F on u o, Vi o; Felici, Tommaso; And eoli, Vania; Bagliani, Ma co Ma ia; Co si,
Alessand o
A icle
Is he e an animal ood kuzne s cu e, and does i ma e ?
Bio-based and Applied Economics (BAE)
P o ided in Coope a ion wi h:
Fi enze Uni e si y P ess
Sugges ed Ci a ion: F on u o, Vi o; Felici, Tommaso; And eoli, Vania; Bagliani, Ma co Ma ia; Co si,
Alessand o (2025) : Is he e an animal ood kuzne s cu e, and does i ma e ?, Bio-based and
Applied Economics (BAE), ISSN 2280-6172, Fi enze Uni e si y P ess, Flo ence, Vol. 14, Iss. 1, pp.
19-35,
h ps://doi.o g/10.36253/bae-16172
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/321808
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/4.0/
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
© 2025 Au ho (s). Open access a icle published, excep whe e o he wise no ed, by Fi enze Uni e si y P ess unde CC-BY-4.0 License
o con en and CC0 1.0 Uni e sal o me ada a.
Fi enze Uni e si y P ess | www. up ess.com/bae
Bio-based and Applied Economics
BAE
Ci a ion: F on u o, V., Felici, T.,
And eoli, V., Bagliani, M. M. & Co -
si, A. (2025). Is he e an Animal Food
Kuzne s Cu e, and does i ma e ?
Bio-based and Applied Economics 14(1):
19-35. doi: 10.36253/bae-16172
Recei ed: May 24, 2024
Accep ed: Sep embe 23, 2024
Published: June 13, 2025
Da a A ailabili y S a emen : All el-
e an da a a e wi hin he pape and i s
Suppo ing In o ma ion iles.
Compe ing In e es s: The Au ho (s)
decla e(s) no con lic o in e es .
Associa ed Edi o : O iana Ga a
Edi o in Chie : Sil ia Code oni
ORCID
VF: 0000-0002-0627-9511
TF: 0000-0002-3779-4225
VA: 0000-0001-9163-011X
MMB: 0000-0003-4918-8852
AC: 0000-0002-8519-6721
Is he e an Animal Food Kuzne s Cu e, and
does i ma e ?
Vi o F on u o1,*, Tommaso Felici2, Vania And eoli3, Ma co Ma ia
Bagliani1,#, Alessand o Co si1
1 Depa men o Economics and S a is ics, Uni e si à degli S udi di To ino, To ino 10153,
I aly
2 U ech Uni e si y School o Economics (U.S.E), K iekenpi plein 21-22, 3584 EC U e-
ch , The Ne he lands
3 School o Biological Sciences, Uni e si y o Wes e n Aus alia,35 S i ling Hwy, C awley,
WA 6009, Aus alia
# Sadly, P o . Ma co Ma ia Bagliani, passed away in July 2024 while his pape was
unde e ision. We wish o dedica e his pape o a dea colleague and iend.
*Co esponding au ho . E-mail add ess: i o. on u o@uni o.i
Abs ac . P o eins om animal sou ces, including mea , and plan -based oods a e
essen ial o a heal hy human die . Howe e , animal-based p o eins ha e signi ican ly
highe en i onmen al impac s (e.g., g eenhouse gas emissions, de o es a ion, and wa e
usage) and heal h isks (e.g., obesi y, ype 2 diabe es, kidney s ones and ca dio ascula
diseases) compa ed o plan -based p o eins. The consump ion pa e ns o hese p o-
eins a e s ongly in luenced by income le els. This s udy in oduces he concep o
an Animal Food Kuzne s Cu e by sys ema ically analyzing he ela ionship be ween
income and animal-based p o ein consump ion. U ilizing a no el panel da ase span-
ning 28 yea s and co e ing 79 coun ies, we unco e an in e ed U-shaped ela ion-
ship be ween income and he consump ion o animal-based and mea p o eins. Ou
indings indica e ha he u ning poin s occu a ound 43,000-45,000 US$, co espond-
ing o he 90 h and 95 h pe cen iles o he pe capi a income dis ibu ion in he sam-
ple. A hese income le els, p o ein consump ion is es ima ed a app oxima ely 25 g/
day o mea and 52 g/day o animal-based p o eins, as compa ed o ecommended
o al p o ein in ake o 45-56 g/day. These insigh s highligh he c i ical need o a -
ge ed policy in e en ions, such as axes, nudges, and in o ma ional campaigns o p o-
mo e sus ainable die a y choices ac oss all income le els. Ou s udy p o ides empi i-
cal e idence o he impo ance o in eg a ing economic and en i onmen al policies o
enhance global ood sus ainabili y.
Keywo ds: p o ein consump ion, consump ion d i e s, En i onmen al Kuzne s
Cu e, mixed e ec s model, panel da a.
JEL Codes: Q54, Q56, C23.
1. INTRODUCTION
O e he pas 50 yea s, he global consump ion o animal-based p o-
eins, including mea , eggs, dai y, and sea ood, has signi ican ly inc eased in
20
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
Vi o F on u o e al.
bo h absolu e and pe capi a e ms (Bonne e al., 2020;
Ma ques e al., 2018; Pais e al., 2021). This g ow h has
been mainly d i en by inc eased mea consump ion
(Bonne e al., 2020; Sans and Comb is, 2015). Acco d-
ing o OECD and FAO (2023), global pe capi a mea
consump ion has nea ly doubled, ising by 87 pe cen
om abou 23 kg pe pe son pe yea in 1961 o 43 kg
pe pe son pe yea in 2021. Simila ly, o he animal-
based oods ha e seen inc eases, wi h milk consump-
ion by 16 pe cen and egg consump ion by 129 pe cen .
This endency is in acco dance wi h he nu i ion an-
si ion ea u ing inc easing demand o animal-based
oods when income ises (Popkin, 1993). Howe e , die s
ich in animal-based p o ein ha e been linked o ad e se
heal h and en i onmen al ou comes (Tilman and Cla k,
2014), while die s wi h a highe composi ion o plan -
based p o eins a e associa ed wi h less damaging impac s
(Galli and Mo e i, 2024). Se e al s udies ha e called o
u gency in shi ing p o ein consump ion om animal-
based sou ces o plan -based sou ces (Wille e al., 2019),
especially in uppe -middle income coun ies wi h sus-
ained economic g ow h a es (Du o e al., 2020). Indeed,
he inc ease in global mea consump ion (kg/yea pe
capi a) be ween 1961 and 2021 has been d i en mainly
by coun ies wi h apid economic g ow h such as Sou h
Ko ea (1,935 pe cen ), China (1,774 pe cen ) and Indone-
sia (398 pe cen ). Se e al s udies ha e demons a ed ha
he i s global p o ein ansi ion, ma ked by a signi ican
inc ease in demand o animal-based p o ein o e he
las cen u y, was closely linked o changes in eal income
(Sans and Comb is, 2015). The mo e ecen second nu i-
ion ansi ion, cha ac e ized by a s abiliza ion o decline
in animal-based p o ein consump ion, pa icula ly mea
(God ay e al., 2018; Ma ques e al., 2018; V anken e al.,
2014), may also be a ibu ed o simila ac o s. Economic
g ow h has ini ially p omo ed animal-based consump ion
and hen i has slowed i down. This b ough some schol-
a s o claim he exis ence o an En i onmen al Kuzne s
Cu e (EKC) o animal-based ood consump ion, which
could be named Animal Food Kuzne s Cu e (AFKC).
Acco ding o he EKC o iginal heo y, he en i onmen al
impac o economic g ow h inc eases in he i s phase
and subsequen ly declines (G ossman, 1995; G oss-
man and K uege , 1991). I such a end p o ed ue
o animal-based ood consump ion, i would dec ease
he u gency o policies aiming a cu bing i s consump-
ion since income g ow h would au oma ically lead o i s
decline. Ne e heless, he exis ence o an AFKC is o be
empi ically e i ied, and i s ac ual e ec on global con-
sump ion is o be assessed.
This pape aims a in es iga ing in e ac ions
be ween p o ein consump ion and income o e he las
30 yea s. The esea ch uniquely analyses p o ein in ake
om animal-based, mea and plan -based sou ces o
unde s and he dynamics o change and he p edomi-
nan ac o o a ia ion, i.e., income. While he exis ing
li e a u e has p edominan ly ocused on mea consump-
ion and i s co ela ion wi h income (Yo k and Gossa d,
2004; V anken e al., 2014), he e is a no iceable gap
conce ning he consump ion o p o ein om di e en
sou ces. This pape aims o b idge his gap by comp e-
hensi ely explo ing di e ences in p o ein consump ion
ac oss animal-based, mea and plan -based sou ces using
a global panel da ase co e ing 28 yea s and 79 coun-
ies. The o iginali y o his s udy is u he highligh ed
by he applica ion o he linea mixed e ec model. This
me hodological ad ancemen add esses c oss-sec ional
dependence in e o s wi hin la ge panel da ase s, hus
enhancing he accu acy o pa ame e es ima es com-
pa ed o con en ional ixed e ec s models.
2. NEGATIVE IMPACTS OF ANIMAL-
BASED PROTEIN CONSUMPTION
Animal-based p oduc s a e an essen ial sou ce o
nu ien s – p o eins, among o he s – o humans. How-
e e , among p o ein- ich oods, hose o animal-based
sou ces p oduce highe g eenhouse gas (GHG) emissions
(Dye and Desja dins, 2022; E ickson e al., 2021), use
mo e land (Van Zan en e al., 2018) and wa e (Mekon-
nen and Ge bens-Leenes, 2020), cause mo e acidi ica-
ion and eu ophica ion (God ay e al., 2018; Poo e and
Nemecek, 2018). Among animal-based oods, mea has
a highe en i onmen al damage po en ial han hose
de i ed om eggs, milk and sea ood (de V ies and de
Boe , 2010). Among mea s, bee p o eins ha e he highes
impac on he en i onmen (de V ies and de Boe , 2010;
Gaillac and Ma bach, 2021).
The e is an u gen need o ansi ioning o mo e
sus ainable p o ein sou ces, such as p o ein o ege al
sou ces – pulses, legumes and no el p o ein- ich oods
(McClemen s and G ossmann, 2021) – which ha e a
lowe en i onmen al impac (Mazac e al., 2022). Plan -
based die s can educe GHG emissions by 49%, land
use by 76%, sca ci y-weigh ed eshwa e wi hd awals
by 19%, acidi ica ion by 50% and eu ophica ion by 49%
(Poo e and Nemecek, 2018).
Ano he eason o educe consump ion o animal-
based p oduc s, pa icula ly mea , is ela ed wi h he
po en ial ad e se e ec s o i s excessi e consump ion on
human heal h. A highe a ailabili y o animal-based p o-
ein consump ion would bene i ood-insecu e coun ies,
whe e ewe al e na i es a e a ailable o access nu ien s
21
Is he e an Animal Food Kuzne s Cu e, and does i ma e ?
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
and mic onu ien s. He e, a highe animal-based p o-
ein consump ion would inc ease ood and nu i ional
secu i y. By con as , he de eloped wo ld, i any hing,
consumes an excessi e amoun o p o eins (Aiking and
de Boe , 2020). Fo ins ance, while he Lance Commis-
sion on heal hy die s sugges s ha an “adequa e p o ein
in ake o adul s is 0.8 g/kg bodyweigh , which is 56 g/
day o a 70-kg indi idual” (Wille e al., 2019) and he
Eu opean Food Sa e y Au ho i y (EFSA) se s an a e -
age equi emen in ake o 46 g p o ein pe capi a pe day
(Agos ini e al., 2012), p o ein in ake in he EU is a ound
82 g pe day, o which 49 g om animal-based sou ces
and 33 g om plan -based sou ces (Simon e al., 2024).
This aspec highligh s subs an ial inequali ies o he
ood sys ems be ween de eloping and de eloped wo ld,
and also ep esen s an inc eased isk o human heal h.
Mea consump ion con ibu es o global obesi y (You and
Hennebe g, 2016), highe isks o ype 2 diabe es (Malik
e al., 2016), kidney s ones (Asoudeh e al., 2022), ca dio-
ascula disease mo ali y (Zheng e al., 2022), cance
mo ali y (Huang e al., 2021) in he speci ic, colo ec al,
b eas and p os a e cance (Cellu a e al., 2022; Gonza-
lez e al., 2020) and mo e gene ally all-cause mo ali y
(Sun e al., 2021). Con e sely, die s ich in plan -based
p o eins, such as legumes, nu s and seeds, while su icien
o achie e ull p o ein adequacy in he de eloped wo ld
(Ma io i and Ga dne , 2019), seem o con e p o ec ion
agains he incidence o cance s (Gonzalez e al., 2020)
and o educe global mo ali y (Sp ingmann e al., 2016).
Inc easing he sha e o plan -based p o eins will p o ide
signi ican heal h and en i onmen al co-bene i s (Bon-
ne e al., 2020; S ylianou e al., 2021). This s udy aims
o assess he ela ionships be ween income and di e en
p o ein sou ces o highligh po en ial di e ences ha can
be use ul o unde s and he impac o policies. The pape
will discuss he ela ionship be ween ood consump ion
and income using exis ing li e a u e, which, howe e ,
a ely ook in o conside a ion p o ein sou ces o he han
mea , and explains he heo y behind he model in Sec-
ion 3. We will hen ou line he da a and he econome ic
s a egy we chose o apply o desc ibe his ela ionship in
Sec ion 4. The esul s o he es ima ed models a e p e-
sen ed in Sec ion 5 and hei implica ions a e discussed
in Sec ion 6 and 7.
3. THE RELATIONSHIP BETWEEN ANIMAL-BASED
PROTEIN CONSUMPTION AND INCOME
Rising eal G oss Domes ic P oduc (GDP a con-
s an p ices) o e he las cen u y has been iden i ied as
he oo -cause o a global nu i ion ansi ion. The an-
si ion encompasses a shi owa ds animal-based sou ced
p o eins in gene al (Ge bens-Leenes e al., 2010; Sans
and Comb is, 2015) and p o eins om mea in pa icu-
la (Mil o d e al., 2019; Yo k and Gossa d, 2004). As
income inc eases, consume s end o shi hei die a y
p e e ences owa d mo e esou ce-demanding oods
(Tilman and Cla k, 2014). This ansi ion is aking
place a di e en s ages and paces wo ldwide (Ge bens-
Leenes e al., 2010). Consump ion o animal-based oods
is much highe in de eloped coun ies han in de elop-
ing and leas de eloped coun ies. Howe e , he upwa d
end is mo e p onounced in de eloping coun ies (Hen-
chion and Zimme mann, 2021), whe e he ise abo e he
po e y line occu s a a as e pace han i did in de el-
oped coun ies (Sans and Comb is, 2015; D ewnowski
and Poulain, 2018). Meanwhile, in highe income coun-
ies a “second nu i ion ansi ion” seems o occu (Pais
e al., 2021; V anken e al., 2014). In hese coun ies,
he consump ion o animal-based p o eins, especially
om mea , seems o s agna e o decline when eaching a
high le el o income. V anken e al. (2014) and Cole and
McCoskey (2017) ha e he e o e ound e idence o an
in e ed U-shape ela ionship be ween mea consump-
ion and income, indica ing ha he consump ion o
unsus ainable p o eins could each a maximum and hen
decline. The e o e, hese s udies sugges ed ha mea
p o ein consump ion ollows an EKC. A guably, he ea-
sons o an AFKC di e om hose o he EKC. The la -
e is jus i ied by he inc easing en i onmen al impac o
he shi om an ag icul u al o an indus ial economy,
ollowed by a dec easing impac due o esou ces-sa ing
echnological p og ess and inc easing en i onmen al
awa eness. In he case o he AFKC, he same easons do
no apply, and he de e minan s ha e o be asc ibed o
he ac o s men ioned abo e.
The easons behind his decline can be a ibu ed o
se e al ac o s: i) inc easing awa eness o he heal h isks
associa ed wi h high mea consump ion, ii) conce ns
abou he en i onmen al impac o mea p oduc ion,
including g eenhouse gas emissions, de o es a ion, and
wa e usage, iii) g owing awa eness o animal wel a e, i )
he ising a ailabili y and popula i y o plan -based mea
al e na i es, ) he sp ead o popula die a y ends, such
as ege a ianism, eganism, and lexi a ianism.
We he e o e p esen a model o animal-based ood
consump ion ha inco po a es he abo e easons o
an in e ed U-shaped consump ion-income pa e n o
animal-based ood (AF) consump ion. The heo e ical
model sheds ligh on pas ends in AF consump ion and
he easons ha ende possible an AFKC. Ne e heless,
while he model may jus i y he exis ence o an AFKC, i
does no p edic i unequi ocally.
22
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
Vi o F on u o e al.
The model ( o a o mal p esen a ion see Appendix
1) assumes ha u ili y om AF consump ion has wo
componen s. The i s one di ec ly s ems om i s con-
sump ion pe se, due o i s as e and appe i e alue. U il-
i y is he e o e a posi i e unc ion o animal-based ood
consump ion, so ha i s ma ginal u ili y is posi i e, bu
dec easing, due o inc easing sa ia ion: addi ional AF
consump ion p o ides less and less addi ional u ili y.
The second componen is he nu i ional and heal h one.
Acco ding o he nu i ional li e a u e, consump ion o
animal-based p o eins has ini ially a posi i e e ec on
nu i ion and heal h ( eceding om amine, mo ali y
declines, see e.g., Ma hijs, 2015) bu , a highe le els, i
b ings se e al ad e se heal h e ec s (e.g., ca dio ascu-
la isks, obesi y- ela ed issues). Hence, i consume s a e
awa e o and ca e abou he nega i e impac s o high
animal-based ood consump ion on heal h, his compo-
nen o u ili y has an in e ed U-shape. In addi ion, as
men ioned abo e, conce n o animal wel a e and o he
en i onmen can be easons o a lowe u ili y associ-
a ed wi h la ge animal-based ood consump ion (F ank,
2008). In his model, o simplici y we include hese
e ec s in he heal h one.
The model assumes ha a consume maximizes his/
he u ili y subjec o a budge cons ain . The equilib ium
condi ion s a es ha he ma ginal u ili y om AF con-
sump ion pe se, plus he ma ginal u ili y s emming om
he a ia ion in nu i ion-heal h due o he e ec on con-
sume s’ heal h o an addi ional AF consump ion, equals
he addi ional u ili y ha could be d awn om o he
goods ha could be pu chased wi h he animal-based
ood p ice, i.e., he ma ginal oppo uni y cos o AF.
The ma ginal u ili y o AF consump ion pe se
dec eases when AF inc eases, and eaches a lowe bound
a ze o o sa ia ion, when u he consump ion p o ides
no addi ional u ili y. The ma ginal u ili y om nu i-
ional-heal h bene i s also dec eases wi h AF consump-
ion and emains posi i e as long as he ma ginal heal h
bene i is posi i e, hen i becomes nega i e. When he
ma ginal heal h bene i s, a high consump ion le els,
become nega i e, hey may de e mine a dec ease in o e -
all u ili y i disu ili y om heal h damages p e ails o e
u ili y due o as e. In his case, an in e ed U-shape o
he income-consump ion ela ionship esul s.
The model implies ha a dec ease in he AF p ice
ela i e o all o he p ices (i.e., a dec ease in eal AF
p ice) leads o highe AF consump ion. This explains
wha ac ually happened in he pas (FAO, 2009) when
he ela i e p ice o AF declined wi h e e ence o o he
ood p ices.
The c ucial ques ion o he exis ence o an AFKC
is ne e heless he shape o he ela ionship be ween
income and AF consump ion. Among necessi ies, ani-
mal-based ood is mo e expensi e han plan -based ood.
A low-income le els, a highe income allows a shi om
cheap s aple ood o animal-based ood, as empi ically
obse ed in all coun ies in he ini ial s ages o de elop-
men and as a gene al end in he ecen decades (Sans
and Comb is, 2015; Delgado e al., 2009; among o he s).
Howe e , he model canno unambiguously p edic
a p io i whe he a u he income g ow h leads o an
inc ease o dec ease o AF consump ion, because he
esul ing equilib ium will depend on how he ma ginal
u ili ies o AF o o he consump ions and o nu i ion-
heal h eac o income, and on hei in e ela ionships.
The model allows o he exis ence o an AFKC, bu
does no imply i s necessi y. The o m o he income-
AF consump ion ela ionship has he e o e o be de e -
mined empi ically.
Plan -based p o ein consump ion also inc eases
wi h income a he ini ial s ages o de elopmen . How-
e e , i s inc ease is p esumably slowe han he one o
animal-based p o eins, since income g ow h allows con-
sump ion o he mo e expensi e animal-based p o eins,
so ha in he die he sha e o animal-based p o eins
g ows. I consump ion o animal-based p o eins declines
a high income le els, i is possible ha plan -based p o-
ein consump ion will inc ease as a subs i u e. The ela-
ionship be ween plan -based p o ein consump ion and
income mus also be de e mined empi ically.
4. MATERIALS AND METHODS
4.1. Va iables and da a
We employ a balanced panel da ase co e ing 79
coun ies om 1991 o 2018 (Table A.1 in Appendix 2).
We d aw on da a om he Food and Ag icul u e O gan-
iza ion (FAO) New Balance Shee s (NBSs; FAO, 2021),
whe e ood supply quan i ies a e used as p oxies o
consump ion (Cole and McCoskey, 2017; You and Hen-
nebe g, 2016). These quan i ies a e measu ed in g ams
pe capi a pe day and e lec ood eaching consume s,
wi h he ca ea ha ac ual consump ion may be lowe
due o was e and spoilage du ing p epa a ion. The s udy
classi ies p o ein consump ion in o h ee ypes: “mea
p o ein” om poul y, po k, goa , mu on and bo ine;
“animal-based p o ein” encompassing all animal p od-
uc s including dai y and eggs; and “plan -based p o-
ein” de i ed om ce eals, ege ables, ui s, beans, nu s,
seeds, oo s and spices.
We explo e po en ial de e minan s o p o ein con-
sump ion ac oss h ee p incipal dimensions: economic,
socio-cul u al, and land use. In he economic dimen-
23
Is he e an Animal Food Kuzne s Cu e, and does i ma e ?
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
sion, he p ima y ocus is income exp essed by GDP
pe capi a (p.c.) a chained Pu chasing powe pa i ies
(PPPs) measu ed in million cons an 2017 US$. The
da a a e collec ed om he Penn Wo ld Table (Feens a
e al., 2015), a se o na ional-accoun s da a o measu e
eal GDP ac oss coun ies and o e ime. In he p esence
o an in e ed U-shape, i.e. Kuzne s cu e, we expec
posi i e es ima ed coe icien s o he linea e ms and
nega i e coe icien o he quad a ic e ms. In addi-
ion o income p.c., we ecognize he subs an ial in lu-
ence o ood p ices on p o ein consump ion pa e ns. To
cap u e his in luence, we build na ional p ice indexes
using da a om FAOSTAT (FAO, 2022b). Speci ically,
we selec he p ice o he mos consumed i em wi hin
each o he h ee p o ein sou ces (mea , animal-based
and plan -based) o e e y coun y and yea unde s udy
and build an index using he i s yea o he ime se ies
(1991) as base yea . This app oach aims o quan i y how
a ia ions in ood p ices ac oss di e en p o ein sou c-
es impac die a y choices and consump ion beha io s
globally. Indeed, ou hypo hesis is o obse e a nega i e
coe icien o he p ice index meaning ha an inc ease
in p ices de e mines a educ ion in p o ein consump-
ion. In addi ion o own p ice o each p o ein sou ce,
we ha e also es ed ela i e p ices. In ac , as sugges ed
by FAO (2009) o e he las 50 yea s he e has been a
decline in he p ices o li es ock p oduc s ela i e o
hose o o he p oduc s, making consump ion o animal-
based and plan -based oods mo e a o dable han mea
e en wi hou ising income. A hi d economic a iable
used in ou empi ical applica ion is he ade openness,
buil as he a io o impo s and expo s o e na ional
GDP. Ou hypo hesis is o obse e a posi i e e ec o
ade on he h ee p o eins consump ion due o he like-
ly la ge a ailabili y o di e en p oduc s and hus p o-
ein sou ces.
Beyond economic ac o s, social and cul u al in lu-
ences could also shape p o ein consump ion pa e ns.
We in eg a e se e al key a iables o explo e hese
dimensions. Fi s , he eligious belie s we e inco po a ed
by using he pe cen age o popula ion adhe ing o Islam
as a p oxy o unde s and die a y es ic ions ha may
a ec consump ion p e e ences, o example by educing
mea consump ion and inc easing plan -based p o ein
in akes. Second, we in eg a e he pe cen age o women
pa icipa ing in he labou o ce as an indica o o e ol -
ing ood p epa a ion p ac ices. Thi d, he pe cen age o
adul s wi h e ia y educa ion le els is used o cap u e
he in luence o educa ional a ainmen on die a y p e -
e ences and awa eness o nu i ional choices, po en ially
a ec ing p o ein in ake pa e ns. We hypo hesize ha
mo e educa ed people end o p e e die s wi h mo e
plan -based ood o bo h heal h and en i onmen al con-
ce ns. Howe e , we a e awa e ha educa ion is s ongly
co ela ed wi h income le els.
Finally, o u he explo e o he con ex ual condi-
ions likely in luencing p o ein consump ion, he s udy
includes wo p oxies o land use: he ha es ed a ea pe
capi a as a measu e o he ele ance o he ag icul u al
sec o o sel -p o ision o p o eins and he pe cen age o
he popula ion li ing in u ban a eas. These a iables a e
used o examine he impac o u baniza ion on die a y
habi s and access o di e se ood op ions, including p o-
ein sou ces.
Table 1 p o ides a comp ehensi e lis o a iables
used in he s udy, and hei desc ip i e s a is ics and
sou ces. Unlike he ypical p ac ice in En i onmen al
Kuzne s Cu e (EKC) li e a u e, he s udy uses a iables
in hei o iginal le els ins ead o loga i hmic ans o ma-
ions, aligning wi h indings by Hasano e al. (2021).1
4.2. Econome ic s a egy
Since he mid-1950s, schola s es ing he Kuzne s
Cu e (KC) hypo hesis on a ious en i onmen al and
non-en i onmen al indica o s ha e p ima ily used c oss-
sec ional da a and longi udinal da a wi h ixed e ec s
es ima o s (e.g., V anken e al. (2014) o consump ion o
mea p o ein). Howe e , adi ional panel da a es ima o s
assume c oss-sec ional independence, basing he models
on homogeneous coe icien s and yielding inconsis en
es ima ed pa ame e s (Heck and Thomas, 2020).
Indeed, c oss-sec ional uni s may exhibi sha ed
cha ac e is ics, such as spa ial e ec s, omi ed common
ac o s, o socioeconomic ne wo ks in e ac ion leading
o c oss-sec ional dependence, calling o es ima o s ha
accoun o in e cep s and slopes he e ogenei y. The li -
e a u e on he e ogeneous panels has e ol ed along wo
main s ands: i) he applica ion o mean g oup (MG)
es ima o s (Pesa an and Smi h, 1995) and subsequen
modi ica ions (Augmen ed MG and Common Co ela ed
E ec s MG; Teal and Ebe ha d , 2010), ii) he applica-
ion o mul ile el o mixed e ec models o panel da a
(McCulloch e al., 2001).
The key dis inc ion be ween panel models (such
as MG es ima o s) and mixed e ec s models lies in he
ea men o he independen a iables. In mixed e ec s
1 Hasano (2021) a gues ha in non-linea loga i hmic En i onmen al
Kuzne s Cu es (EKC), he signs o es ima ed coe icien s and he s a-
is ical signi icance o lowe -o de polynomial e ms can a y a bi a ily
based on he uni s o measu emen chosen o he independen a ia-
bles. Consequen ly, Hasano sugges s ha esea che s should i s s udy
he EKC in le els conside ing he po en ial issues wi h he loga i hmic
speci ica ion.
24
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
Vi o F on u o e al.
models, independen a iables a e ea ed as non- an-
dom a iables, whe eas in panel da a models, hey a e
always assumed o be andom. Ano he signi ican di -
e ence is in es ima ing he a e age e ec s (in a ian
be ween indi iduals) and indi idual (o andom) e ec s.
In he case o MG es ima o s, indi idual-speci ic o di-
na y leas -squa es (OLS) eg essions a e es ima ed hen
he indi idual-speci ic pa ame e s a e a e aged ac oss
he panel o de e mine an o e all e ec . In he case o
mixed e ec models, he es ima ed pa ame e s a e he
common e ec wi h he andom e ec s ep esen ing
indi idual de ia ions om his a e age, in e ed om
es ima ed a iances and co a iances (Dinda, 2004).
A me a-analysis conduc ed by Saqib and Benhmad
(2021) on mo e han i e hund ed s udies concluded ha
he econome ic s a egy does no signi ican ly impac
he es o he EKC hypo hesis. Howe e , hey high-
ligh ed he g ea e eliabili y o longi udinal da a and
he obus ness o me hods ha deal wi h he e ogeneous
panels such as MG es ima o s and mixed e ec models.
In his pape , we employ a mixed e ec model
because o ou ocus on he a ia ion in eg ession coe -
icien s a he han a global beha iou as an a e age o
coun y-speci ic dynamics. Coun y-speci ic es ima es,
limi ed by he income anges, canno p ope ly iden-
i y he cu a u e o a gene al unc ion. To accoun o
in e cep and slopes he e ogenei y in pa ame e s he
unknown pa ame e s a e decomposed in a ixed e m γ
(cons an ac oss coun ies) and a andom e m δ (speci -
ic o each coun y). Thus, he ela ionship be ween p o-
ein consump ion pe capi a (animal-, plan -based and
mea ) and GDP pe capi a is modelled as:
(P o eins/P)i = (γs0 + δsi0) + (γs1 + δsi1) (GDP/P)si +
(γs2 + δsi2) (GDP/P)2si + ∑Jj=3 βsijXsi j + εi (1)
whe e s= a,m,p iden i ies he p o ein sou ce (animal-,
plan -based and mea ), i=1,…,N indica es he coun ies,
=1,…,T he ime pe iods, GDP is de ined as abo e and
P is popula ion, Xj he j-co a ia es. No e ha ep-
esen s he po en ial non-linea e ec o GDP pe capi a
on p o eins consump ion and i is used in he Kuzne s
amewo k o check he in e ed U-shaped cu a u e o
he ela ion.
This model has been es ima ed using maximum
likelihood es ima o s o he h ee sou ces o p o ein
(Rabe-Heske h and Sk ondal, 2008) and likelihood- a io
es s ha e been employed o compa e di e en models
and o alida e he use o andom coe icien s. Mo eo-
e , he models a e i s es ima ed wi h an uns uc u ed
andom-e ec s co a iance ma ix, which allows o dis-
inc a iances and co a iances be ween all andom-
Table 1. Lis o a iables wi h desc ip i e s a is ics.
Va iable Desc ip ion Mean S d. Min Max Sou ce
Dependen a iables
Mea P o Pe capi a Mea -based P o ein consump ion (g/day) 16.4 11.1 1.2 46.9 FAO (2021)
AnimalP o Pe capi a Animal-based P o ein consump ion (g/day) 35.2 21.4 3.2 79.7 FAO (2021)
Plan P o Pe capi a Plan -based P o ein consump ion (g/day) 44.0 10.1 22.9 82.7 FAO (2021)
Independen a iables
GDPPc Pe capi a expendi u e-side eal GDP a chained PPPs (000 US$) 16.6 16.3 0.4 90.3
Penn wo ld able
(Feens a e al., 2015)
GDPPc2Squa ed GDPPc 541.7 880.9 0.2 8154.2
Penn wo ld able
(Feens a e al., 2015)
P ice Index Animal-based p oduc s 0.99 0.16 -0.88 2.65
FAO (2022b)
Mea p oduc s 0.98 0.17 -0.45 2.50
Plan -based p oduc s 1.11 0.52 -2.73 4.67
T ade (Impo s+expo s) / GDP (%) 68.7 34 13.8 227.4 Wo ld Bank (2022a)
Educa ion Sha e o pos -seconda y educa ion (%) 10.2 8.9 0.15 48.3 Wo ld Bank (2022b)
Pe Mus Sha e o Muslims o e popula ion (%) 23.4 35.7 0 99.8 ARDA (2022)
Pe FemWo k Sha e o emale employmen (%) 40.2 9.3 10.7 56 Wo ld Bank (2022c)
U baniza ion Sha e o people li ing in u ban a eas (%) 58.6 22.2 5.5 95.3 Wo ld Bank (2022d)
Ha A ea Ha es ed a ea/popula ion (pe capi a ha) 0.172 0.2 0 1.4 FAO (2022a)
N. obs. 2212
N. g oups 79
Sou ces: FAO, Penn Wo ld Table, Wo ld Bank, ARDA and own calcula ion.
25
Is he e an Animal Food Kuzne s Cu e, and does i ma e ?
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
e ec s co a ia es. Howe e , inconsis en es ima ions o
he plan -based p o ein model necessi a ed an iden i y
co a iance s uc u e, assuming equal a iances.
Acco ding o he li e a u e on es ing he na u e o
he ime-se ies o selec he app op ia e panel es ima o
(Pe man and S e n, 2003; Ebe ha d , 2012), he model
o equa ion [1] was es ed ela i e o: i) c oss-sec ional
dependence; ii) p esence o uni oo s (i.e., s a iona i y);
iii) long- un ela ionship (i.e., coin eg a ion).
To selec he app op ia e es o in es iga ing uni
oo s, we ini ially checked he c oss-sec ional depend-
ence o he se ies using he Pesa an es (Pesa an, 2021)
unde he null hypo hesis o c oss-sec ional independ-
ence. Mos a iables exhibi ed c oss-sec ional depend-
ence (excep o ade) (see Table A.2 in Appendix 3).
Subsequen ly, we es ed he s a iona i y o he se ies by
implemen ing he modi ied pCADF es (Cos an ini
and Lupi, 2013) which conside c oss-sec ional depend-
ence unde he null hypo hesis o non-s a iona i y. The
esul s sugges ed ha he null hypo hesis o non-s a ion-
a i y can be ejec ed only a e ans o ming he se ies
in hei i s di e ences excep o he u baniza ion a e
and he educa ion (Table A.3 in Appendix 3). Then, we
checked he coin eg a ion assump ion o p e en he
eg ession om p o iding biased s a is ical e idence
o he ela ionship among a iables. Coin eg a ion was
in es iga ed h ough a ious es s, including he Phil-
lips-Pe on, he Modi ied Phillips-Pe on, he Augmen -
ed Dickey-Fulle es s (Ped oni, 1999; Ped oni, 2004)
and he so-called Wes e lund es (Wes e lund, 2005)
by assuming he p esence o c oss-sec ional dependenc-
es (Table A.4 in Appendix 3). The ejec ion o he null
hypo hesis o all hese es s indica es ha ou models a e
coin eg a ed. The indings suppo he selec ion o he
mixed e ec model as app op ia e o es ima e he e og-
enous coe icien s o in e cep s and slopes.
5. RESULTS
Table 2 p esen s he es ima ed coe icien s o he
model o mea animal-based and plan -based p o ein
consump ion which exhibi o e all signi icance. Likeli-
hood- a io es s ha e been applied o compa e di e en
models and di e en co a ia es. Educa ion and u bani-
za ion a e u ned ou o be non-s a iona y e en when
ans o med in hei i s di e ences and we e he e o e
no used o a oid spu ious es ima ed coe icien s (see
Table A.3 in he Appendix 3). P ices o he h ee p o ein
sou ces a e no s a is ically signi ican and hence a e no
included in ou p e e ed speci ica ion in Table 2 (see
Table A.5 o he es ima ed coe icien s o model includ-
ing p ices).2 Table 2 also epo s he es ima ed s anda d
de ia ions o he in e cep , he GDP pe capi a and he
GDP2 pe capi a coe icien s. All o hese s anda d de ia-
ions a e s a is ically signi ican , indica ing he in e cep
and slopes he e ogenei y and hus suppo ing he use o
he mixed e ec s model.
The mos impo an de e minan o mea p o ein
consump ion is pe capi a income, wi h bo h i s es i-
ma es o he linea and he quad a ic e m highly sig-
ni ican . The es ima es indica e ha a housand dol-
la inc ease in pe capi a income induces a 0.725 g/
day inc ease in mea p o ein consump ion. No ably, he
nega i e sign o he squa ed e m sugges s ha mea p o-
ein consump ion does inc ease wi h income, bu a a
dec easing pace.
Among he a iables aside om income, he T ade
and he pe cen age o Muslims a e signi ican . Speci i-
cally, e e y addi ional pe cen age poin in he a io o
impo s plus expo s o e GDP implies a 0.008 g/day
inc ease in he a e age mea p o ein consump ion. A pe -
cen age poin inc ease in he sha e o Muslims o e he
popula ion ansla es in o a 0.11 g/day dec ease in he
a e age mea p o ein consump ion, ce e is pa ibus. These
esul s a e consis en wi h And eoli e al. (2021) and Mil-
o d e al. (2019). Female pa icipa ion, howe e , does no
show s a is ical signi icance, as in Mil o d e al. (2014).
The posi i e sign o he GDP pa ame e and he
nega i e sign o he GDP2 pa ame e , bo h signi ican ,
sugges he exis ence o an in e ed U-shaped ela ion-
ship, hus suppo ing he exis ence o an AFKC whe e
mea p o ein consump ion inc eases wi h pe capi a
income up o a maximum be o e dec easing. A c ucial
poin o assessing he policy implica ions o he AFKC
is ne e heless de e mining he le el o he u ning
poin . This can be calcula ed as whe e
is he es ima ed pa ame e o pe capi a income and
he es ima ed pa ame e o i s squa e.3 This simple
calcula ion esul s in a u ning poin o 42,923 US$,4
loca ed be ween he 90 h and he 95 h pe cen iles o
he pe capi a income dis ibu ion in he whole sample,
and abo e he 80 h pe cen ile o he income dis ibu ion
in he las yea o he panel (2018). I could be a gued
ha he u ning poin should also be es ima ed consid-
2 We used e sion 18 o STATA o Windows o ca y ou he analysis o
he da a in his pape . The mixed command has been used o es ima e
he mixed models p esen ed in Table 2.
3 The o mula o he maximum income in he es ima ed second-deg ee
equa ion is ob ained by se ing he de i a i e o he equa ion o ze o
and sol ing o he income a iable.
4 To p esen a mo e concise able o esul s, he coe icien s ha e been
ounded o h ee decimal places. Consequen ly, he u ning poin alue
de i ed om ounded coe icien s di e s om he one p esen ed in he
ex , which uses es ima ed coe icien s o six decimal places.
26
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
Vi o F on u o e al.
e ing he a ia ion o he es ima ed pa ame e s. Un o -
una ely, he u ning poin esul s om he a io o wo
no mal andom a iables, which esul s in a Cauchy dis-
ibu ion, whose expec ed alue and a iance a e unde-
ined. Howe e , i s mode and median a e de ined, and
he dis ibu ion is symme ical. We he e o e pe o m
a Mon e Ca lo simula ion o he median u ning poin .
We andomly d aw couples o and pa ame e s
om a bi a ia e no mal dis ibu ion, calcula e he u n-
ing poin , epea ed o 1000 d aws, and indi idua e he
median u ning poin o hese simula ions. By epea ing
he p ocedu e 10,000 imes we ob ain an empi ical dis-
ibu ion o he medians, om which we calcula e hei
mean and s anda d de ia ion. The esul o 42,891 US$
is sensibly simila o he simple calcula ion om he
es ima ed pa ame e s. The s anda d de ia ion is ela-
i ely modes , 318 US$, and he ange wen om a mini-
mum o 41,867 o a maximum o 44,159 US$. The mini-
mum alue is a ound he 90 h pe cen ile.
Compa isons wi h p e ious s udies e eal simila
u ning poin es ima es, i.e. 46,000-66,000 cons an 2017
In e na ional US$ p.c. (And eoli e al., 2021); 36,375-
49,848 cons an 2005 US$ p.c. (Cole e al., 2013); 35,000-
53,000 cons an 2005 in e na ional US$ p.c. (V anken e
al., 2014), indica ing consis ency ac oss analyses. How-
e e , employing mixed e ec s models alongside Mon e
Ca lo simula ion p oduces mo e e icien es ima es wi h
educed a iabili y.
Fu he mo e, he esul s allow us o p edic he
mea p o ein consump ion co esponding o he u n-
ing poin , by using he es ima ed pa ame e s and he pe
capi a income o he u ning poin and se ing he o he
a iables a hei mean. To app ecia e he a ia ion o
he p edic ion, we also calcula e he p edic ed consump-
ion when he o he a iables a e aken a he minimum
and maximum o hei obse ed alues,5 and when hey
a e aken a hei mean plus/minus hei s anda d de ia-
ion. Table 3 p esen s he esul s.
The calcula ed mea p o ein consump ion a he
income u ning poin and he mean o he o he a i-
ables is 24.61 g/day, sligh ly below he 75 h pe cen ile.
The maximum alue (29.57 g/day) is be ween he 85 h
5 When calcula ing he maximum and minimum consump ion, a i-
ables wi h a nega i e pa ame e we e aken as posi i e, so o iden i y he
maximum possible ange.
Table 2. Resul s o he models o p o ein consump ion.
Indep. a iables
Dependen a iables
Mea P o ein Animal-based P o ein Plan -based P o ein
Coe . S d.E . Coe . S d.E . Coe . S d.E .
GDPPc 0.725*** 0.135 1.255*** 0.206 1.506*** 0.419
GDPPc2-0.008*** 0.003 -0.015*** 0.004 -0.003 0.002
T ade 0.008*** 0.003 0.019*** 0.005 0.032*** 0.005
Ha A ea 1.097 1.317 1.930 1.863 -15.479*** 1.739
Pe Mus -0.105*** 0.0281 -0.177*** 0.484 0.212*** 0.421
Pe FemWo k -0.008 0.033 0.058 0.048 0.113** 0.046
Cons an 11.092*** 2.191 25.517*** 3.689 30.675*** 2.613
sd(GDPPc) 1.086 0.124 1.686 0.161 3.644 0.361
sd(GDPPc2) 0.021 0.006 0.030 0.004 0.005 0.002
sd(Cons an ) 13.364 1.414 25.170 2.378 12.892 1.120
sd(Residual) 1.541 0.028 2.167 0.035 2.093 0.034
N. obs. 2212 2212 2212
N. g oups 79 79 79
Wald Chisq(6) 78.33*** 108.10*** 194.50***
Log likelihood -4533.46 -5302.99 -5300.83
*, **, *** indica e ha s a is ics a e signi ican a he 10%, 5% and 1% le el o signi icance espec i ely.
Table 3. P edic ed p o ein consump ion a he u ning poin s.
Mea P o ein Animal-based P o ein
Mean 24.61 51.62
Min 16.16 35.02
Max 29.57 61.96
Mean-SD 20.45 43.79
Mean+SD 28.76 59.45
33
Is he e an Animal Food Kuzne s Cu e, and does i ma e ?
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
APPENDIX 1
In o mal e ms, he model o AF consump ion
assumes he consume maximizes his/he u ili y subjec
o a budge cons ain :
Max U[a, H(a), C] [A1]
s. .: C + pa = I
whe e a is animal-based p o ein consump ion, H indi-
ca es heal h-nu i ion componen s o u ili y as a unc-
ion o animal ood consump ion, C is expendi u e o
all o he consump ion goods, I is income, pa is he p ice
o a and he p ice o C is aken as nume ai e. The usual
gene al assump ions hold: U’a > 0, U’’a < 0; U’C > 0, U’’C
< 0; U’H > 0, U’’H < 0. To ep esen he U-shape o nu i-
ional-heal h bene i s, i is assumed ha H’a ≥ 0 o a ≤
a , H’a < 0 o a > a whe e a is he animal-based p o ein
consump ion yielding he maximum nu i ion-heal h
bene i ; H’’a is assumed < 0.
The i s o de condi ions (FOCs) is:
U’a + U’HH’a = paU’C [A2]
Equa ion [A2] simply s a es ha , a equilib ium, he
ma ginal u ili y om consump ion o AF ( he i s le -
side e m), plus he ma ginal u ili y om he nu i ion-
al-heal h bene i s om i s consump ion ( he second le -
side e m) is equal o he addi ional u ili y ha could be
d awn om o he goods ha could be pu chased wi h
he animal ood p ice, i.e., he ma ginal oppo uni y cos
o AF ( he igh -side e m).
The e ec o income on AF consump ion can be
compu ed as he de i a i e o a wi h espec o I in eqn.
[A2]. The esul is ne e heless a complex unc ion o
he second di ec and c oss de i a i es o a, H, C, and
i s sign canno be unambiguously de e mined, i can be
posi i e o nega i e.
APPENDIX 2
Table A.1. Lis o coun ies analysed.
Coun y ISO CODE
Alge ia DZA
A gen ina ARG
Aus alia AUS
Aus ia AUT
Bangladesh BGD
Boli ia BOL
B azil BRA
Coun y ISO CODE
Belize BLZ
Came oon CMR
Canada CAN
Cabo Ve de CPV
S i Lanka LKA
Chile CHL
China CHN
Colombia COL
Congo COG
Cyp us CYP
Denma k DNK
Dominican Republic DOM
Ecuado ECU
Egyp EGY
El Sal ado SLV
Finland FIN
F ance FRA
Gambia GMB
Ge many DEU
Ghana GHA
G eece GRC
Guinea GIN
Hondu as HND
Hunga y HUN
India IND
Indonesia IDN
I an (Islamic Republic o ) IRN
I eland IRL
Is ael ISR
I aly ITA
Cô e d’I oi e CIV
Japan JPN
Jo dan JOR
Kenya KEN
Cambodia KHM
Republic o Ko ea KOR
Lao People’s Democ a ic Republic LAO
Lebanon LBN
Madagasca MDG
Malaysia MYS
Mali MLI
Mau i ius MUS
Mexico MEX
Mo occo MAR
Mozambique MOZ
Namibia NAM
Nepal NPL
Ne he lands NLD
New Zealand NZL
Nica agua NIC
Nige NER
Nige ia NGA
34
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
Vi o F on u o e al.
Coun y ISO CODE
No way NOR
Pakis an PAK
Panama PAN
Pa aguay PRY
Pe u PER
Philippines PHL
Poland POL
Po ugal PRT
Rwanda RWA
Saudi A abia SAU
Sou h A ica ZAF
Spain ESP
Sweden SWE
Swi ze land CHE
Togo TGO
Tu key TUR
Uni ed Kingdom GBR
Uni ed S a es o Ame ica USA
Bu kina Faso BFA
U uguay URY
APPENDIX 3
Table A.2. Tes o c oss-sec ional dependence o a iables.
Va iable#Pesa an es
AnimalP o 2.678***
Mea P o 2.398**
Plan P o 2.798***
GDPPc 9.247***
GDPPc216.342***
T ade 0.158
Ha A ea 5.03***
Pe FemWo k 3.804***
U baniza ion 1.948*
Educa ion 85.837***
Animal-based P ice Index 12.883***
Mea P ice Index 6.536***
Plan -based P ice Index 24.181***
#The pe cen age o Muslim (Pe Mus) has no been es ed because
ime in a ian .
*, **, *** s and o he signi icance le el o 10%, 5% and 1% espec-
i ely. The null hypo hesis is he absence o c oss-sec ional depend-
ence.
Table A.3. Uni oo es on a iables and hei i s di e ence.
Va iable pCADF
es Va iable pCADF
es
AnimalP o 2.965 ∆AnimalP o -8.341***
Mea P o 1.408 ∆Mea P o -10.136***
Plan P o 3.616 ∆Plan P o -19.977***
GDPPc 5.426 ∆GDPPc -3.426***
GDPPc26.295 ∆GDPPc2-1.592*
T ade 4.052 ∆T ade -7.158**
Ha A ea 2.232 ∆Ha A ea -7.700***
Pe FemWo k 1.631 ∆Pe FemWo k -2.565***
U baniza ion 4.302 ∆U baniza ion 6.272
Educa ion 11.310 ∆Educa ion 3.016
Animal-based P ice
Index -12.225*** ∆Animal-based P ice
Index -20.497***
Mea P ice Index -12.825*** ∆Mea P ice Index -5.473***
Plan -based P ice
Index -16.698*** ∆Plan -based P ice
Index -9.583***
*, **, *** s and o he signi icance le el o 10%, 5% and 1% espec-
i ely. The null hypo hesis is non-s a iona i y.
Table A.4. Coin eg a ion es assuming c oss-sec ional dependence.
Tes name AnimalP o Mea P o Plan P o
Ped oni
Modi ied Phillips–Pe on 4.590*** 3.697*** 1.88**
Phillips–Pe on -7.509*** -8.279*** -9.710***
Augmen ed Dickey–Fulle -8.99*** -10.195*** -10.324***
Wes e lund
Va iance a io -2.579 *** -2.751*** -1.6434*
*, **, *** indica e ha s a is ics a e signi ican a he 10%, 5% and
1% le el o signi icance espec i ely. The null hypo hesis is no-coin-
eg a ion.
35
Is he e an Animal Food Kuzne s Cu e, and does i ma e ?
Bio-based and Applied Economics 14(1): 19-35, 2025 | e-ISSN 2280-6172 | DOI: 10.36253/bae-16172
Table A.5. Resul s o he ull models o p o ein consump ion.
Indep. a iables
Dependen a iables
Animal-based P o ein Mea P o ein Plan -based P o ein
Coe . S d.E . Coe . S d.E . Coe . S d.E .
GDPPc 1.256*** 0.206 0.725*** 0.135 1.505*** 0.419
GDPPc2-0.015*** 0.004 -0.008** 0.003 -0.002 0.001
T ade 0.018*** 0.005 0.008** 0.003 0.032*** 0.004
Ha A ea 1.190 1.862 1.087 1.317 -15.486*** 1.738
Pe Mus -0.177*** 0.048 -0.105*** 0.028 0.211*** 0.421
Pe FemWo k 0.057 0.047 -0.008 0.033 0.111** 0.046
P ice Index (animal-based) 0.304 0.289
P ice Index (mea ) 0.065 0.200
P ice Index (plan -based) -0.184 0.148
Cons an 25.245*** 3.697 11.033*** 2.199 30.910*** 2.619***
sd(GDPPc) 1.686 0.161 1.087 0.125 3.644 0.361
sd(GDPPc2) 0.029 0.003 0.021 0.006 0.005 0.002
sd(Cons an ) 25.17 2.378 13.369 1.415 12.894 1.119
sd(Residual) 2.167 0.035 1.541 0.028 2.092 0.033
N. obs. 2212 2212 2212
N. g oups 79 79 79
Wald Chisq(6) 109.25*** 78.31*** 196.14***
Log likelihood -5302.43 -4533.41 -5300.07
*, **, *** indica e ha s a is ics a e signi ican a he 10%, 5% and 1% le el o signi icance espec i ely.
