Ga ay, U bi; Pulga, F edy
A icle
Ca ego izing wo ld egional a p ices by a is ic
mo emen : an analysis o La in Ame ican a
Jou nal o Economics, Finance and Adminis a i e Science
P o ided in Coope a ion wi h:
Uni e sidad ESAN, Lima
Sugges ed Ci a ion: Ga ay, U bi; Pulga, F edy (2025) : Ca ego izing wo ld egional a p ices
by a is ic mo emen : an analysis o La in Ame ican a , Jou nal o Economics, Finance and
Adminis a i e Science, ISSN 2218-0648, Eme ald Publishing Limi ed, Leeds, Vol. 30, Iss. 59, pp.
116-149,
h ps://doi.o g/10.1108/JEFAS-02-2024-0065
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/319674
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/
Ca ego izing wo ld egional a p ices
by a is ic mo emen : an analysis o
La in Ame ican a
U bi Ga ay
Depa men o Finance, IESA, Ca acas, Venezuela and
EICEA, Uni e sidad de La Sabana Campus del Puen e del Com�
un,
Chia, Colombia, and
F edy Pulga
EICEA, Uni e sidad de La Sabana Campus del Puen e del Com�
un,
Chia, Colombia
Abs ac
Pu pose –The li e a u e on he po en ial bene i s o a in es ing has ye o conside he e ec s o ca ego izing
wo ld egional a ma ke s (e.g. La in Ame ican a ) by a is ic s yles o mo emen s (e.g. La in Ame ican
su ealism, La in Ame ican concep ual a , e c.). We p opose ha such ca ego iza ion should be ca ied ou and
analyze he La in Ame ican a ma ke as an example.
Design/me hodology/app oach –Ele en a is ic s yle p ice indices wi hin he La in Ame ican a ma ke
(30,288 a wo ks c ea ed by 293 a is s and sold a auc ion be ween 1970 and 2014) a e es ima ed using hedonic
eg essions: Abs ac -geome ic, abs ac -in o mal, concep ual, cos umb ismo, cubism, igu a i e, mu alism,
landscape, su ealism, nine een h cen u y and a an -ga de. We ind ha se e al a iables ha ely on he
co esponding La in Ame ican a mo emen index ha e a signi ican ela ionship wi h pain ing p ices.
Findings –The e is signi ican a ia ion in he inancial pe o mance o he a ious p ice indices o La in
Ame ican a s yles: he concep ual (10.33% annual eal e u n), abs ac geome ic (1.97%), cubism (0.97%)
and cos umb ismo (0.91%) mo emen s o e pe o med a ma ke ha exhibi ed an agg ega e nega i e cumula ed
eal e u n o 0.9% du ing he sample pe iod. The a e age co ela ion be ween each o he s yles was only 0.12.
The es ima ed p ice index o pain ings sold a Ch is ie’s and So heby’s clea ly ou pe o med he index
es ima ed o he o he auc ion houses, and we also ound ha pain ings c ea ed by La in Ame ican women
a is s yielded highe e u ns.
P ac ical implica ions –Ou esul s ha e p ac ical applica ions o in es o s, collec o s, auc ion houses and
policymake s.
O iginali y/ alue –This is he i s pape o highligh he need o decompose a p ice indices by a is ic
mo emen s a he egional le el.
Keywo ds Hedonic p icing model, La in Ame ican a , S yle e ec s, Di e si ica ion, A p ice indices,
Women a is s
Pape ype Resea ch pape
JEFAS
30,59
116
JEL Classi ica ion — G10, G11, G12, G20, Z11
© U bi Ga ay and F edy Pulga. Published in Jou nal o Economics, Finance and Adminis a i e
Science. Published by Eme ald Publishing Limi ed. This a icle is published unde he C ea i e Commons
A ibu ion (CC BY 4.0) licence. Anyone may ep oduce, dis ibu e, ansla e and c ea e de i a i e wo ks
o his a icle ( o bo h comme cial and non-comme cial pu poses), subjec o ull a ibu ion o he
o iginal publica ion and au ho s. The ull e ms o his licence maybe seen a h p://c ea i ecommons.o g/
licences/by/4.0/legalcode
The au ho s would like o hank he edi o and he anonymous e e ees o hei e y insigh ul
commen s and sugges ions. We would also like o exp ess ou g a i ude o he pa icipan s a he Business
Associa ion o La in Ame ican S udies (BALAS) Annual Con e ence (Mexico Ci y, Mexico, June o
2023); and a he Semina io Anual de In es igaciones, held a IESA (Ca acas, Venezuela) in Sep embe o
2022, o hei cons uc i e sugges ions. The usual disclaime applies.
The cu en issue and ull ex a chi e o his jou nal is a ailable on Eme ald Insigh a :
h ps://www.eme ald.com/insigh /2077-1886.h m
Recei ed 12 Ap il 2024
Re ised 1 Oc obe 2024
24 No embe 2024
Accep ed 26 No embe 2024
Jou nal o Economics, Finance and
Adminis a i e Science
Vol. 30 No. 59, 2025
pp. 116-149
Eme ald Publishing Limi ed
e-ISSN: 2218-0648
p-ISSN: 2077-1886
DOI 10.1108/JEFAS-02-2024-0065
While a ma ke s may seem monoli hic he e a e mic o ends cons an ly occu ing wi hin each
collec ing ca ego y
ARTBnk
1. In oduc ion
O e he pas ew decades, bo h indi idual and ins i u ional in es o s as well as a collec o s
ha e become inc easingly in e es ed in he global a ma ke . The la e belie e ha he a
ma ke could o e be e po olio di e si ica ion. In 2023, 65 billion dolla s’ wo h o a was
aded, which was ma ginally mo e han he p e-pandemic le els (McAnd ew, 2024). In a
su ey on a collec ing, McAnd ew (2023) disco e ed ha , ou o he six easons gi en o
collec o s wo ldwide, sel -iden i y and sel -es eem we e anked highes (37%), ollowed by
inancial mo i a ions (28%), which we e also anked h oughou he wo ld. Addi ionally,
millennials made up he la ges pe cen age o collec o s who we e d i en p ima ily by a desi e
o passion o collec ing, whe eas boome s made up he la ges pe cen age o in es o s (32%)
who we e d i en p ima ily by he possibili y o inancial gains.
On a e age, p ices ep esen he wo h and quali y o a wo ks (Coslo and Spaenje s,
2016). E en hough his p oposi ion is accep ed by mos economis s, i is s ill con o e sial
among a his o ians and he public (Edwa ds, 2004). Pain ings and o he collec ibles
cons i u e a unique ype o asse since, o example, hey do no gene a e income o hei
owne s like s ocks, bonds and eal es a e do. In addi ion o he possible inancial gains om
p ice app ecia ion, collec o s also enjoy he aes he ic pleasu e ha comes wi h owning wo ks
o a .
Acco ding o Baumol (1986), a wo ks can also be ega ded as a consume good.
Acco ding o he empi ical e idence cu en ly a ailable, in es ing in a wo ks (di e si ied by
coun y and a is ic s yle) is gene ally expec ed o yield e u ns ha a e lowe han s ocks and
mo e compa able o bonds, bu wi h less isk han s ocks ( o a e iew o he li e a u e, see
Ga ay, 2018). The possibili y o ob aining supe io e u ns when pu chasing wo ks by a
pa icula a is o a is ic mo emen du ing a speci ic ime pe iod is no excluded by he
a o emen ioned; in ac , such high e u ns a e mo e likely o a ac media a en ion.
The exis ing li e a u e on he pe o mance o a in es ing a he egional le el (e.g. La in
Ame ica, Sou h-Eas Asia, A ica, Scandina ian a , e c.) has been, hus a , been pe o med a
an agg ega e le el, and such geog aphical classi ica ions has only been ca ego ized by
coun ies (na ionali ies), i any hing (e.g. Higgs and Wo hing on, 2005, o he case o
Aus alia; K aeussl and Loghe , 2010, o China, India and Russia; Renneboog and Spaenje s,
2014, o Aus alia, Aus ia, Belgium, Canada, Denma k, F ance, Ge many, I aly, he
Ne he lands, Sweden, Swi ze land, he UK and he USA; Shi e al., 2017, o China; Ga ay
e al., 2017, o A gen ina). Al hough a ew ecen pape s ha e b oken down coun y a
e u ns by mo emen s (see Ga ay, 2021, o Venezuela, Gu ja and Anan hakuma , 2023, o
India, and Wang, 2023, o China), o he bes o ou knowledge, no conside a ion has been
gi en o he impac ha a ious a is ic mo emen s o s yles (such as su ealism, concep ual
a , e c.) ha may exis wi hin wo ld egions (such as Scandina ian a , La in Ame ican a ,
e c.) may ha e on he pe o mance o such indices up o his poin [1]. Gi en he signi icance
ha collec o s and auc ion houses a ach o speci ic a ma ke s, his is somewha su p ising.
Fo example, 30 o he 60 exis ing specialis depa men s a Ch is ie’s a e egional o coun y
speci ic (e.g. Indian, Himalayan and Sou heas Asian A ; and Sou h Asian Mode n and
Con empo a y A ). In he case o So heby’s, he p opo ion is e en highe , as 40 ou o 73 o i s
depa men s a e egional o coun y speci ic (e.g. Ge man, Aus ian and Cen al Eu opean
Pain ings; and Mode n and Con empo a y Sou heas Asian A ).
In his pape , we ake an in-dep h analysis o he La in Ame ican a ma ke and p opose a
p ice indices o his speci ic egion, and classi y pain ings acco ding o he ollowing ele en
a is ic s yles: Abs ac (geome ic), abs ac (in o mal), concep ual, cos umb ismo, cubism,
igu a i e, mu alism, landscape, su ealism, nine een h cen u y and a an -ga de. Ou analysis
Jou nal o
Economics,
Finance and
Adminis a i e
Science
117
o La in Ame ican a is s’ auc ion sales om 1970 o 2014 yields 30,288 sales, c ea ed by 293
a is s. We look a a books and place each o hose a is s in one o he ele en p e iously
men ioned a is ic mo emen s o s yles. A e adjus ing o se e al con ol a iables ha a e
common in he li e a u e, we es ima e hedonic p icing eg essions o each s yle.
The pe o mance o he a ious a is ic s yle indices om La in Ame ica, he egion we
speci ically s udy, does, in ac , di e signi ican ly om one s yle o ano he . Mo e
speci ically, he cumula ed geome ic eal e u ns du ing he pe iod we e posi i e o only ou
o he a is ic s yles: concep ual (10.33%, 1996–2014), abs ac geome ic (1.97%,
1971–2014), cubism (0.97%, 1981–2014) and cos umb ismo (0.91%, 1971–2014). The
emaining se en s yles had a cumula i e annual eal e u n ha was nega i e. Fu he mo e,
and mo e impo an ly o ou pape , he a e age co ela ion be ween he s yles was e y low
(only 0.12) and in some cases e en nega i e.
Ou s udy highligh s he impo ance o b eaking down agg ega e a p ice indices a he
egional le el by s yles so ha a in es o s can coun wi h a be e isk/ e u n assessmen o he
di e en cons i uen s (s yles) o he a ma ke . This is compa able o wha is o en p oclaimed
o s ocks, whe e he pe o mance o di e en ca ego ies ( o example, alue e sus g ow h
s ocks, s ocks o di e en indus ies, e c.) a ies signi ican ly wi hin and ac oss egions and
coun ies. Addi ionally, we cons uc and e alua e La in Ame ican a p ice indices a he
na ional le el and e alua e he pe o mance o La in Ame ican a wo ks sold a Ch is ie’s and
So heby’s in compa ison o hose sold a o he auc ion houses. Las ly, we disco e ha , in he
case o La in Ame ican a , wo ks c ea ed by women a is s command highe p ices, which is
in con as o indings epo ed in o he ecen and ela ed li e a u e.
The pape is o ganized as ollows. Sec ion 2 o e s a li e a u e e iew on he po en ial
bene i s o a in es ing. Sec ion 3 discusses he speci ici y o he La in Ame ican a and i s
auc ion ma ke , which make i sui able o conduc a s udy such as ou s. Sec ion 4 is dedica ed
o he da a and me hodology used in his s udy, and Sec ion 5 p esen s he esul s ob ained,
which a e discussed in Sec ion 6.Sec ion 7 deals wi h a se ies o obus ness es s and
ex ensions o he main esul s. Finally, in Sec ion 8 we o e he conclusions, implica ions o
ou pape (including p ac ical applica ions o in es o s, collec o s, auc ion houses and
policymake s) and p opose po en ial a enues o u u e esea ch.
2. Li e a u e e iew on he inancial pe o mance o a
The hedonic p icing me hod and he epea -sales me hod a e he wo p ima y app oaches ha
ha e been pu o h in he li e a u e o es ima e a p ice indices. Only a wo ks ha ha e sold
a leas wice du ing a sample pe iod a e conside ed using he epea -sales me hod when
es ima ing e u ns. The epea -sales me hod has he bene i o calcula ing a p ice changes
based on pu chases o he same piece o a . Howe e , since only a small pe cen age o
auc ion sales a e usually epea sales, his me hod’s d awback is ha i only includes a e y
small po ion o all a wo ks sold. An addi ional d awback is ha a wo ks whose alues a e
hough o ha e inc eased mo e a e mo e likely o be o e ed a auc ion, hus c ea ing an
upwa d bias in a e u ns when his me hodology is used (see Goe zmann, 1993;Ko eweg
e al., 2016).
Using epea -sales da a o he yea s 1650–1960 om Rei lingle ’s (1961) book, Baumol
(1986) ound ha a p ices beha e andomly and luc ua e subs an ially. In he long un, eal
annual e u ns we e only 0.55%. A e upda ing he da a and applying he epea -sales me hod,
Goe zmann (1993) also used da a om he book by Rei lingle (1961) and concluded ha
weal h and equi y ma ke e u ns signi ican ly impac ed a ma ke p ices. In his analysis o he
1977–1992 p in ma ke , Pesando (1993) ound ha a wo k pe o med wo se han adi ional
inancial asse s. Finally, E o and S epano a (2019) ound, using he epea -sales me hod, ha
a ma ke s a e e icien , as e u n a es do no depend sys ema ically on pas p ices.
The hedonic p icing model (Rosen, 1974) es ima es he implici p ice (i.e. he hedonic
p ice) o con ibu ion o each o he a ibu es o a wo k o a (e.g. name o he a is , i s size,
JEFAS
30,59
118
e c.) wi hin he o al p ice. The e is a weal h o esea ch on he applica ion o he hedonic
p icing model o es ima e a p ice indices and analyze he ac o s ha in luence a p ices. Fo
example, Higgs and Wo hing on (2005),K aeussl and Loghe (2010),Renneboog and
Spaenje s (2012),Ga ay e al. (2017) and Ga ay (2021) ha e all used he hedonic p icing
model o analyze a ma ke s [2]. Hedonic p icing eg ession is by a he me hod ha is mos
commonly used in he cul u al economics li e a u e o es ima e a e u ns, acco ding o a s udy
conduc ed by Rade mecke and Al a ez de Toledo (2022).
Acco ding o Higgs and Wo hing on’ (2005) hedonic p icing model analysis o he
Aus alian a ma ke om 1973 o 2003, he nominal a e age annual e u n o Aus alian a
was 6.96%, wi h a s anda d de ia ion o 16.51%. The in es men bene i s o a wo ks c ea ed
by Aus alian Abo iginal a is s be ween 1982 and 2007 we e examined by Taylo and
Coleman (2011), who calcula ed an annual nominal e u n o 6.6% wi h a s anda d de ia ion o
17.9%. Renneboog and Spaenje s (2012) used da a om 1.1 million auc ion sales a ound he
wo ld (1957–2007), and applied a hedonic eg ession analysis, inding ha ha he ollowing
a ibu es a ec a p ices: size, signa u e, da e, echnique, au ho a ibu ion dummies, subjec
( opic), auc ion house and loca ion o he auc ion house. They also epo ed ha a exhibi ed
annual eal e u ns o 3.97% du ing hei sample pe iod. Renneboog and Spaenje s (2012), and
Li e al. (2022) also ound ha he co ela ion o a and s ocks and bonds was low (and was
e en nega i e wi h o he asse s).
Resea ch on a in es men pe o mance in eme ging ma ke s is s ill ela i ely ecen .
K €
aussl and Loghe (2010) s udied he a ma ke s o Russia (1985–2008), China
(1990–2008) and India (2002–2008). The au ho s calcula ed, using hedonic eg essions,
ha hese h ee eme ging ma ke s’ a e age nominal annual dolla e u ns we e,
espec i ely, 10%, 5.7% and 42.2% highe han he in la ion a es in he USA. Using he
epea -sales me hod, Shi e al. (2017) examined he in es men pe o mance o Chinese a
and calcula ed ha , on a e age, Chinese a is s’ wo ks p o ided an annual eal app ecia ion
o 8.42% be ween 2000 and 2015. Acco ding o a 2017 s udy by Ga ay, Vielma and
Villalobos, he a ma ke o A gen ine a is s p o ided an annual eal dolla e u n o
3.81% (1980–2014). Acco ding o Ga ay’s (2021) analysis o he Venezuelan a ma ke
om 1970 o 2014, geome ic e u ns we e ma ginally highe han in la ion in US dolla s,
and a wo ks yielded a nominal annual a e age e u n o 7.96%. In hese las wo s udies,
hedonic p icing eg essions we e used.
Addi ionally, a body o li e a u e has su aced ega ding he es ima ion o a p ice indices.
Fo example, Candela and Sco cu (1997) p oposed a p ice index me hodology ha is based on
es ima es and auc ion p ices and sugges ed a selec ion p ocedu e ha enables he c ea ion o
indexes o a speci ic a ma ke segmen and/o a high equencies. In u n, Candela e al.
(2004) de eloped an annual p ice index compu ed as he a io o he a e age ma ke p ice o
he a e age es ima ed p ice o pain ings sold by a is s each yea . The au ho s sugges adjus ing
he a e age es ima ed p ice a a speci ic momen based on his o ical p ice dynamics in o de o
es ima e a ios because hey belie e ha expe s a e likely o aise (dec ease) es ima ed p ices
i he ma ke is bullish (bea ish). They also concluded ha indices compu ed o a is s who a e
equen ly aded a e less e a ic.
3. Li e a u e e iew on he La in Ame ican a ma ke
“La in Ame ican a ” encompasses he collec i e a is ic exp ession o Sou h and Cen al
Ame ica, he Ca ibbean, La in Ame icans esiding ab oad and a is s who ha e mig a ed o
La in Ame ica, acco ding o Ba ni z (2006). Acco ding o U ibe (in The an, 1999), he e m
“La in Ame ican a ” also includes a wo k p oduced by o eigne s who ha e isi ed he
egion, such as Du ch a is s who we e among he i s o cap u e he exo ic lands o he New
Wo ld when hey a eled o wha is now B azil in he 1630s. A unique adi ion o shapes,
colo s and mo i es has been c ea ed h oughou he egion as a esul o he blending o
Na i e Ame ican, A ican and Eu opean cul u es. Fo T aba (1994), he mos signi ican
Jou nal o
Economics,
Finance and
Adminis a i e
Science
119
aspec o La in Ame ican a is s’ wo k is hei cons an a emp o engage wi h hei
cul u es. T aba’s p oposal is con o e sial because some schola s con end ha La in
Ame ican a should also be examined om he s andpoin o he global a scene, whe e
La in Ame ican a is ic mo emen s like geome ic abs ac ion and kine ic a ha e had a
signi ican impac (F os A Museum and The Pa icia & Phillip F os A Museum, 2010).
Ga ay (2018) ela es ha “acco ding o some au ho s such as P�
e ez-Ba ei o (see Cas o,
2013), when a eling o con inen al coun ies such as B azil, Colombia, Chile, U uguay o
Ca ibbean coun ies, o example, people end o iden i y wi h hei coun y and no wi h he
con inen . Tha is why he e m La in Ame ican a should be conside ed wi h cau ion, no
only because o he p imacy ha collec o s om La in Ame ican coun ies ha e
adi ionally gi en o c ea ed wo ks p oduced by a is s om hei own coun y ( a he
han o wo ks execu ed by o he a is s om he con inen ), bu also because La in
Ame ican a includes some cu en s o s yles, such as geome ic abs ac ion o kine icism,
which can be concei ed a he as insc ibed wi hin in e na ional abs ac -geome ic o
kine ic a ” ( he ansla ion o English is ou s).
The e a e h ee p ima y bene i s o s udying La in Ame ican a auc ions om an economic
pe spec i e (Edwa ds, 2004). Fi s , since 1979, he e ha e been equen in e na ional auc ions
de o ed o La in Ame ican a , wi h So heby’s and Ch is ie’s domina ing he ma ke
(pa icula ly o he mos expensi e lo s) as well as a o he smalle bu highly ac i e auc ion
houses in Eu ope and he USA. This spans mo e han 40 yea s [3]. Addi ionally, in ce ain
La in Ame ican na ions, local a auc ions ha e been egula ly held since a leas he 1950s. As
a esul , La in Ame ican a auc ions ha e a leng hy his o y, which is essen ial o conduc ing a
s udy like he one ha is being p esen ed he e. Second, al hough museum in e es in La in
Ame ican a has undoub edly g own since Edwa ds’ a icle, i is s ill ela i ely small
compa ed o Ame ican a is s and Imp essionis s, o ins ance. Fo he pu poses o ou s udy,
his is a posi i e ea u e o he ma ke because i sugges s ha he La in Ame ican a ma ke is
no skewed by he pu chasing decisions o big museums, which equen ly pu chase and
“ e i e” some o he bes pieces a ailable.
Thi d, a wo ks om mos La in Ame ican g ea mas e s ( he “Big-Fi e” La in Ame ican
a is s a e Ru ino Tamayo, Diego Ri e a, Robe o Ma a, Wil edo Lam and Fe nando Bo e o)
a e liquid. Fu he mo e, Ga ay (2018) a gues ha i may be wo hwhile o in es iga e he
ela ionship be ween inancial and economic ac o s and he p ices o a wo ks due o he
compa a i ely high economic ola ili y o many La in Ame ican na ions. In hei 2014 s udy
o he a ma ke s in Wes e n Eu ope and he USA, Renneboog and Spaenje s disco e ed ha
local ac o s, such as GDPs and s ock e u ns, ha e an impac on he p ices o pain ings c ea ed
by local a is s, hough no as much in he case o he high-end ma ke .
Edwa ds (2004) conduc ed one o he i s s udies on he pe o mance o La in Ame ican
a , analyzing 12,690 auc ion sales o La in Ame ican pain ings. He ound ha , be ween 1982
and 1990, he eal annual a es o e u n o La in Ame ican a we e consis en ly posi i e and
ha , in a ew yea s, hey we e ela i ely high. Fo he yea s 1991–2000, howe e , he a es o
e u n we e signi ican ly lowe and in some cases e en nega i e. The eal a e age annual a e
o e u n o he en i e po olio om 1981 o 2000 was 9%, wi h a s anda d de ia ion
o 12.6%.
Campos and Ba bosa (2009) analyzed pain ings by La in Ame ican a is s auc ioned a
So heby’s be ween 1995 and 2002 and ound ha a wo k p ices we e highe when, ce e is-
pa ibus: he a is was epu able, he a wo k was pain ed using oil and he la ge i s a ea.
Simila ly, he au ho s ound, con a y o wha would be expec ed, ha a signed wo k did no
e ch a highe p ice. Addi ionally, hey examined he e ec s o o he ac o s ha a e exclusi e
o So heby’s and a ew o he upscale ma ke ca alogs, like p o enance – whe he a pain ing has
been ea u ed in a books o has been shown in galle ies o museums. Table 1 p esen s a
summa y o he li e a u e on he isk- e u n o a in es ing by mo emen s/s yles and o La in
Ame ican a ( o bo h he egion as a whole and o speci ic coun ies).
JEFAS
30,59
120
Acco ding o Ga ay e al. (2017), he e is a home bias (local bias) in a pu chases because
in es o s and collec o s a e d awn o he c ea ions o a is s om hei own na ion, whe he
hey a e o e ed o sale a domes ic o in e na ional auc ions. The e o e, domes ic economic
condi ions should ha e an impac on he local a ma ke , suppo ing he c ea ion o a p ice
indices o local a is s and consis en wi h he logic p esen ed by Renneboog and Spaenje s
Table 1. Summa y o he li e a u e on he isk- e u n o a in es ing by mo emen s/s yles and o La in
Ame ican a
S yle(s)/mo emen (s)
Annual
e u ns
S anda d
de ia ion
Annual
e u ns
S anda d
de ia ion Annual e u ns
Renneboog and
Spaenje s (2012) ( eal
dolla s) 1957–2007 1957–2007 1982–2007 1982–2007
Li e al. (2022)
1958–2016
( eal dolla s)
Medie al and
enaissance
3.01% 27.13% 6.44% 19.59% 3.32%
Ba oque 4.76% 17.69% 5.82% 12.57% 3.42%
Rococo 3.69% 25.42% 5.03% 12.15% 4.40%
Neoclassicism 6.32% 45.93% 5.36% 22.45% 7.48%
Roman icism 4.28% 17.34% 4.79% 15.24% 3.35%
Realism 2.57% 21.42% 4.16% 15.46% 3.20%
Imp essionism and
symbolism
4.10% 24.01% 4.55% 16.70% 3.47%
Fau ism and
exp essionism
3.72% 22.84% 4.90% 18.36% 3.68%
Cubism, u u ism and
cons uc i ism
5.53% 22.40% 6.01% 20.55% 4.74%
Dada and su ealism 5.85% 32.32% 5.58% 19.42% 5.32%
Abs ac exp essionism – – 7.78% 21.91% 5.28%
Pop-a – – 10.35% 29.33% 7.93%
Minimalism and
con empo a y
– – 7.07% 23.68% 12.88%
Ko eweg e al. (2016) 1961–2013 1961–2013
(Nominal dolla s)
Pos -wa and con empo a y 7.43% 11.63%
Imp essionism and mode n 6.09% 13.30%
Old mas e s 4.56% 13.75%
US a is s 6.83% 10.28%
Eu opean XIX cen u y 6.81% 11.70%
O he s yles 6.53% 13.92%
Top 100 a is s 9.50% 13.86%
Edwa ds (2004) 1981–2000 1981–2000
La in Ame ica ( eal dolla s) 9.00% 12.60%
Campos and Ba bosa (2009) 1995–2002
La in Ame ica (nominal dolla s) 5.23% –
K €
aussl e al. (2016) 1970–2013
La in Ame ica (nominal dolla s) 6.11% –
Ga ay e al. (2017) 1980–2014 1980–2014
A gen ina (nominal dolla s) 6.81% 29.11%
Ga ay (2021) 1969–2014 1969–2014
Venezuela (nominal dolla s) 7.96% 33.66%
Sou ce(s): Upda ed om Cine a e al. (2019)
Jou nal o
Economics,
Finance and
Adminis a i e
Science
121
(2014). P e iously, Goe zmann e al. (2011) asse ed ha B i ish in es o s and collec o s had
pu chased mos o he a wo k sold a auc ion in G ea B i ain. In he pape ’s discussion
sec ion, we o e addi ional analysis on he possible ami ica ions o he likely p esence o a
home bias in he a ma ke .
3.1 On he issue o a is s’ ca ee e olu ion o e ime
A is s’ ca ee s change o e ime. The a gumen pu o h by Galenson (2000) and Galenson
and Weinbe g (2000,2001) is ha concep ual a is s – hose who plan hei ideas ahead o ime
and execu e hem me hodically – achie e he highes p ices o hei wo ks and each he
pinnacles o hei ca ee s be o e expe imen al pain e s, who use inc emen al echniques and
aim “ o pe ec ion in hei wo ks.” In his con ex , Ga ay e al. (2022b) examined he Big-5
La in Ame ican a is s and ound ha , in he cases o Wil edo Lam, Robe o Ma a and Diego
Ri e a, he p ices o hei wo ks we e nega i ely and signi ican ly co ela ed wi h he ages (in
he case o Ru ino Tamayo he coe icien was nega i e bu no s a is ically signi ican ), and
posi i ely and signi ican ly ela ed o he age o he a is o he case o Fe nando Bo e o’s
pain ings. The e o e, Bo e o could be conside ed an expe imen alis and he i s ou a is s
we e concep ual a is s. These esul s a e consis en wi h he indings o Edwa ds (2004).
Hodgson and Hellmanzik (2019) and Hodgson (2022), wo mo e ecen and ela ed pape s,
looked a how mo emen associa ions migh in luence ca ee c ea i i y p o iles, which could
hen in luence c ea i i y. Using auc ion da a o 272 well-known mode n a is s – he majo i y o
whom we e ca ego ized in o mo emen s based on a his o ical e idence – Hodgson and
Hellmanzik (2019) applied a hedonic eg ession model and ound ha he a mo emen s o
which an a is is a ibu ed signi ican ly con ibu e o explaining auc ion p ices. Howe e , i is
impo an o conside ha a some poin du ing hei ca ee , an a is may ha e begun wo king in a
di e en a is ic mo emen . In hese si ua ions, we ha e included hose a is s in he ca ego y
mos closely ela ed o hem in ou pape based on he a his o y books we e iewed (we lis hose
books in Sec ion 5). Fu he mo e, as was p e iously men ioned, shi s in a is ic mo emen a e
no always o blame o he ac ha a is s’ p ices o pain ings di e acco ding o he pe iod o
hei li es in which hey p oduced hem [4]. One Venezuelan a is in ou da abase, Ca los C uz-
Diez (1923–2019), o ins ance, was a pa o he kine ic a mo emen , which is a subse o he
op-a mo emen , nea ly om he s a o his leng hy ca ee in he 1950s. Howe e , acco ding o
Galenson’s (2000) c i e ia, C uz-Diez could be conside ed a concep ual a is because his wo ks
om he la e 1950s and 1960s a e by a he mos aluable (see Ga ay e al., 2024).
Al hough i is easonable o assume ha some a is s in ou pape would ha e changed
ce ain aspec s o hei pain ings o e ime, such as he subjec s o mo i a ions, he e a e a
numbe o a is s who s ayed in he same a is ic mo emen h oughou hei li es. In he case
o Fe nando Bo e o (1932–2023), a Colombian a is who is also included in ou da abase,
Edwa ds (2004) ound ha his mos ecen wo ks (which a e cen e ed on Colombian socie y
and people and always ea u e a local na a i e) a e mo e aluable han his ea lie wo ks,
which include s ill li es and po ai s (no e ha we adjus ou eg essions o accoun o he
opic o pain ings, a con ol a iable ha has equen ly been o e looked in he li e a u e, as we
explain in he nex sec ion). Despi e hese changes in opics h oughou his long ca ee ,
Bo e o’s pain ings could be asc ibed o one a is ic s yle: he cos umb ismo mo emen , and we
hus ca ego ized him as belonging o ha a is ic s yle. By using an inc emen al echnique and
s i ing o pe ec ion in his pain ings, Bo e o may be classi ied as an expe imen al a is
acco ding o Galenson’s (2000) c i e ia.
4. Da a and me hodology
Ou da a se includes 30,288 auc ion sales o d awings and pain ings c ea ed by 293 a is s
om La in Ame ica [5]. The i s sale in he da abase was in Janua y o 1970, and he las sale
ook place in Decembe o 2014. The in o ma ion was ob ained om he Blouin A Sales
JEFAS
30,59
122
da abase. We hypo hesize, based on he hedonic p icing model o iginally p oposed by Rosen
(1974), ha he p ices o pain ings a e signi ican ly ela ed o a se o a iables in he
ollowing o m:
ln Pi
k ¼αþX
M
m¼1
βm3Xmk þμiþλ þυk (1)
Whe e:
ln Pi
k : P ice, in na u al loga i hm, o pain ing kauc ioned a yea (including he buye ’s
p emium o commission cha ged by he auc ion house o he buye ).
β
m
e lec s he cha ac e is ics o a ela i e shadow p ice o each o he a ibu es, whe e mis
he numbe o he conside ed a wo ks’ cha ac e is ics o a ibu es.
X
mk
ep esen s he cha ac e is ics (m) o each sold pain ing k, a pe iod (yea ) .
μi: S yle ixed e ec s.
λ : Time (yea ) ixed e ec s
The unc ional o m o he model is semi-loga i hmic. Higgs and Wo hing on (2005),
Campbell (2008),Campos and Ba bosa (2009),K €
aussl and Loghe (2010),Taylo and
Coleman (2011),Renneboog and Spaenje s (2012),S epano a (2016),Vosilo (2015a,b),
Pownall and G addy (2016),Ga ay e al. (2017),Cine a e al. (2019) and Ga ay (2021),
among o he au ho s, use his speci ica ion, as i p o ides a be e adjus men o he
eg ession. Equa ion (1) assumes ha he ma ke alua ion o each a ibu e does no change
h ough ime. Following he li e a u e (see, among o he s, Edwa ds, 2004;Higgs and
Wo hing on, 2005;Campbell, 2008;Campos and Ba bosa, 2009;K aeussl and Loghe , 2010;
Taylo and Coleman, 2011;Renneboog and Spaenje s, 2012;Pownall and G addy, 2016;
Ga ay e al., 2017;Ga ay, 2021), he hedonic eg ession model will be ini ially es ima ed by
unning an o dina y leas squa es eg ession (OLS).
Ou da ase comp ises bo h ime se ies and c oss-sec ional a iables. The e o e, we
pe o med he B eusch-Pagan Lag ange Mul iplie es o choose he app op ia e me hodology
o be applied (ei he an OLS pooled eg ession o a andom-e ec s panel eg ession). The null
hypo hesis ha a iances ac oss he p ices o pain ings pe a is a e ze o is ejec ed, and
he e o e, a andom-e ec s eg ession is p e e ed o a pooled eg ession. We applied a
Hausman es o selec ei he a andom-e ec s o a ixed-e ec s model and ejec ed he null
hypo hesis ha he e o s pe a is /pain ing a e no co ela ed wi h he eg esso s. The e o e,
we un panel eg essions wi h ixed e ec s. The s anda d e o es ima es a e obus o
dis u bances being au oco ela ed and he e oscedas ic.
An impo an ad an age o panel da a eg essions is ha hey enable us o examine he
impac o unobse able a iables on pain ings by indi idual a is s, as well as he e ec o
a iables ha change o e ime (bu no ac oss pain ings pe a is , see Ga ay e al., 2022b). We
es ima e ou model wi h s yle and ime ixed e ec s conside ing ha he g oup a iable in ou
panel is s yle. Wi h s yle ixed e ec s, we con ol o hose c oss-sec ional pain ing a ibu es
ha a ec pain ing p ices pe s yle bu ha do no change h ough ime. Wi h yea ixed e ec s,
we analyze he sys ema ic impac o ime- a ying a iables on pain ing p ices.
The hedonic p icing model ha we use includes he ollowing a iables:
4.1 Dependen a iable
The dependen a iable o he eg essions is he auc ion sale p ice o each pain ing (including
he buye ’s commission), exp essed in Napie ian loga i hm, and in eal dolla e ms (2014 US
dolla s).
Jou nal o
Economics,
Finance and
Adminis a i e
Science
123
0
20
40
60
80
100
120
140
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
La�n Ame ican a �s�c s yles p ice index
Panel B: Ele en La in Ame ican a is ic s yles a p ice indices (1970-2014, USD eal, 1970 = $100)
0
100
200
300
400
500
600
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Panel B1: La�n Ame ican Con empo a y (abs ac -geome ic, abs ac -in o mal and concep ual) a
p ice indices
Abs ac -geome ic
Abs ac -in o mal
Concep ual
Panel A: Agg ega e La in Ame ican s yle a p ice index (1970-2014, USD eal, 1970 = $100)
Figu e 1. La in Ame ican a is ic s yles p ice indices
JEFAS
30,59
130
impo ance, as i highligh s he need o decompose a p ice indices by a is ic s yles/
mo emen s o gain an enhanced di e si ica ion when in es ing in La in Ame ican a ac oss
di e en a is ic s yles.
0
50
100
150
200
250
300
350
400
450
500
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Panel B3: La�n Ame ican cubism, su ealism, a an -ga de, and XIX cen u y a p ice indices
Cubism
Su ealism
XIX Cen u y
A an -ga de
0
50
100
150
200
250
300
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Panel B2: La�n Ame ican cos umb ismo, figu a� e, mu alis and landscape a p ice indices
Cos umb ismo
Figu a� e
Mu alism
Landscape
Figu e 1. (con inued)
Jou nal o
Economics,
Finance and
Adminis a i e
Science
131
Table 4. In es men pe o mance by La in Ame ican a is ic s yle, S&P 500 and US 10-yea Go e nmen bonds (USD eal e u ns)
Yea ly
e u ns/s yle
Abs ac -
geome ic
Abs ac -
in o mal Concep ual Cos umb ism Cubism Figu a i e Mu alism Landscape Su ealism
Nine een h
cen u y
A an
ga de S&P 500 US Bonds
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)
1971–2014 1971–2014 1996–2014 1971–2014 1981–2014 1971–2014 1971–2014 1971–2014 1971–2014 1971–2014 1971–2014 1971–2014 1971–2014
A i hme ic
mean e u n
(%)
9.99 3.54 18.41 14.76 12.95 2.39 1.35 0.01 0.12 5.35 5.10 7.57 5.48
Geome ic
mean e u n
(%)
1.97 �0.70 10.33 0.91 2.02 �1.14 �0.73 �3.37 �2.20 �1.77 �3.60 5.86 0.72
S anda d
de ia ion
(%)
47.19 31.34 59.01 53.13 52.03 28.65 21.10 27.97 21.72 39.94 44.99 17.37 8.96
Sha pe a io 0.20 0.09 0.30 0.26 0.23 0.06 0.03 �0.03 �0.03 0.12 0.10 0.39 0.53
Sou ce(s): Bloombe g and own calcula ions
JEFAS
30,59
132
Table 5. Co ela ion ma ix o La in Ame ican a is ic s yles, S&P 500 and US 10-yea go e nmen bonds (USD eal e u ns)
Abs ac -
geome ic
Abs ac -
in o mal Concep ual Cos umb ismo Cubism Figu a i e Mu alism Landscape Su ealism
Nine een h
cen u y
A an
ga de
S&P
500
US
Bonds
Abs.-geome ic 1.00 �0.04 �0.26 �0.09 0.19 �0.20 0.08 �0.03 0.34 0.24 �0.08 0.06 �0.17
Abs.-in o mal 1.00 �0.09 0.09 0.12 0.31 0.38 0.28 0.11 0.28 0.04 �0.11 0.07
Concep ual 1.00 0.05 0.02 0.55 0.45 0.07 �0.01 0.37 �0.04 �0.32 �0.12
Cos umb ismo 1.00 0.15 0.11 �0.01 0.17 �0.18 �0.20 �0.02 0.01 0.05
Cubism 1.00 0.12 0.05 0.16 0.04 �0.04 �0.36 �0.31 �0.39
Figu a i e 1.00 0.51 0.40 �0.08 0.25 0.06 �0.08 0.13
Mu alism 1.00 0.44 0.40 0.35 �0.12 �0.13 �0.07
Landscape 1.00 0.18 0.29 0.16 0.05 �0.03
Su ealism 1.00 0.31 0.01 0.02 �0.01
Nine een h
Cen u y
1.00 0.32 0.11 0.00
A an Ga de 1.00 0.03 0.12
S&P 500 1.00 0.49
US Bonds 1.00
No e(s): All he e u n co ela ions a e o he pe iod 1971–2014, excep o he cases o concep ual a (1996–2014) and cubism (1981–2014)
Sou ce(s): Bloombe g and own calcula ions
Jou nal o
Economics,
Finance and
Adminis a i e
Science
133
The mean annual geome ic eal e u n o he La in Ame ican a p ice index was �0.9%
be ween 1970 and 2014, compa ed o a posi i e mean annual eal e u n o 0.8% o he wo ld
a ma ke p ice index du ing he same pe iod (es ima ing he a e age e u ns o he wo ld a
ma ke om he esul s epo ed by Li e al., 2022). As commen ed be o e, he highes La in
Ame ican a s yle e u ns we e p o ided by concep ual a (10.33%), cubism (2.02%) and
abs ac -geome ic a (1.97%). Abs ac -geome ic and concep ual a a e pos -wa and
con empo a y a mo emen s ha , du ing he sample pe iod, also p o ided he highes e u ns
a he wo ld le el ( o example, when compa ed o he Abs ac Exp essionism/Pop/
Minimalism and Con empo a y a p ice index es ima ed by Li e al., 2022). Howe e , du ing
ou sample pe iod, he a p ice index o Dada and Su ealism, Fu u ism and Cons uc i ism,
Cubism and Fau ism and Exp essionism es ima ed by Li e al. (2022) was i ually la on he
wo ld a ma ke s. The a e age annual eal e u ns p o ided by he La in Ame ican a p ice
indices es ima ed he e and ela ed o ha index we e: Cubism (2.02%), su ealism (�2.20%)
and a an -ga de (�3.60%).
The a e age o he h ee indices yields an annual eal e u n o �1.26%, which is
ma ginally wo se han he La in Ame ican mean annual eal e u n p e iously epo ed. These
s yles also unde pe o med he Abs ac Exp essionism/Pop/Minimalism and Con empo a y
a p ice index es ima ed by Li e al. (2022). This sugges s ha he e may be a posi i e
co ela ion be ween some o he La in Ame ican a p ice s yle index e u ns calcula ed he e
and hei global coun e pa s. This conclusion is consis en wi h ou ea lie quo a ion om
Pe ez-Ba ei o (in Cas o, 2013), who con ends ha ce ain s yles, such as La in Ame ican
geome ic abs ac ion, can be classi ied as belonging o he global abs ac -geome ic a
mo emen , and a gues ha some s yles, like La in Ame ican geome ic abs ac ion, can be
ca ego ized as pa o he global abs ac -geome ic a mo emen .
Acco ding o Schulze (1999),Goodwin (2008),Renneboog and Spaenje s (2012),S eine
e al. (2013),Shi e al. (2017) and Ga ay (2018), he e is a home bias in a in es ing since
collec o s a e mo e likely o buy wo ks by a is s who a e o he same na ionali y. S eine e al.
(2013) a gue ha he e is a signi ican home bias in p i a e a collec ions ac oss all
con inen s and na ions, which can be pa ially explained by e y s ic impo and expo
egula ions. Also, acco ding o Ma in (1999, pp. 4–5), who analyzed he La in Ame ican a
ma ke : “The audience a hose i s (La in Ame ican A ) auc ions (la e 1970s, ea ly 1980s)
ended o si oge he in li le g oups acco ding o na ionali y. In he g ea es numbe s we e
he Mexicans, who bough abou 40 pe cen o he o e ings. Ano he g oup was he
Venezuelans, mo e “pan-La in” in ha hey we e in e es ed in a om a a ie y o coun ies.
Fo example, a Venezuelan migh buy a Mexican pain ing like a Ri e a o a Tamayo (a
Mexican a is ), bu a Mexican would no buy a Re e on (a Venezuelan a is ). The emaining
La in Ame icans bough a om hei own coun y only and ailed o see any pa allels wi h
he a o close neighbo s.” Towa d he mid o he la e 1980s, and a e a ew b eak h ough
museum exhibi ions, US collec o s also began o acqui e he wo ks o impo an La in
Ame ican mas e s. Some 20 yea s la e , a he ime o he publica ion o he book, The an
admi ed ha he “home bias” was no as p e alen as be o e, bu ha i was s ill an
impo an o ce.
In he p eceding quo a ion, Ma in (1999) e e s o La in Ame ican a auc ions ha ook
place in he USA. Since auc ions a e p i a e ma ke s whe e buye s’ iden i ies a e sa egua ded,
in o ma ion abou he names o he buye s o a wo k is i ually nonexis en . As a esul , a
sys ema ic eco d o he na ionali ies o auc ion bidde s canno be ound. I may be in e ed
ha , wi h he possible excep ion o some auc ions held in New Yo k Ci y and London, he
majo i y o bidde s a local auc ions wo ldwide a e ci izens o he na ion whe e he sale is being
held. Since he au ho was p esen a hose auc ions and hus had i s -hand knowledge, we
hink Ma in’s quo e is e y ele an . Ma in (in o mally) con i med ha he majo i y o
buye s a he La in Ame ican a auc ions held in he USA be ween he la e 1970s and a leas
un il he end o he 1990s we e om La in Ame ica. Acco ding o Ga ay (2018), he imposi ion
o ex emely high axes on he impo and expo o a wo k in ce ain coun ies exace ba es
JEFAS
30,59
134
he home bias in he a ma ke . Addi ionally, anspo ing a wo ks, pa icula ly sculp u es,
can be e y expensi e (see Vosilo , 2015a,b).
A o eigne who buys an a wo k in A gen ina mus na iga e a numbe o challenges,
acco ding o A eaga (2017) ( he ollowing ansla ion o English is ou s): “[. . .] 1) eques an
app aisal om he Banco Ciudad, which akes a leas 24 h; 2) once he buye has his
app aisal, he o she has o go o he Depa men o Visual A s, which depends on he Na ional
Sec e a ia o Cul u e, o i o au ho ize he wo k o be aken ou o he coun y, a p ocedu e
which akes abou en days; 3) i he buye has al eady gone back o his o he coun y o o igin
o o ano he coun y, he galle is will ha e o hi e a cus oms b oke o send he piece, which
cos s a minimum o 1,000 dolla s, ai po o ai po , and 3,000 dolla s i he inal des ina ion is
he buye ’s home [. . .]. Bu , in addi ion, he impo a ion o wo ks o a in o ou coun y is
axed a app oxima ely 17.5%”. In he case o B azil, a 40% ax is le ied when impo ing
a wo ks (Ga ay, 2018). Finally, well-known a is s’ c ea ions a e equen ly designa ed as
na ional he i age, a p ac ice ha makes i ex emely di icul o nego ia e and expo a wo k
om he espec i e na ions. An example o his is ep esen ed by he pain ings c ea ed by
Mexican a is F ida Kahlo (1907–1954), which we e decla ed as pa o he na ional he i age
by he Mexican Go e nmen in 1984.
Acco ding o Angelini e al. (2023), an expo e o ha has been in place in I aly since 1939
may ha e an impac on p ices i an a wo k was p oduced mo e han 50 yea s be o e he da e o
sale by he a is (who is no longe ali e a ha ime). The au ho s es ima e a model o accoun
o and analyze he impac o he expo e o law on p ices while con olling o he possibili y
o sample selec ion bias using hand-collec ed da a co e ing all a wo ks c ea ed by non-li ing
mode n and con empo a y I alian a is s and sold a Ch is ie’s and So heby’s in London and
Milan be ween 2012 and 2016. Fo a wo ks sold in I aly and hose made mo e han 50 yea s
be o e he sale da e, he e ec o ising a p ices be ween he yea o c ea ion and he sale da e
is e e sed, acco ding o Angelini e al. (2023). P e-sale es ima es likewise exhibi a simila
pa e n. The au ho s hypo hesize ha ansac ion cos s – such as anspo a ion, insu ance and
s o age expenses, which a e signi ican in he a indus y – a e he p ima y sou ce o ma ke
ine iciencies and ha expo e o ules may be d i en by o combined wi h a home bias
impac .
Na ionals a e mo e in e es ed in some a is ic gen es han o eign collec o s (The an, 1999).
Fo ins ance, when i comes o landscape pain ings, collec o s a e mo e likely o alue hose
ha ea u e hei own na ions and a eas han hose ha ea u e loca ions ha a e un amilia o
hem. S eine e al. (2013) a gue ha collec o s “. . . a e said o ha e a endency o acqui e a
objec s ela ed o hei own coun y. This can e e o he objec s shown in a piece o a , such
as local landscapes o pe sons d essed acco ding o local cus om. Mos impo an ly, many
collec o s ocus on he a p oduced by a na ional, egional o local a is . I is a gued ha he
collec o s eel a special a achmen o, o a special as e o , a is s sha ing he same cul u e,
his o y, and na ionali y and whom hey some imes know pe sonally.”
Fo h ee o La in Ame ica’s a s yle indices ha we es ima ed: Landscapes (�3.37%),
cos umb ismo (0.91%) and nine een h cen u y a (�1.77%), and pe haps less so in he case o
mu alism (�0.73%), hei in e io geome ic a e age eal annual e u ns du ing ou sample
pe iod (on a e age, �1.24% e sus �0.9% o he agg ega e La in Ame ican index) is
consis en wi h he highe desi abili y ha locals end o display o hese a is ic s yles
(compa ed o in e na ional collec o s). This is in addi ion o he ac ha a numbe o La in
Ame ican na ions wen h ough se ious economic c ises h oughou he esea ch pe iod,
including he ea ly 1980s deb c isis and hype in la ion occu ences in he same decade. One
could a gue ha a compa a i ely poo e g oup o local (La in Ame ican) collec o s ended up
wi h lowe e u ns a e buying a la ge pe cen age o a wo ks om a is ic s yles c ea ed by
local (La in Ame ican) a is s. I would be necessa y o es ima e a s yle p ice indices o e e y
na ion o es his hypo hesis mo e p ecisely. Un o una ely, ou in e ence can only be
conside ed a conjec u e because we lack su icien obse a ions o a numbe o a s yle p ice
indices a he na ional le el o pe o m such an analysis.
Jou nal o
Economics,
Finance and
Adminis a i e
Science
135
When inancial analys s b eak down s ocks by di e en ca ego ies (e.g. alue e sus
g ow h s ocks), ou me hod o b eaking down a egional a p ice index (e.g. La in Ame ica) by
a is ic s yles wi hin ha egion migh be compa able o ha p ocedu e.
Las ly, wo undamen al ideas in inance and ma ke ing a e he complemen a i y and
subs i u abili y o goods. Two i ems a e conside ed complemen a y when hey a e ypically
bough oge he , bu in e changeable when a buye can swap ou one o he o he in
ma ke ing and e ail (Tian e al., 2021). Cul u al e en s ha e been used o in es iga e he
impac s o complemen a i y and subs i u ion. Meleddu and Pulina (2024), o ins ance,
e alua e he e ec s o cul u al e en s in u al communi ies on he Medi e anean island o
Sa dinia (I aly) in e ms o complemen a i y and subs i u ion. People who a end an e en in
one communi y a e mo e likely o isi ano he communi y whe e a simila e en is held,
acco ding o he complemen a i y e ec (complemen a i y e ec s a e likely o a ise unde a
coope a ion s a egy ac oss communi ies). Con e sely, he subs i u ion e ec sugges s ha
he e is a ade-o as isi ing one g oup educes he likelihood o isi ing ano he . When
communi ies compe e o achie e a sel -ad an ageous ou come, a subs i u ion e ec is likely
o occu .
The deg ee o which he e u ns o wo o mo e asse s mo e in opposing di ec ions unde
ce ain ci cums ances is e e ed o as complemen a i y in he con ex o po olio
di e si ica ion (Ga ay, 2005;Co e, 2021). As p e iously men ioned, we disco e ed a e y
low a e age co ela ion o 0.12 be ween he ele en a is ic s yles o La in Ame ican a .
We belie e his esul is ex emely signi ican because i emphasizes he need o b eak down a
p ice indices by a is ic mo emen s o s yles in o de o enhance po olio di e si ica ion when
in es ing in La in Ame ican a ac oss a ious a is ic s yles. We conjec u e ha he
impo ance o dissec ing egional a collec ing by a is ic mo emen s could also a ise in o he
egions o sub-con inen s o which auc ion houses o ganize sales (e.g. Ch is ie’s Indian,
Himalayan and Sou heas Asian A and So heby’s Mode n and Con empo a y Middle Eas ).
Fu he mo e, we highligh ha he a e age co ela ion o he La in Ame ican a s yles wi h US
s ocks and US bonds is e en lowe , a �0.06 and �0.04, hus sugges ing ha hese asse s a e
complemen a y.
7. Robus ness es s and ex ensions
Fou obus ness es s and ex ensions o ou esul s a e p esen ed in his sec ion. Fi s , we
es ima e a p ice indices o each La in Ame ican coun y in he sample (de ined acco ding o
he na ionali y o he a is s in he sample). Hedonic eg essions we e used o es ima e a p ice
coun y indices o each g oup o a is s, b oken down by na ionali y. Figu e 2 shows ou
es ima ed La in Ame ican a p ice index (Panel A) [8], and he es ima ed indices o
A gen ina, B azil, Chile, Colombia, Cuba, Mexico, Venezuela and o he coun ies (Ecuado ,
Pe u and U uguay, “EPU” o “OTH”) [9], Panel B. The highes cumula ed eal e u ns we e
eco ded by Colombia, Chile, EPU and B azil, espec i ely. A gen ina, Mexico and Cuba
exhibi ed ela i ely modes e u ns, and Venezuela had nega i e cumula ed e u ns. Mos o
he a p ice coun y indices peaked a ound 1990, a inding ha coincides wi h o he e idence
on wo ld a p ices a ound ha ime (see Li e al., 2022). A na u al ex ension o hese esul s
would be o es ima e a is ic s yle indices wi hin each La in Ame ican coun y. Un o una ely,
as we p e iously s a ed, we we e unable o do his analysis due o he small sample size ha
was a ailable in he majo i y o he coun ies in ou sample [10].
Second, we es ima ed a p ice indices o each o he 13 opic/mo i e a iables included in
ou eg essions: Abs ac , animal, landscape, nude, objec , people, po ai , sel -po ai ,
eligion, s ill li e, un i led, u ban and o he s. A p ice indices we e es ima ed using hedonic
eg essions o each g oup o a is s, ca ego ized by opic. The opics ha did be e han he
es o he La in Ame ican ma ke we e abs ac and eligion. In he case o abs ac , he las
20 yea s ha e seen he mos o e pe o mance. In keeping wi h global a ma ke s, he abs ac
opic also saw a sha p ise and all a ound 1990. The landscape and u ban opics had a nega i e
JEFAS
30,59
136
0
20
40
60
80
100
120
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
La�n Ame ican coun y a p ices
Panel A: Agg ega e La in Ame ican coun y a p ice index (1970-2014, USD eal, 1970 = $100)
0
500
1000
1500
2000
2500
J-69
J-70
J-71
J-72
J-73
J-74
J-75
J-76
J-77
J-78
J-79
J-80
J-81
J-82
J-83
J-84
J-85
J-86
J-87
J-88
J-89
J-90
J-91
J-92
J-93
J-94
J-95
J-96
J-97
J-98
J-99
J-00
J-01
J-02
J-03
J-04
J-05
J-06
J-07
J-08
J-09
J-10
J-11
J-12
J-13
J-14
A gen�na
B azil
Chile
Colombia
Cuba
Mexico
Venezuela
Ecuado , Pe u and U uguay
Panel B: Eigh La in Ame ican coun y a p ice indices (1970-2014, USD eal, 1970 = $100)
Figu e 2. La in Ame ican coun y a p ice indices
Jou nal o
Economics,
Finance and
Adminis a i e
Science
137
cumula i e e u n and pe o med wo se han he o he indexes. These wo opics end o ha e a
highe pe cen age o local buye s, as we men ioned in he p e ious sec ion [11].
Thi d, we di ided ou sample by gen es and es ima ed a p ice indices. Ou sample had 17
women a is s and 276 man a is s. Figu e 3 shows he esul s. The e u n o he women a
p ice index clea ly o e pe o med ha o he male a p ice index, a inding ha is consis en
wi h he posi i e coe icien on he a iable Women ha we ob ained ac oss he eg essions
shown in he ou speci ica ions o he eg essions o he agg ega e La in Ame ican a ma ke
shown in Table A1 (Appendix 2). Fu he mo e, his o e pe o mance exhibi ed an upwa d
end h oughou he sample pe iod. Adams e al.’s (2021) s udy ound a 42.1% emale-women
discoun in in e na ional auc ion p ices o pain ings o a pe iod ha is e y simila o ou s
(1970–2013), con adic ing ou indings. Addi ionally, LeBlanc and Sheppa d (2021)
disco e ed ha emale a is s ecei ed unadjus ed discoun s o o e 40%. Howe e , La in
Ame ican women a is s in Edwa ds’s (2004) sample had he highes a es o e u n (32.04%),
indica ing ha his indings we e in line wi h ou s.
Ou esul s a e also in line wi h he indings o Came on e al. (2019), who es ed o gende
impac s in he a ma ke using auc ion p ices o Yale School o A s g adua es. They ound
ha he a wo ks o emale g adua es we e signi ican ly less likely o be o e ed a auc ion a e
con olling o ime e ec s, base g adua ion yea a e and o he a iables. Howe e ,
condi ioning on appea ing a auc ion, he a e age p ice ob ained om women g adua es was
highe , con olling o a hos o hedonic cha ac e is ics. Acco ding o Boca e al. (2022), a
glass ceiling exis s o women a is s, which is e idenced by hei inding ha a wo ks
execu ed by women a is s sell o an a e age p emium o 4%. These au ho s also disco e ed
ha he e is an abno mal demand o he ew wo ks c ea ed by a small numbe o excep ional
emale a is s, which is consis en wi h he idea ha he p esence o a ew emale supe s a s
skews he esul s. Con e sely, ea nings a e spli mo e equi ably o male a is s. Fu u e s udies
should in es iga e possible causes o La in Ame ican women a is s’ o e pe o mance in
compa ison o hei male coun e pa s.
0
100
200
300
400
500
600
700
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Gende Indexes
Male
Female
No e(s): 1977–2014, USD eal, 1977 = $100
Figu e 3. La in Ame ican emale and male a p ice indices
JEFAS
30,59
138
Fou h, we es ima ed a p ice index o La in Ame ican a wo ks sold a So heby’s and
Ch is ie’s, and a p ice index o he emaining lo s in he ull sample (lo s o La in Ame ican
a wo ks a o he auc ion houses, nea ly all o hem we e local auc ion houses, meaning hey
we e loca ed in he coun y whe e he a is s’ wo k was sold). Figu e 4 shows he e olu ion o
bo h p ice indices. The Ch is ie’s and So heby’s index clea ly ou pe o med he local auc ion
houses index, yielding an annual compounded eal e u n o 1.65%, compa ed o a g im e u n
o �2.23% o he o he auc ion houses index. The esul s o Ch is ie’s and So heby’s also
imply an unde pe o mance o La in Ame ican a when compa ed o he esul s epo ed by Li
e al. (2022) o he agg ega e wo ld a ma ke s. Du ing he same pe iod as ou s (1970–2014),
hese au ho s ound ha an index o a wo ks sold a Ch is ie’s and So heby’s o e ed an annual
compounded eal e u n o 2.37%, whe eas an index o local auc ion houses o e ed an annual
compounded eal e u n o �0.51%. Addi ionally, since he i s a bubble o he la e 1980s,
he di e ence be ween he Ch is ie’s and So heby’s index and he local a ma ke s index has
g own, which is consis en wi h he indings o Li e al. (2022). Fu he mo e, compa ed o he
in e na ional a ma ke s (as desc ibed in Li e al., 2022), he gap be ween he Ch is ie’s and
So heby’s index and he o he auc ion houses index is conside ably wo se in he case o he
La in Ame ican a ma ke s.
The beha io o he wo indices ha we calcula ed o La in Ame ica in he la e 1970s and
ea ly 1980s shows a di e ence be ween he esul s epo ed in he wo pape s. Speci ically, he
local auc ion house index expe ienced a sha p decline, which likely e lec ed he global
ecession o ha ime and he onse o he La in Ame ican deb c isis (Mexico de aul ed on i s
deb s in 1982, s a ing he egion’s deb c isis). In con as , he Ch is ie’s and So heby’s indices
(ou s o La in Ame ican a , and Li e al., 2022 o he global a ma ke ) show signi ican
p ice inc eases in he la e 1980s, peaking a ound 1990 be o e plumme ing p ecipi ously. They
also show accele a ed p ice inc eases p io o he Global Financial C isis o 2008–2009, which
we e ollowed by signi ican declines.
Las ly, we also ied o sepa a e he sample in o “medium and small” and “big” auc ion
houses. None heless, we ound ha he co esponding indices exhibi ed ema kably
compa able beha io o he a p ice indices de eloped o Ch is ie’s, So heby’s and he
o he auc ion houses men ioned in he p eceding pa ag aphs. This is ha dly su p ising gi en
ha , be ween 1970 and 2014, Ch is ie’s and So heby’s; and he o he auc ion houses di e ed
signi ican ly no jus in e ms o p es ige bu also in e ms o he o al amoun o money ha was
sold a auc ion o he a e age p ice pe lo . When i comes o he a e age p ice pe lo sold,
Phillips is he only auc ion house ha can ma ch Ch is ie’s and So heby’s. Howe e , i s
0
50
100
150
200
250
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Ch is�e’s and So heby’s
O he auc�on houses
No e(s): USD eal, 1970 = $100
Figu e 4. La in Ame ican a p ice indices o Ch is ie’s and So heby’s, and o o he auc ion houses
Jou nal o
Economics,
Finance and
Adminis a i e
Science
139
Vosilo , R. (2015a), “Sculp u e as an al e na i e in es men : an analysis o p ice dynamics be ween
sculp u e and equi y and bond ma ke s”, Jou nal o Al e na i e In es men s, Vol. 17 No. 4,
pp. 21-45, doi: 10.3905/jai.2015.17.4.021.
Vosilo , R. (2015b), “A auc ion p ices: home bias, amilia i y and pa io ism”, a ailable a : h ps://
ss n.com/abs ac 52686527
Wang, F. (2023), “Do eme ging a ma ke segmen s ha e hei own p ice dynamics? E idence om
he Chinese a ma ke ”, In e na ional Re iew o Economics and Finance, Vol. 84, pp. 318-331,
doi: 10.1016/j.i e .2022.11.015.
Appendix 1
Topic/mo i e dummies
He e, we explain he c i e ia used o de e mine he opic/mo i e o each pain ing in he sample. Following
Renneboog and Spaenje s (2012), we examined he i le wo ds o each pain ing in he sample o alloca e
wo ks o a o a se o opic dummies. Mos o he i les we e in Spanish, ollowed by English, Po uguese
and F ench. Renneboog and Spaenje s (2012) ca ego ized hei sample by opic/mo i e as: People,
po ai , sel -po ai , landscape, nude, eligion, s ill li e and o he . We ollowed a simila c i e ion and
added i e new ca ego ies, o be e e lec he mos popula La in-Ame ican opics/mo i es: People,
po ai and sel -po ai , landscape, nude, eligion, s ill li e, abs ac , animals, objec s, u ban, un i led and
o he . Following he p ocedu e used by Ga ay (2021) and De Ridde e al. (2024), in se e al cases we had
o analyze he image o a pain ing o asce ain he espec i e opic dummy a iable.
JEFAS
30,59
146
Appendix 2
Agg ega e eg essions (all La in Ame ican a is ic s yles), eal dolla p ices
Table A1. Agg ega e eg essions (all La in Ame ican a is ic s yles), eal dolla p ices
(1) (2) (3) (4)
Ln(P ice) Ln(P ice) Ln(P ice) Ln(P ice)
A ea Ln(a ea) 0.441*** 0.477*** 0.520*** 0.516***
(0.00703) (0.00723) (0.00687) (0.00693)
Technique (omi ed a iable:
ac ylic)
Cha coal 0.696*** 0.999***
(0.151) (0.157)
Gouache 0.00633 0.223
(0.159) (0.165)
Ink �0.00563 0.298
(0.193) (0.200)
Mixed 0.309*** 0.331***
(0.0249) (0.0258)
Oil 0.690*** 0.819***
(0.0205) (0.0210)
O he 0 0
(.) (.)
Pas el 0.408*** 0.795***
(0.104) (0.108)
Pencil 1.299*** 1.671***
(0.104) (0.107)
Tempe a 0.203 0.589
(0.366) (0.380)
Wa e colo 0.489*** 0.585***
(0.0555) (0.0575)
Wo k on pape 0.0463 0.153***
(0.0289) (0.0298)
Auc ion house (omi ed a iable:
o he houses)
Topic (omi ed a iable: o he s yles)
Ch is ie’s 1.032*** 1.202***
(0.0310) (0.0322)
So heby’s 1.007*** 1.193***
(0.0319) (0.0330)
Abs ac 0.856*** 1.042***
(0.0319) (0.0328)
Animal 0.117** 0.107**
(0.0462) (0.0483)
Landscape 0.225*** 0.241***
(0.0314) (0.0328)
Nude 0.186** 0.137**
(0.0643) (0.0673)
Objec 0.212*** 0.315***
(0.0518) (0.0542)
People 0.401*** 0.413***
(0.0259) (0.0271)
Po ai 0.198*** 0.251***
(0.0427) (0.0447)
Religion 0.438*** 0.438***
(0.0419) (0.0438)
Sel -po ai 0.487*** 0.489***
(0.0836) (0.0875)
S ill li e 0.211*** 0.252***
(0.0357) (0.0372)
Un i led 0.0417 �0.0396
(0.0294) (0.0307)
U ban 0.186*** 0.260***
(0.0368) (0.0385)
(con inued)
Jou nal o
Economics,
Finance and
Adminis a i e
Science
147
Table A1 p esen s he esul s o he eg essions o he agg ega e sample o La in Ame ican a is s,
using he ollowing ou di e en speci ica ions: including all he con ol a iables (column 1),
excluding auc ion houses and opics (column 2), excluding echniques and opics (column 3) and
excluding echniques and auc ion houses (column 4). O e all, he p ices o pain ings inc ease wi h he
a ea o he a wo k (1% le el o signi icance ac oss all he eg essions). This inding has al eady been
epo ed in he li e a u e (see, o example, Higgs and Wo hing on, 2005;Campbell, 2008;Taylo and
Coleman, 2011;Renneboog and Spaenje s, 2012; and he e iew p esen ed in Ga ay, 2018). Pain ings
execu ed in oil, wa e colo , pas el, pencil, mixed ma e ials and cha coal eco ded p ices ha we e highe
han ac ylic, he echnique a iable ha was omi ed om he eg ession (1% le el o signi icance). The
inding o oil is consis en wi h he exis ing li e a u e (see, o example, Renneboog and Spaenje s,
2012;Ga ay, 2021).
P ices o a wo ks auc ioned a Ch is ie’s and So heby’s we e signi ican ly highe (a he 1% le el)
han hose sold a o he auc ion houses. This inding is consis en wi h he e iew p esen ed in Ga ay
(2018). Ch is ie’s and So heby’s a e ega ded as he wo mos epu able auc ion houses. Wo ks ha
we e da ed e ched highe p ices (1% le el o signi icance). Resul s also sugges ha wo ks sold in
New Yo k Ci y commanded highe p ices (1% le el o signi icance). These wo esul s a e consis en
wi h he e idence epo ed by Renneboog and Spaenje s (2012) and Ga ay (2021), among o he
au ho s. Signed wo ks had lowe p ices. This coun e in ui i e esul has also been ound by o he
au ho s, such as Campos and Ba bosa (2009), o La in Ame ican wo ks sold a So heby’s, and Ga ay
(2021), o he case o Venezuela. The ac ha some auc ion houses do no o e accu a e in o ma ion
on hei ca alogues ega ding whe he a wo k o a is signed o no may pe haps explain his puzzling
inding.
Table A1. Con inued
(1) (2) (3) (4)
Ln(P ice) Ln(P ice) Ln(P ice) Ln(P ice)
O he cha ac e is ics Signed �0.249*** �0.277*** �0.254*** �0.201***
(0.0208) (0.0216) (0.0216) (0.0217)
Da ed 0.268*** 0.274*** 0.210*** 0.245***
(0.0154) (0.0160) (0.0160) (0.0160)
New Yo k Ci y 0.654*** 1.475*** 0.560*** 1.575***
(0.0294) (0.0172) (0.0305) (0.0170)
Ali e �0.424*** �0.455*** �0.482*** �0.503***
(0.0174) (0.0180) (0.0181) (0.0181)
Women 0.524*** 0.463*** 0.407*** 0.570***
(0.0375) (0.0388) (0.0390) (0.0392)
_cons 5.887*** 6.109*** 5.911*** 5.599***
(0.140) (0.143) (0.143) (0.145)
N
2_o 30,288 30,288 30,288 30,288
2_w 0.506 0.461 0.456 0.457
2_b 0.496 0.454 0.446 0.447
2_a 0.671 0.637 0.581 0.502
Time ixed
e ec s
0.495 0.453 0.445 0.446
S yle ixed
e ec s
Yes Yes Yes Yes
Con ols All Ma e ial Auc ion
House
Topic
No e(s): S anda d e o s in pa en heses
*p< 0.10, **p< 0.05, ***p< 0.01
Sou ce(s): Own calcula ions, based on a ma ke in o ma ion ob ained om Blouin A
JEFAS
30,59
148
Wo ks by a is s ha ha e passed away by he ime o he auc ion end o ha e highe p ices. This
inding is consis en wi h he e idence p esen ed by Higgs and Wo hing on (2005). Ve y in e es ingly,
wo ks by women a is s end o e ch highe p ices (we discuss his issue wi h mo e de ail on he
obus ness and ex ensions sec ion). Finally, all he pain ing opics had p ices ha we e signi ican ly highe
han hose o o he s yles, he ca ego y ha was le ou o he eg ession.
Co esponding au ho
U bi Ga ay can be con ac ed a : [email p o ec ed]
Fo ins uc ions on how o o de ep in s o his a icle, please isi ou websi e:
www.eme aldg ouppublishing.com/licensing/ ep in s.h m
O con ac us o u he de ails: [email p o ec ed]
Jou nal o
Economics,
Finance and
Adminis a i e
Science
149
