o he commodi y
spo ma ke s which ha e been igno ed in p e ious esea ch, and ind ha a ela i ely less
e icien spo ma ke may ha e esul ed in mo e equen bubbles. Ou indings a e also
ela ed o he li e a u e o nonlinea p ice ansmission along supply chain (Azzam, 1999;
Bachmeie & G i in, 2003; Bacon, 1991; Benza i e al., 2020; Loy e al., 2015, 2016,
2018). We poin ou ha specula i e s o age and ma ke powe sugges ed by nonlinea
ansmission e ec s could p e en he spo p ice bubbles om being a bi aged away.
The s uc u e o he pape is as ollows: Sec ion 2 in oduces he heo e ical amewo k
o his pape , including he de ini ion o bubbles and he nonlinea ansmission e ec on
bubbles. Sec ion 3 in oduces he bubble es ing me hod, he uni oo es s based on
M-TAR model, and he TV-PC-VECM model. Sec ion 4 desc ibes he da a. Sec ion 5
p esen s he main es ima ion esul s. In sec ion 6, we summa ize he main indings and
p esen some conclusion.
2. Theo e ical amewo k
2.1. The de ini ion o p ice bubbles
To de ine p ice bubbles, we ollow he s udy o Blancha d and Wa son (1982). The p ice
p ocess o one asse should be:
whe e P ep esen s he p ice a ime , D ep esen s he di idend o payo o he asse a
ime , is he isk- ee in e es a e and E �½ �is he expec a ion based on he in o ma ion se
a ime . Take he con enience yields as he di idends o commodi ies, equa ion (1) can also
be used o explain he o ma ion o commodi y u u es p ice (Pindyck, 2001). Fo wa d
i e a ing equa ion (1) o in ini e pe iods, we can ge he undamen al p ice o he commodi y:
JOURNAL OF APPLIED ECONOMICS 3
only when he ans e sali y condi ion is ul illed, namely he p ice a he in ini e u u e
poin is ze o:
equa ion (2) is he unique solu ion o equa ion (1). Howe e , when equa ion (3) does no
hold, equa ion (2) will no longe be he unique solu ion. Conside a bubble componen B
wi h he p ope y:
adding his B in o equa ion (2) will also sa is y equa ion (1). Tha is
In his case, he bubble componen g ows a a e and he c oss-pe iod non-a bi age
condi ion s ill holds. Thus, he a ional expec a ion o in es o s is no biased and his
kind o p ice bubbles is called as a ional p ice bubbles.
Mo eo e , unde he plausible assump ion ha he di idends would ollow a andom
walk wi h a d i μ.
whe e ε is a whi e noise p ocess. Subs i u ing equa ion (6) in o equa ion (2), we ge
The i s e m o he igh side o equa ion (7) is cons an , while he second e m is a andom
walk based on equa ion (6). Thus, equa ion (7) shows ha he undamen al p ice o he
commodi y should ollow a andom walk, while equa ion (5) shows ha he p ice would
become an explosi e p ocess when he e is a bubble componen B . Fo mo e de ails, please
e e o he s udy o Blancha d and Wa son (1982), Gü kaynak (2008) and Miao (2014).
2.2. Nonlinea p ice ansmission and asynch onous p ice bubbles
As men ioned abo e, he agg essi e inancializa ion o he commodi y u u es ma ke s
has long been conside ed o induce p ice bubbles (Basak & Pa lo a, 2016; Mas e , 2008,
2009; Tang & Xiong, 2012). Too many specula o s en e he u u es ma ke and ake long
posi ions o he u u es con ac s wi hou conside ing he undamen al alue o he
unde lying commodi ies, which may u he dis o he p icing signal and gene a e
bubbles o u u es p ices. These s udies ocus on commodi y u u es ma ke s and
assume ha he spo p ice will simply ollow he u u es p ice p ocess.
Speci ically, he undamen al alues o commodi y u u es p ice in equa ion (2) and
(7) ollow a andom walk p ocess (in eg a ed o o de one, I(1)). When assuming no
4Q. MAO ET AL.
a bi age condi ion be ween he u u es and spo p ices o he same commodi y, he
heo y o s o age, o he cos o ca y heo y indica es ha (Yang e al., 2001, 2021)
whe e p
and ps
a e he (log) u u es and spo p ices, β0 is he cons an e m which could
e lec s all kinds o s o age cos s including anspo a ion, wa ehousing, and insu ance cos s,
and β1 is he slope pa ame e . ec is he esidual pa and becomes he e o -co ec ion e m
in he VECM ep esen a ion when p
and ps
a e coin eg a ed wi h each o he :
whe e Δ is he i s di e ence ope a o , α and αs a e he long- un adjus men pa ame e s
which con ol how quickly he p
and ps
adap o de ia ions om hei long- un equilib ium
ela ionship. Bi is he ma ix o he sho - un adjus men coe icien s.
and s
a e whi e noise
p ocess.
The e o e, linea ansmission be ween p
and ps
sugges ed by ime-in a ian β0 and
β1 in equa ion (8), o α
αs
� �and Bi in equa ion (9) means ha he spo p ice ps
would
ollow he p ocess o p
igh ly. This u he implies bubble synch oniza ion be ween he
u u es and spo p ices.
Howe e , he assump ion o linea ansmission has long been in doub . Fo ins ance,
he “ ocke s and ea he s” p icing beha iou , i.e., p ices ise like ocke s bu all like
ea he s, has been con i med o many ma ke s (Bacon, 1991; Loy e al., 2015; Tappa a,
2009). The di ec ion (sign) o p ice changes could lead o a ious dynamic p ice eac ions
wi h espec o he speed o adjus men and he magni ude o he long- un p ice
equilib ium. Thus, nonlinea p ice ansmission would complica e he ela ionship
be ween he u u es and spo p ices o he same commodi y.
P e ious s udies on commodi y p ice bubbles mainly ocus on he u u es p ice
bubbles and explo e he ela ionship be ween bubbles and specula ion (E ienne e al.,
2015, 2017; Mao e al., 2021; Sande s & I win, 2017). This igno es he pe o mance o
commodi y spo ma ke s. Howe e , gi en he possible nonlinea ansmission e ec ,
bubbles may no synch onize be ween he u u es and spo ma ke s. Especially, he
asymme ic ansmission e ec be ween he u u es and spo p ices may lead o bubbles
in spo ma ke s only. One possible case can be desc ibed by he ollowing equa ion:
whe e he e o co ec ion e m ec is spli in o h ee egimes by wo h esholds θand
θþ. I1
, I2
and I3
a e dummy a iables. I1
¼1 i ec 1<θand ze o o he wise; I2
¼1 i
θ<ec 1<θþand ze o o he wise; I3
¼1 i ec 1>θþand ze o o he wise. Equa ion (10)
JOURNAL OF APPLIED ECONOMICS 5
is commonly used o cap u e he asymme ic cos pass- h ough e ec measu ed by he
di e ence be ween αsþ han αs, whe e αsþmeasu es he adjus men speed o he spo
p ice owa d he long un equilib ium when he u u es p ice inc eases, and αsmeasu es
he adjus men speed o he spo p ice owa d he long un equilib ium when he u u es
p ice dec eases (Loy e al., 2015, 2016; Tappa a, 2009).
I he u u e p ice e lec s he undamen al alue o he unde lying commodi y and
ollows a andom walk as in equa ion (2) and (7), he spo p ice may de ia e om a andom
walk and expe ience bubbles due o he asymme ic ansmission e ec . Highe absolu e
alue o αsþ han αsmeans ha he spo p ice would adjus as e when he u u es p ice
ises compa ed wi h when i alls. This may sugges mo e bubbles o he spo p ice, because
he g ow h o he spo p ice ends o las longe and de ia es om he undamen al alue o
he unde lying commodi y.
F om he es ima es o nonlinea p ice ansmission e ec s and he heo y o “con-
jec u al a ia ions”, esea che s o en make in e ences o he exis ence o ma ke powe
o specula i e s o age con olled by some ma ke pa icipan s (such as e aile s, whole-
sale s o p oduce s) (Asse a e al., 2017; Loy e al., 2018; Nakamu a & Ze om, 2010;
Sex on, 2013; Sex on & Zhang, 2001; Ve e h e al., 2015). Speci ically, ocusing on he
Ge man po k supply chain wi h a me s, slaugh e house, and e aile s, Asse a e al.
(2017) ind ha when eac ing o expo p ices de i ed om compe i i e ma ke s, he
ma ke powe o domes ic slaugh e house would enable hemsel es o inc ease domes ic
po k p ices when compe i i e expo p ices inc ease, bu a oid a la ge domes ic p ice
d op in imes o low expo p ices.
This same logic could also be applied o he ela ionship be ween he u u es p ice and
local spo p ices in he co n and soybeans ma ke s in China. The u u es p ice eme ges
om bidding by all kinds o ade s na ionwide. When he e a e no p ice bubbles, he
u u es p ice can be conside ed as an compe i i e ex e nal p ice o any local spo
ma ke s. Gi en ha he in en o y o co n and soybeans is o a la ge ex en unde he
con ol o local s a e-owned companies (SOCs) in China (Gale, 2013), hese local SOCs
may ha e ce ain ma ke powe o e he supply chain and a ec he imely and e ec i e
adjus men o spo p ices o he u u e p ice. This may u he lead o empo a y
de ia ions o he spo p ices om he long un equilib ium and make oom o bubble
occu ences.
3. Me hodology
3.1. Bubble es ing me hod
The de ini ion o p ice bubbles abo e p o ides he basis o he igh - ailed uni
oo es o iden i y bubbles. Phillips e al. (2012, 2015). de elop he Gene alized
Sup emum-ADF (GSADF) es o da e-s amp p ice bubbles, which has been
widely accep ed o de ec p ice bubbles in a ious ma ke s (Caspi & G aham,
2018; Engs ed e al., 2016; E ienne e al., 2015; Ts e ano e al., 2016). Compa ed
wi h o he bubble es ing me hods (such as he sequen ial Chow- es and CUSUM
es ), he ad an ages o he GSADF me hod a e ha i can iden i y he poin s o
o igina ion and e mina ion o a bubble. Mo eo e , i wo ks sa is ac o ily o p ice
se ies wi h s uc u al b eaks and will no su e om educed powe when
6Q. MAO ET AL.
de ec ing he pe iodically collapsing bubbles (Ha ey e al., 2016; Homm &
B ei ung, 2012; Mao e al., 2021).
Acco ding o Phillips e al. (2015), he GSADF es applies he ADF- es o sequen ial
subse s ( olling windows) o he en i e sample. Suppose ha he olling window uns om
he h
1 ac ion o he o al sample (T) o he h
2 ac ion, whe e 2¼ 1þ w and w>0 is
he ac ional window size o he eg ession. Equa ion (8) shows he empi ical model:
whe e p is he p ice se ies and k is he lag leng h. The ADF-s a is ic alue based on his
eg ession is deno ed as ADF 2
1. The GSADF elies on he epea ed es ima ion o he ADF
es on he subsamples o p ice da a. I a ies he endpoin o he ADF eg ession 2 om 0
( he minimum window wid h) o 1, and i allows he s a ing poin 1 o change wi hin
a easible ange, ha is, om 0 o 2 0. The GSADF- es s a is ic o 2 is hen ob ained as
he sup eme alue o he co esponding ADF-s a is ic sequence (see Equa ion (9)).
The numbe o obse a ions in he model is TW¼T w, whe e :j jis he loo unc ion (gi en
he in ege pa o he a gumen ). The o igina ion da e o a bubble T e is calcula ed as he i s
ch onological obse a ion wi h a GSADF-s a is ic abo e he c i ical alue. The calcula ed
o igina ion da e is deno ed by Tb e. The es ima ed e mina ion da e o a bubble Tb is he i s
ch onological obse a ion a e Tb eþLT wi h a GSADF-s a is ic below he c i ical alue. The
bubble du a ion mus exceed he leng h o log Tð Þ. This equi emen helps o exclude sho
li ed blips in he i ed au o eg essi e coe icien (Phillips e al., 2012). Fo he sample unde
s udy, we calcula e log 460ð Þ ¼ 2:66. Thus, he bubble du a ion should a leas be 3 weeks.
Gu ie ez (2013) and Ha ey e al. (2016) sugges o use he wild boo s ap me hod o
calcula e he c i ical alues, which will conside he unde lying s uc u al b eak o he ime
se ies. The numbe o i e a ions o wild boo s apping in his pape is se a 2000.
3.2. Uni oo es s based on Momen um-Th eshold Au o eg essi e Model (M-TAR)
In o de o es he asymme ic adjus men e ec s o indi idual p ice se ies, we adop he
uni oo es s based on Momen um-Th eshold Au o eg essi e Model (M-TAR).
Acco ding o he wo k o Ende s and G ange (1998), he model is lis ed as ollows:
whe e,
As explained by Ende s and G ange (1998), i ρ1
����<ρ2
����, he M-TAR model exhibi s
li le decay o posi i e p ice e u ns (Δp 1>0) bu subs an ial decay o nega i e p ice
e u ns (Δp 1<0). Namely, o ρ1
����<ρ2
����, i means ha p ice inc eases end o pe sis
JOURNAL OF APPLIED ECONOMICS 7
G ange -causali y es s only ind ha he lagged alues o u u es p ice e u ns p edic spo
p ice e u ns (see Table A2 in he Appendix). Based on hese con en ional es s, he u u es
ma ke i s disco e s p ices and hen ansmi s he signals o he spo ma ke .
Howe e , hese esul s canno explain why he e a e mo e equen bubbles o spo p ices.
Ghosh ay (2018) and Ende s and G ange (1998) poin ou ha he s anda d ADF es
assumes symme ic adjus men s and canno iden i y asymme ic ea u es wi hin ime se ies
da a. They sugges using he uni oo es based on he Momen um-Th eshold
Au o eg essi e Model (M-TAR) o measu e he asymme ic adjus men s o indi idual
p ice se ies. Ou esul s o M-TAR based uni oo es a e lis ed in Table 3. Based on he
esul s o F- es , all p ice e u n se ies a e s a iona y and he alues o ρ1 and ρ2 a e all
signi ican ly di e en om ze o.
5
Fo he u u es p ice e u ns o co n and soybeans ( he i s
and hi d column in Table 3), i shows ha ρ1
����>ρ2
����, which means ha p ice dec eases
would pe sis bu p ice inc eases would e e quickly owa d he long un equilib ium; while
o he spo p ice e u ns ( he second and ou h column in Table 3), he esul ρ1
����<ρ2
����
means ha p ice inc eases would pe sis bu p ice dec eases would e e quickly owa d he
long un equilib ium. Thus, we ind asymme ic adjus men e ec s bo h o he u u es and
spo p ices, bu in di e en di ec ions. The longe las ing upwa d mo emen o he spo
p ices may ha e esul ed in mo e spo p ice bubbles o co n and soybeans.
Mo eo e , h ough h eshold es ima ion on equa ion (10), we measu e he asymme ic
ansmission e ec be ween he u u es and spo p ices, based on he alues o he long un
adjus men pa ame e s αs a di e en in e als. The esul s a e p esen ed in Table 4. Fo
bo h commodi ies, he absolu e alues o αþ
s a e highe han hose o αs, implying ha he
spo p ice o ei he commodi y adjus s as e when i s u u es p ice inc eases compa e wi h
he ime when he u u es p ice alls.
Table 3. TAR based uni oo es .
M-TAR based Uni Roo Tes
Co n: Soybeans:
Fu u es P ice Re u ns:
Δp
1
Spo P ice Re u ns:
Δps
1
Fu u es P ice Re u ns:
Δp
1
Spo P ice Re u ns:
Δps
1
Region 1: (I ¼1)
ρ1 (coe icien o lag.
p ice p 1)
0.0300
(0.0210)
−0.0050*
(0.0030)
−0.0430*
(0.0230)
−0.0070
(0.0080)
cons an −0.1050
(0.068)
0.0110*
(0.0060)
0.1560*
(0.0840)
0.0140
(0.0160)
Region 2: (I ¼0)
ρ2 (coe icien o lag.
p ice p 1)
−0.0220***
(0.0080)
−0.0210***
(0.0080)
−0.0090
(0.0110)
−0.1030***
(0.0290)
cons an 0.0750***
(0.0250)
0.0400**
(0.0170)
0.0340
(0.0410)
0.2080***
(0.0580)
F-Tes 10.8500*** 9.8400*** 4.1500** 13.5000***
*** s a is ically signi ican a 1% con idence le el; **s a is ically signi ican a 5% con idence le el; * s a is ically signi ican
a 10% con idence le el. The c i ical alues o F- es a e ob ained om Ende s and G ange (1998) and hey we e ound
o ha e be e powe han he T- es s a is ics.
Sou ce: own calcula ions based on da a om DCE and he China G ain Rese es G oup, L d. using S a a 15.
5
The c i ical alues o F- es a e ob ained om Ende s and G ange (1998) and hey we e ound o ha e be e powe han
he T- es s a is ics. Thus, based on he F- es , he alues o ρ1 and ρ2 in Table 3 a e all signi ican ly di e en om ze o.
14 Q. MAO ET AL.
Combining wi h he esul s om he M-TAR based uni oo es s, he upwa d
mo emen o he spo p ice ends o adjus as e and las longe , compa ed wi h i s
downwa d mo emen . I he u u es p ice e lec s he undamen al alue o he unde -
lying commodi y and ollows a andom walk p ocess, he asymme ic ansmission e ec
sugges s ha he spo p ice would de ia e om he u u es p ice. This may u he lead o
asynch onous bubbles be ween hem.
5.2.2. Time- a ying p ice ansmission and bubbles
Resul s om p e ious sec ions sugges a possible link be ween he mo e equen bubbles
o commodi y spo p ices and he asymme ic ansmission e ec (wi hin and be ween
p ice se ies). To es whe he he nonlinea ansmission e ec di ec ly ela es wi h spo
p ice bubbles, we con inue o adop he TV-PC-VECM model o de i e he ime- a ying
adjus men pa ame e s and o hogonal impulse esponse unc ions (OIRF), and use he
logi model o es ima e he in e ac ion mode be ween he u u es and spo p ices du ing
he bubble pe iods.
We i s implemen he pa ial coin eg a ion es on he u u es and spo p ices o
each commodi y. Based on he esul s lis ed in panel A o Tables 5 and 6, he e is no
pa ial coin eg a ion ela ionship be ween he u u es and spo p ices. Thus, he pe ma-
nen pa R be ween he u u es and spo p ices in equa ion (15) ends o emain
cons an o e ime. This indica es ha he de ia ions o he spo p ice om he long
un equilib ium a e unlikely o be explained by ime- a ying s o age o anspo a ion
cos s. Meanwhile, since he e is no pa ial coin eg a ion ela ionship, he M e m in he
TV-PC-VECM model o equa ion (19) and (20) will become he no mal e o -co ec ion
e m ec o he VECM model.
S ill, combining he VECM model o equa ion (9) wi h he s a e space me hod
in equa ion (21), we can de i e he ime- a ying adjus men pa ame e s α
αs
� �and
Bi; . Speci ically, he ime- a ying adjus men pa ame e s αs
, b1;s
, b1;ss
, b2;s
, and
b2;ss
o he spo p ices a e p esen ed in Figu es 3 and 4.
6
These ime- a ying
Table 4. Th eshold VECM model.
Asymme ic e ec s: Δps
¼I1
αsec 1þI2
αs0ec 1þI3
αsþec 1þP
k
i¼1
BiΔp
i
Δps
i
� �þ s
Co n: Soybeans:
Down in e al
(ec 1<θ):
αs−0.0340***
(0.0116)
αs0.0211
(0.0280)
Middle in e al
(θ<ec 1<θþ):
α0
s−0.0100
(0.0237)
α0
s−0.0795**
(0.0396)
Up in e al
(ec 1>θþ):
αþ
s−0.0441**
(0.0206)
αþ
s−0.1468***
(0.0263)
*** s a is ically signi ican a 1% con idence le el; **s a is ically signi ican a 5% con idence le el; * s a is ically signi ican
a 10% con idence le el.
Sou ce: own calcula ions based on da a om DCE and he China G ain Rese es G oup, L d. using S a a 15.
6
The lag leng h o he sho un adjus men pa ame e s k = 2 is de e mined by AIC. The ime- a ying adjus men
pa ame e s α
, b1;
, b1; s
, b2;
, and b2; s
o he u u es p ices a e p esen ed in Figu es A1 and A2 in he Appendix.
JOURNAL OF APPLIED ECONOMICS 15
pa ame e s oge he e lec he dynamic in e ac ion p ocess be ween he u u es and spo
p ices. Fo ins ance, he ime- a ying alues o αs
in he i s g aph o Figu es 3 and 4
indica e uns eady mo emen o he spo p ice owa d he long un equilib ium. Fo be e
and in ui i e unde s anding he esponse o he spo p ice o i s own and u u es p ice
shocks, we calcula e he OIRF o each ime poin , see Figu es 5 and 6. “OIRF_s o ”
e e s o he esponse o he spo p ice o he u u es p ice shocks and “OIRF_s os” e e s
o he esponse o he spo p ice o i s own p ice shocks. Compa ing he alues be ween
he “OIRF_s o ” and “OIRF_s os”, we can easily see ha he esponse o he spo p ice o
i s own shocks a e always la ge han ha o i s u u es p ice shocks.
Wi h hese ime- a ying pa ame e s, we i s use he logi model o es whe he he
nonlinea ansmission e ec di ec ly ela es wi h he spo p ice bubbles. The esul s o
he logi model a e p esen ed in panel B o Tables 5 and 6. We i s ind no e ec s o he
specula ion on he spo p ice bubbles o bo h commodi ies and highe liquidi y ends o
educe p ice bubbles o co n. This is consis en wi h he e idence om expe imen al
economics which shows ha u u es ma ke s dampen, hough do no elimina e p ice
bubbles (Po e & Smi h, 2003).
Meanwhile, he e ec s o hese ime- a ying pa ame e s a e signi ican , bu di icul o
in e p e . Fo co n, he long un adjus men pa ame e αs
has a signi ican posi i e e ec
on i s spo p ice bubbles, while he αs
o soybeans has no signi ican e ec s. The sho
Table 5. TV-PC-VECM model o co n.
Pa ial coin eg a ion es o co n u u es and spo p ices: Tes s a is ics p- alue
Panel A:
HR
0: esidual se ies ollows a pu e uni oo p ocess (no coin eg a ion) −20.3200 0.0000***
HM
0: esidual se ies ollows a pu e AR(1) p ocess (linea coin eg a ion) −1.0300 0.1526
Logi model o spo p ice bubbles
Bubble Bubble Bubble Bubble
Panel B:
Cons an −1.4892***
(0.5123)
−0.7313***
(0.2644)
−0.9080***
(0.2929)
−0.9100***
(0.2919)
Specula ion −0.3999
(0.3466)
−0.6205
(0.4170)
−0.4652
(0.3871)
−0.4660
(0.3882)
ln T ade olume
ð Þ −0.1724***
(0.0413)
−0.1165***
(0.0361)
−0.1314***
(0.0382)
−0.1310***
(0.0380)
αs
429.1173****
(150.9391)
b1;s ; −10.4752***
(2.1626)
b1;ss; 3.3089*
(1.6969)
b2;s ; 11.4598***
(3.471038)
b2;ss; −18.1662***
(2.4944)
OIRF _s o
( esponse o spo p ice o u u es shocks)
−3.1908
(11.4757)
1.2804
(11.8599)
OIRF _s os ( esponse o spo p ice o own shocks) 17.4735***
(4.1894)
17.5108***
(4.2307)
Obse a ions 456 456 456 456
*** s a is ically signi ican a 1% con idence le el; **s a is ically signi ican a 5% con idence le el; * s a is ically signi ican
a 10% con idence le el. The lag leng h o p ice e u ns a e de e mined by in o ma ion c i e ia (AIC).
Sou ce: own calcula ions based on da a om DCE and he China G ain Rese es G oup, L d. using S a a 15.
16 Q. MAO ET AL.
un adjus men pa ame e s b1;s
, b1;ss
, b2;s
, and b2;ss
show mo e complica ed e ec s o
bo h commodi ies. Thus, we eplace hese pa ame e s wi h he OIRF, which is a mo e
in ui i e indica o o he nonlinea e ec s.
Fo co n, he esul s in he second and ou h column o panel B in Table 5 show ha
he OIRF caused by u u es p ice shocks has no signi ican e ec s on he spo p ice
bubbles, while he OIRF caused by spo p ice own shocks has signi ican posi i e e ec s
on bubbles. This sugges s ha he (o hogonal) u u es p ice shocks a leas don’
con ibu e o spo p ice bubbles. In o he wo ds, u u es p ice inc eases alone canno
lead o mo e spo p ice bubbles, while highe esponse o he spo p ice o i s own shocks
con ibu es o mo e spo p ice bubbles. Mo eo e , p e ious esul s om he M-TAR
based uni oo es s and he h eshold VECM model indica e ha he spo p ice inc eases
end o adjus as e and las longe when he u u e p ice inc eases. Ou esul s based on
he logi model u he p o e he nonlinea adjus men e ec is signi ican in explaining
he o igin o spo p ice bubbles. Thus, he spo p ice ails o ollow wi h he u u es p ice
igh ly. Once a upwa d end o momen um is es ablished o he spo p ice, i is mo e
likely o con inue in ha di ec ion han o mo e agains he end.
Fo soybeans, we ind almos he same esul s as co n, excep ha he coe icien s o
he OIRF caused by spo p ice own shocks a e posi i e, bu i is only e y close o he 10%
Table 6. TV-PC-VECM model o soybeans.
Pa ial coin eg a ion es o soybeans u u es and spo p ices: Tes s a is ics p- alue
Panel A:
HR
0: esidual se ies ollows a pu e uni oo p ocess (no coin eg a ion) −11.2400 0.0000***
HM
0: esidual se ies ollows a pu e AR(1) p ocess (linea coin eg a ion) −0.0000 1.0000
Logi model o spo p ice bubbles
Bubble Bubble Bubble Bubble
Panel B:
Cons an −4.1051***
(0.6487)
−1.9670***
(0.3106)
−2.0238***
(0.3079)
−2.0111***
(0.3102)
Specula ion 0.0090
(0.0372)
−0.0444
(0.0697)
−0.0426
(0.0607)
−0.0438
(0.0632)
ln T ade olume
ð Þ 0.0158
(0.0636)
−0.0043
(0.0459)
0.0047
(0.0449)
−0.0002
(0.0463)
αs
96.2165
(125.3267)
b1;s
9.3231***
(2.9634)
b1;ss
−6.9644***
(1.4290)
b2;s
−7.5480**
(2.9519)
b2;ss
3.3498*
(1.7940)
OIRF _s o
( esponse o spo p ice o u u es shocks)
−2.0693
(1.6469)
1.8019
(1.6922)
OIRF _s os
( esponse o spo p ice o own shocks)
5.3931
(3.7605)
4.8610
(3.8646)
Obse a ions 456 456 456 456
*** s a is ically signi ican a 1% con idence le el; **s a is ically signi ican a 5% con idence le el; * s a is ically signi ican
a 10% con idence le el. The lag leng h o p ice e u ns a e de e mined by in o ma ion c i e ia (AIC).
Sou ce: own calcula ions based on da a om DCE and he China G ain Rese es G oup, L d. using S a a 15.
JOURNAL OF APPLIED ECONOMICS 17
Figu e 3. Co n: ime- a ying long un and sho un adjus men pa ame e s o he spo p ice. Sou ce:
own calcula ions based on da a om DCE and he China G ain Rese es G oup, L d. using S a a 15.
Figu e 4. Soybeans: ime- a ying long un and sho un adjus men pa ame e s o he spo p ice.
Sou ce: own calcula ions based on da a om DCE and he China G ain Rese es G oup, L d. using S a a
15.
18 Q. MAO ET AL.
signi icance le el. Fu u e s udy using mo e de ailed o disagg ega ed p ice da a may be
needed o iden i y and analyze he easons o soybeans p ice bubbles.
These esul s abo e indica e ha he adjus men e ec o spo p ices owa d he long-
un equilib ium becomes weak when spo p ice bubbles occu . Spo p ices can ha dly
adjus o a new ma ke clea ing p ice le el when esponding o u u e p ice changes. Ou
esul s a e also consis en wi h p e ious s udies, which p o e a di e ence be ween he
commodi y spo and u u es ma ke s in he abili y o inco po a e ele an p ice in o ma-
ion (C ain & Lee, 1996; Yang e al., 2001). The sel -pe sis ence o p ice e u ns du ing
p ice inc easing p ocesses may con ibu e o mo e spo p ice bubbles.
Mo eo e , om he es ima es o nonlinea p ice ansmission e ec s, esea che s
o en make in e ences as o he exis ence o ma ke powe o specula i e s o age
con olled by some ma ke pa icipan s (such as e aile s, wholesale s o p oduce s)
(Loy e al., 2018; Nakamu a & Ze om, 2010; on C amon-Taubadel & Goodwin,
2021).
7
S o age by specula o s can be expec ed o mo e commodi y om pe iods o
low p ices o pe iods o high p ices, hus inducing au oco ela ion and asymme y
momen um o p ice adjus men s (Dea on, 1999; Dea on & La oque, 1996), which is
consis en wi h ou esul s om he M-TAR based uni oo es s and h eshold VECM
model.
In addi ion, p ices end o be highe in less compe i i e ma ke s and ma ke pa ici-
pan s wi h ma ke powe could keep “p ice going up bu no coming down” (Asse a e al.,
2017; Benza i e al., 2020). Gi en ha he in en o y o co n and soybeans is o a la ge
ex en unde he con ol o s a e-owned companies (SOCs) in China (Gale, 2013), hese
SOCs ha e a signi ican ma ke powe o e he supply chain and could a ec he imely
Figu e 5. Co n: ime- a ying OIRF. Sou ce: own calcula ions based on da a om DCE and he China
G ain Rese es G oup, L d. using S a a 15.
7
See he li e a u e e iew on p ice ansmission by on C amon-Taubadel and Goodwin (2021).
JOURNAL OF APPLIED ECONOMICS 19
and e ec i e adjus men o ag icul u al spo p ices. Though hese SOCs a e buil o
s abilize he ag icul u al ma ke s, hey also assume sole esponsibili y o hei own
p o i s o losses. I is plausible ha hese SOCs wi h signi ican ma ke powe may
“ ide he bubbles” empo a ily o gain mo e p o i s. In his case, spo p ice bubbles
canno be ully a bi aged away in a sho ime (Wang & Tomek, 2007). Simila
phenomenon o “ iding he bubbles” has also been p o ed in he equi y ma ke (Temin
& Vo h, 2004). Compa ed wi h p e ious esea ch me ely ocusing on he des abilizing
e ec s o u u es ins i u ional in es o s, ou es ima ion esul s sugges ha ag icul u al
p ice bubbles may also easily occu in a ma ke whe e some ma ke pa icipan s hold
signi ican ma ke powe .
6. Conclusions
P e ious s udies on ag icul u al p ice bubbles ha e mos ly igno ed spo ma ke s and
ag icul u al u u es ma ke s ha e been blamed o hei po en ially nega i e e ec s o
o e - inancializa ion. Ou s udy aims o iden i y and analyze he p ice bubbles in
ag icul u al commodi y ma ke s, highligh ing he nonlinea ansmission ac oss u u es
and spo ma ke s.
We i s iden i y he bubble da es o he wo highly aded ag icul u al commodi ies
in China, co n and soybeans. Limi ed synch oniza ion o bubbles ac oss he ag icul u al
u u es and spo ma ke s is ound, and he spo p ice se ies shows mo e equen and
du able bubbles han he u u es p ices. This may imply ha commodi y u u es ma ke s
p o ide a mo e e ec i e p ice signal o ade s. We use he M-TAR based uni oo es s
and h eshold VECM model o cap u e he nonlinea (asymme ic) p ice ansmission
wi hin and ac oss he u u es and spo ma ke s. The spo p ice indica es a s ong sel -
Figu e 6. Soybeans: ime- a ying OIRF. Sou ce: own calcula ions based on da a om DCE and he
China G ain Rese es G oup, L d. using S a a 15.
20 Q. MAO ET AL.
pe sis ence o i s upwa d p ocess. Meanwhile, we ind a signi ican asymme ic ansmis-
sion e ec be ween he u u es and spo p ices. Thus, he spo p ice adjus s as e o i s
u u e p ice inc eases han dec eases. A quick and las ing esponse o he spo p ice o he
u u es and i s own p ice inc eases may ha e esul ed in mo e bubbles o he spo
ma ke .
We u he es whe he he nonlinea ansmission e ec di ec ly ela es wi h mo e spo
p ice bubbles h ough he Time- a ying Pa ially Coin eg a ed VECM model and he logi
model. The ime- a ying (long un and sho un) adjus men pa ame e s and OIRF alues
es ima ed om he Time- a ying Pa ially Coin eg a ed VECM model a e used o cap u e
he dynamic in e ac ion mode be ween he u u es and spo p ices a each pe iod. The
esul s o he logi model show ha hese ime- a ying indica o s ha e signi ican e ec s on
spo p ice bubbles, and highe esponse o he spo p ice o i s own p ice shocks con ibu es
o mo e bubbles. Thus, consis en wi h he esul s om he M-TAR based uni oo es s
and he h eshold VECM model, spo p ice bubbles a e mo e likely o occu when he spo
p ice adjus s quickly o i s u u es p ice and keeps his upwa d end longe . Fu u es p ice
inc eases alone canno lead o mo e spo p ice bubbles.
Abo e all, hough he e is a coin eg a ion ela ionship be ween he u u es and spo
p ices, bubbles occu mo e equen ly o he spo p ice se ies, which shows highe sel -
pe sis ence o inc easing e u ns du ing he upwa d end. This u he implies poo abili y
o he spo ma ke o adjus i sel o a new equilib ium. Specula i e s o age and impe ec ly
compe i i e (spo ) ma ke s uc u e may accoun o his slow and asymme ic p ice
adjus men s and mo e spo p ice bubbles. Mo eo e , ou conclusion is limi ed due o
ha we only ha e agg ega ed p ice se ies o spo p ices and in e he exis ence o he
ma ke powe based on he nonlinea ansmission e ec . Fu u e s udy using disagg ega ed
spo p ice da a o be e measu es on he ma ke powe o ag icul u al ma ke pa icipan s
could gi e a mo e in ui i e explana ion on he ela ionship among he nonlinea ansmis-
sion e ec , p ice bubbles and ma ke powe .
Disclosu e s a emen
No po en ial con lic o in e es was epo ed by he au ho (s).
Funding
This wo k was suppo ed by he Educa ion O ice o Zhejiang P o ince [P ojec numbe :
Y202248790], and Zhejiang Uni e si y o Technology [P ojec numbe : SKY-ZX-20220258 and
GB202301003].
No es on con ibu o s
Qianqian Mao, PhD, Assis an P o esso a School o Economics, Zhejiang Uni e si y o
Technology (China). His email is [email p o ec ed]. His esea ch opics co e ag icul-
u al ma ke analysis (p ice ansmission, ma ke powe and ma ke in eg a ion), u u es ma ke s,
ime se ies analysis, and beha io al economics.
JOURNAL OF APPLIED ECONOMICS 21
Yanjun Ren, PhD, Full P o esso in Ag icul u al and Food Economics a No hwes A&F
Uni e si y (China). His esea ch opics co e ood (nu i ion) economics, ag icul u al economics,
causal in e ence, and beha io al economics.
Jens-Pe e Loy, PhD, co esponden au ho . Full P o esso a Depa men o Ag icul u al
Economics, Kiel Uni e si y (Ge many). His email is [email p o ec ed]. His esea ch in e es s
co e ma ke ing, p icing and policy issues on ma ke s along he ag icul u al and ood alue chain
wi h a ocus on e ail ma ke s. Some ecen opics a e e ical p ice ansmission, he measu emen
o ma ke powe in he ood indus y, p ice se ing in he ood e ail sec o and economic
expe imen s o auc ions, and p edic ion ma ke s.
Re e ences
Adämme , P., & Bohl, M. T. (2015). Specula i e bubbles in ag icul u al p ices. The Qua e ly
Re iew o Economics & Finance, 55(2015), 67–76. h ps://doi.o g/10.1016/j.q e .2014.06.003
Adämme , P., & Bohl, M. T. (2018). P ice disco e y dynamics in Eu opean ag icul u al ma ke s.
Jou nal o Fu u es Ma ke s, 38(5), 549–562. h ps://doi.o g/10.1002/ u .21891
Alexakis, C., Bagna osa, G., & Dowling, M. (2017). Do coin eg a ed commodi ies bubble oge he ?
The case o hog, co n, and soybean. Finance Resea ch Le e s, 23, 96–102. h ps://doi.o g/10.
1016/j. l.2017.02.007
Asse a, T. T., Meuwissen, M. P. M., Ga deb oek, C., & Oude Lansink, A. G. J. M. (2017). P ice and
ola ili y ansmission and ma ke powe in he Ge man F esh Po k supply chain. Jou nal o
Ag icul u al Economics, 68(3), 861–880. h ps://doi.o g/10.1111/1477-9552.12220
Azzam, A. M. (1999). Asymme y and igidi y in a m- e ail p ice ansmission. Ame ican Jou nal
o Ag icul u al Economics, 81(3), 525–533. h ps://doi.o g/10.2307/1244012
Bachmeie , L. J., & G i in, J. M. (2003). New e idence on asymme ic gasoline p ice esponses.
The Re iew o Economics and S a is ics, 85(3), 772–776. h ps://doi.o g/10.1162/
003465303322369902
Bacon, R. W. (1991). Rocke s and ea he s: The asymme ic speed o adjus men o UK e ail
gasoline p ices o cos changes. Ene gy Economics, 13(3), 211–218. h ps://doi.o g/10.1016/
0140-9883(91)90022-R
Basak, S., & Pa lo a, A. (2016). A model o inancializa ion o commodi ies. The Jou nal o
Finance, 71(4), 1511–1556. h ps://doi.o g/10.1111/jo i.12408
Benza i, Y., Ca loni, D., Ha ju, J., & Kosonen, T. (2020). Wha goes up may no come down:
Asymme ic incidence o alue-added axes. Jou nal o Poli ical Economy, 128(12), 4438–4474.
h ps://doi.o g/10.1086/710558
Blancha d, O. J., & Wa son, M. W. (1982). Bubbles, a ional expec a ions and inancial ma ke s
(Wo king Pape 945). Na ional Bu eau o Economic Resea ch. h ps://doi.o g/10.3386/w0945
Bohl, M. T., B ange , N., & T ede, M. (2021). Measu emen e o s in index ade posi ions da a: Is
he p ice p essu e hypo hesis s ill in alid? Applied Economic Pe spec i es and Policy, 44(3),
1534–1553. h ps://doi.o g/10.1002/aepp.13186
Boyd, N. E., Ha is, J. H., & Li, B. (2018). An upda e on specula ion and inancializa ion in
commodi y ma ke s. Jou nal o Commodi y Ma ke s, 10, 91–104. h ps://doi.o g/10.1016/j.
jcomm.2018.05.005
Caspi, I., & G aham, M. (2018). Tes ing o bubbles in s ock ma ke s wi h i egula di idend
dis ibu ion. Finance Resea ch Le e s, 26, 89–94. h ps://doi.o g/10.1016/j. l.2017.12.015
Clegg, M., & K auss, C. (2018). Pai s ading wi h pa ial coin eg a ion. Quan i a i e Finance, 18
(1), 121–138. h ps://doi.o g/10.1080/14697688.2017.1370122
C ain, S. J., & Lee, J. H. (1996). Vola ili y in whea spo and u u es ma ke s, 1950–1993:
Go e nmen a m p og ams, seasonali y, and causali y. The Jou nal o Finance, 51(1),
325–343. h ps://doi.o g/10.1111/j.1540-6261.1996. b05211.x
Dea on, A. (1999). Commodi y p ices and g ow h in A ica. Jou nal o Economic Pe spec i es, 13
(3), 23–40. h ps://doi.o g/10.1257/jep.13.3.23
22 Q. MAO ET AL.
Dea on, A., & La oque, G. (1996). Compe i i e s o age and commodi y p ice dynamics. Jou nal o
Poli ical Economy, 104(5), 896–923. h ps://doi.o g/10.1086/262046
Ende s, W., & G ange , C. W. J. (1998). Uni - oo es s and asymme ic adjus men wi h an
example using he e m s uc u e o in e es a es. Jou nal o Business & Economic S a is ics, 16
(3), 304–311. h ps://doi.o g/10.1080/07350015.1998.10524769
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 es ing. Econome ica, 55(2), 251–276. JSTOR. h ps://doi.o g/10.2307/1913236
Engs ed, T. (2006). Explosi e bubbles in he coin eg a ed VAR model. Finance Resea ch Le e s, 3
(2), 154–162. h ps://doi.o g/10.1016/j. l.2006.03.004
Engs ed, T., H iid, S. J., & Pede sen, T. Q. (2016). Explosi e bubbles in house p ices? E idence
om he OECD coun ies. Jou nal o In e na ional Financial Ma ke s, Ins i u ions and Money,
40, 14–25. h ps://doi.o g/10.1016/j.in in.2015.07.006
Espos i, R., & Lis o i, G. (2013). Ag icul u al p ice ansmission ac oss space and commodi ies
du ing p ice bubbles. Ag icul u al Economics, 44(1), 125–139. h ps://doi.o g/10.1111/j.1574-
0862.2012.00636.x
E ienne, X. L., I win, S. H., & Ga cia, P. (2015). P ice explosi eness, specula ion, and g ain u u es
p ices. Ame ican Jou nal o Ag icul u al Economics, 97(1), 65–87. h ps://doi.o g/10.1093/ajae/
aau069
E ienne, X. L., I win, S. H., & Ga cia, P. (2017). New e idence ha index ade s did no d i e
bubbles in g ain u u es ma ke s. Jou nal o Ag icul u al and Resou ce Economics, 1(23), 45–67.
h ps://doi.o g/10.22004/ag.econ.252754
Fan, C. S., & Wei, X. (2006). The law o one p ice: E idence om he ansi ional economy
o China. The Re iew o Economics and S a is ics, 88(4), 682–697. h ps://doi.o g/10.1162/
es .88.4.682
Fishe , T. C. G., & Konieczny, J. D. (2000). Synch oniza ion o p ice changes by mul ip oduc
i ms: E idence om Canadian newspape p ices. Economics Le e s, 68(3), 271–277. h ps://
doi.o g/10.1016/S0165-1765(00)00231-7
Gale, F. (2013). G ow h and e olu ion in China’s ag icul u al suppo policies (Economic Resea ch
Repo 153). Uni ed S a es Depa men o Ag icul u e.
Ga bade, K. D., & Silbe , W. L. (1983). P ice mo emen s and p ice disco e y in u u es and cash
ma ke s. The Re iew o Economics and S a is ics, 65(2), 289–297. h ps://doi.o g/10.2307/
1924495
Ghosh ay, A. (2018). A e shocks ansi o y o pe manen ? An inqui y in o ag icul u al commodi y
p ices. Jou nal o Ag icul u al Economics, 70(1), 26–43. h ps://doi.o g/10.1111/1477-9552.
12277
Gü kaynak, R. S. (2008). Econome ic es s o asse p ice bubbles: Taking s ock*. Jou nal o
Economic Su eys, 22(1), 166–186. h ps://doi.o g/10.1111/j.1467-6419.2007.00530.x
Gu ie ez, L. (2013). Specula i e bubbles in ag icul u al commodi y ma ke s. Eu opean Re iew o
Ag icul u al Economics, 40(2), 217–238. h ps://doi.o g/10.1093/e ae/jbs017
Ha ey, D. I., Leybou ne, S. J., Sollis, R., & Taylo , A. M. R. (2016). Tes s o explosi e inancial
bubbles in he p esence o non-s a iona y ola ili y. Jou nal o Empi ical Finance, 38, 548–574.
h ps://doi.o g/10.1016/j.jemp in.2015.09.002
He nandez, M. A., & To e o, M. (2010). Examining he dynamic ela ion be ween spo and u u es
p ices o ag icul u al commodi ies. FAO Commodi y Ma ke Re iew, No.2009/2010, 47–86.
Homm, U., & B ei ung, J. (2012). Tes ing o specula i e bubbles in s ock ma ke s: A compa ison
o al e na i e me hods. Jou nal o Financial Econome ics, 10(1), 198–231. h ps://doi.o g/10.
1093/jj inec/nb 009
I win, S. H., & Sande s, D. R. (2012). Tes ing he mas e s hypo hesis in commodi y u u es
ma ke s. Ene gy Economics, 34(1), 256–269. h ps://doi.o g/10.1016/j.eneco.2011.10.008
Li, M., & Xiong, T. (2019). Do bubbles al e con ibu ions o p ice disco e y? E idence om he
Chinese Soybean Fu u es and spo ma ke s. Eme ging Ma ke s Finance & T ade, 55(15),
3417–3432. h ps://doi.o g/10.1080/1540496X.2019.1608178
JOURNAL OF APPLIED ECONOMICS 23
