637
WHAT DETERMINES THE PRICE OF PORT WINE?
Rui Cou o Viana
Lúcia Lima Rod igues
RESUMO:
A indús ia do inho do Po o ca ac e iza-se pela longe idade das suas exis ências. De ido à na u eza
das p á icas e dos no ma i os con abilís icos, es as exis ências de longa du ação são egis adas no
ac i o co en e e medidas ao cus o. Es e g upo de ac os, em associação com a in lação, p o oca um
p oblema con abilís ico: a suba aliação das exis ências. Es e a igo analisa a ques ão: O que de e mina
o p eço do Vinho do Po o? Um modelo hedónico oi es imado de o ma aexplica as a iações do
p eço em ace das al e ações das ca ac e ís icas do Vinho do Po o. Os esul ados indiciam que a idade
é um dos de e minan es do p eço (além de ou os). Es e ac o con i ma a con icção de que o p ocesso
de en elhecimen o dos inhos, em conjugação com as no mas e as p á icas con abilís icas, bem como
a in lacção, conduzem a um p oblema con abilís ico.
PALABRAS CLAVES:.Con abilidade de Vinhos, Modelos hedónicos
ABSTRACT:
Po wine indus y is cha ac e ized by i long-li ed in en o ies. Due o accoun ing s anda ds and
p ac ices he long-li ed in en o ies a e ecognized in he cu en asse s accoun and a e measu ed a
cos . This g oup o ac s, in associa ion wi h in la ion, leads o an accoun ing p oblem: he
unde es ima ed long-li ed in en o ies. This pape add esses he ques ion: wha de e mines he p ice o
Po Wine? A hedonic model is es ima ed o explain he a ia ion in p ice be ween di e en Po wine
cha ac e is ics. Ou esul s sugges ha age is ma ke p ice de e minan . This con i m he idea ha
ma u ing o ageing he wine has an e ec on ma ke p ice ha , associa ed wi h accoun ing s anda ds
and p ac ices and in la ion, esul s in an accoun ing p oblem.
KEY WORDS: Wine accoun ing, hedonic model
______________________________________________________________________
1. INTRODUCTION
The Po wine indus y ep esen s one he ew Po uguese global p oduc s. Ve y li le economic esea ch has
been comple ed on he indus y, in pa icula , no sys ema ic s udy has been unde aken o iden i y and de e mine
ma ke alue o he indi idual a ibu es, which make up he p oduc called "Po Wine".
Po wine is cha ac e ized by i s la ge he e ogenei y. When consume s wan o buy a bo le o Po wine hey
ace se e al "Po Wines" wi h di e en p ices – young, old, Vin age, LBV, Tawny, Ruby, special ca ego ies,
e c. A Po wines company p oduces bo les wi h a wide ange o p ice, age and o he cha ac e is ics. This
he e ogenei y makes Po wine a sui able p oduc o a hedonic p ice analysis.
CITIES IN COMPETITION
638
Accoun ing in he wine indus y has de eloped g ea ly in ecen yea s. The p emium wine was he s a ing poin
o ha de elopmen . As Blake e al. (1998) s a ed he applica ion o a adi ional accoun ing amewo k o he
wine indus y has a nega i e e ec on p emium wine as compa ed o low-quali y wine. So, in o de o p omo e
quali y p oduc ion in he wine indus y majo amendmen s o inancial accoun ing and managemen accoun ing
a e needed. Ou wo k in ends o be a con ibu ion o he de elopmen o wine accoun ing.
Viana and Rod igues (2004) showed ha he adi ional accoun ing amewo k is no a sui able amewo k
ega ding long-li ed quali y Po wine in en o ies and ad oca ed ha Po wine indus y need, ega ding he
measu emen o he long-li ed in en o ies, a special accoun ing ea men . In his pape , we analyse he Po
wine de e minan s, in conc e e, i he age (o he ageing winemaking p ocess) a ec s he ma ke p ice.
In his s udy he de e mina ion o ma ke alues o wine a ibu es is unde aken h ough he es ima ion o a
hedonic p ice unc ion.
The hedonic (hedonic om he G eek wo d o pleasu e o sa is ac ion) p ice me hod is a alid app oach o s udy
he p ice-a ibu es ela ionship o a p oduc . The wine consume aces di icul choices due o he ac ha a po
wine bo le embodies a bundle o cha ac e is ics ha di e en consume s may iden i y and alue di e en ly.
The hedonic me hod amoun s o a eg ession analysis o he p ice on he cha ac e is ics o he p oduc . Basically,
any a iable which in luences consume s' u ili ies o p oduce s' cos is a candida e o inclusion in he eg ession
analysis. The implici p ice o a cha ac e is ic is de ined as he de i a i e o he p ice wi h espec o he p oduc
a ibu es. Despi e o he in luences om se e al au ho s, oday is accep able o all ha he esponsible om
Hedonic P ice Theo y is She win Rosen in his 1974 pape Hedonic P ices and Implici Ma ke s: P oduc
Di e en ia ion in Pu e Compe i ion.
Rosen (1974) has shown unde which ma ke condi ions he implici p ice can be in e p e ed as he alue
consume s place on addi ional uni o he cha ac e is ic. I he es ima ed implici p ice u ns ou o be no
signi ican ly di e en om ze o, hen consume s do no alue he cha ac e is ic, o he cha ac e is ic is no
conside ed impo an o ele an in connec ion wi h he p oduc .
The pape is s uc u ed as ollow. We i s e iew p e ious s udies ha adop a hedonic app oach ega ding wine
indus y. Hedonic p ice heo y and he app op ia eness o he app oach o Po wine a e hen desc ibed. Nex he
da a se employed and he es ima ed unc ion a e ou lined. This is ollowed by a discussion o he esul s and he
implica ions o he analysis.
2. LITERATURE REVIEW
Wine accoun ing li e a u e
We do no in end o do an in ensi e li e a u e e iew ega ding wine accoun ing, o his see Viana and
Rod igues (2004). In his pape , we will b ie ly p esen he majo accoun ing p oblems s a ed in he li e a u e:
ime ela ed p oblems and ax ela ed p oblems.
Time ela ed p oblems
This accoun ing p oblems a i e om he long ma u a ion ime and s ocking p ocess ela ed o he p emium o
high quali y wines. The ime needed by he p emium wine winemaking p ocess is in con lic wi h he adi ional
ACCOUNTING TRENDS
639
accoun ing amewo k, in conc e e, wi h he his o ical cos measu emen ule. The ime lag be ween cos and
income ecogni ion p oduces an impo an accoun ing p oblem. The way ha his long-li ed wine is ecognized
is also an impo an accoun ing wine issue. Rod iguez and Ocejo (2002), Juchau (1996), (AECA, 1999), Ama e
al. (1998), Ama and Blake (1995) pape s s a ed hese wine accoun ing p oblems.
Tax ela ed p oblems
The majo impac comes ou om he ac ha he p oduce o long cella wine be ax penalized - due o he
in la ion accoun ing ea men and he his o ical cos - esul ing in a non-equi y axa ion compa ing low quali y
o high quali y wine companies. Bu he e a e o he accoun ing issues ega ding ax: he esidual alue, a e
dep ecia ion, o a wine plan a ion is nega i e; non- ealis ic and non-economic dep ecia ion a e o some angible
ixed asse s like wood ba els. Rod iguez and Ocejo (2002), Ama e al. (1998), Juchau (1996) iden i y hese
p oblems.
Wine Hedonic li e a u e
The hedonic p ice me hod has been applied mainly o du able goods. The housing ma ke , he compu e indus y,
he au omobile indus y a e good examples o his du able he e ogenei y goods.
The e a e wo di e ences ega ding Po wine: one, Po wine is a ood p oduc and wo, no all ele an
cha ac e is ics a e easy iden i ied by consume s. Ne e heless, wine hedonic p ice unc ion had been es ima ed
in las decade.
The e is a ine li e a u e e iew, al hough no e y ecen , in Unwin (1999) pape . This pape p o ides a c i ique
o he use o hedonic p ice echniques o unde s anding ma ke p ices o wines. Acco ding wi h Unwin, he
applica ion o hedonic p ice indexes o he wine ma ke is inapp op ia e because o he da a (nei he su icien ly
obus no consis en ), he explana o y a iables (closely in e dependen ), he lack o pu e compe i ion
(necessa y o Rosen) and insu icien knowledge abou consume s' de ini ion o quali y.
This c i ique does a ec ou esea ch because he empi ical model used he e is used in ou con ex aim and
goals. We a e no ying o s udy consume beha iou 349. We do claim o y o expose he in insic
cha ac e is ics o a bo le o wine ha indeed poin o i s p ice, and we a emp o quan i y ha impac .
S ill, pa o he deba e abou hedonic p icing is he ela ion be ween subjec i e judgmen s o wine i sel and
objec i e cha ac e is ics. Ou li e a u e e iew is s uc u ed by he na u e o he explana o y a iables in he
hedonic unc ion: objec i e, subjec i e and bo h.
Objec i e explana o y a iables
Blai and Bu ley (1998) in es iga ed he p oposi ion ha a ai deg ee o he p ice a ia ion se e al able ed
Aus alian wines can be explained by he objec i e such g ape a ie y, he egion o he g apes and he wine's
in age. All h ee we e s a is ically ele an .
Wood and Ande son (2002) es ima ed a hedonic model o explain he a ia ion in he auc ion p ice be ween
di e en in ages o pa icula wine, using has explana o y a iables he wea he , ineya d managemen
echniques, winemaking echniques and he age oh he wine. They ound a s a is ically signi ican ela ionship
be ween age, wea he and echnological change and he seconda y ma ke p ice o some Aus alian icon wines.
349 The pape is a pa o one's au ho Ph.D. esea ch in Financial Accoun ing. The main goal o he Ph.D. esea ch is o de elop a Po Wine
Index, wi h quali y adjus men , in o de o e alua e he long- e m In en o ies.
CITIES IN COMPETITION
640
Ribei o and San os (2003) showed ha some egions o o igin ha e a signi ican impac on p ice. O he wine
a ibu es such as colou ( ed), age (mo e yea s) and special a ibu es ( ese e) all ha e posi i e e ec s on p ice.
Subjec i e explana o y a iables
Landon and Smi h (1997) examined he impac o quali y and epu a ion on consume decisions ega ding
Bo deaux wine. Thei model o consume decision making inco po a ed in o ma ion on epu a ion and collec i e
epu a ion and ejec ed cu en quali y. Landon and Smi h (1998) es ima ed he impac on p ice o cu en
p oduc quali y and epu a ion using da a om he ma ke o Bo deaux wine. The empi ical indings show ha
he p ice p emium associa ed wi h a be e epu a ion a exceeds ha associa ed wi h imp o emen s in cu en
quali y.
Objec i e and subjec i e explana o y a iables
Al hough no di ec ly using a hedonic me hod, Ashen el e e al (1995) a emp ed o explain empi ically he
a ia ion in p ice be ween di e en in ages. They s udied he ela ionship be ween he p ice o some Bo deaux
wine and hei age o he wea he du ing he g ape g owing season. They ound ha he wea he is a undamen al
and easily measu ed de e minan o he quali y o he ma u e wines.
Oczkowski (1994) es ima ed a hedonic p ice unc ion o Aus alian wine ha included bo h objec i e
cha ac e is ics like g ape a ie y, egion, in age o ha es and size o p oduce and also subjec i e a ibu es
like o e all quali y and cella ing po en ial. He ound ha quali y, cella ing po en ial, g ape a ie y, g ape egion,
g ape in age and p oduce size is s a is ically impo an in explaining p ice de ia ions om a e age p ice.
Comb is e al. (1997) es ima ed a hedonic p ice unc ion o Bo deaux wine ha included bo h objec i e
cha ac e is ics appea ing in he bo le label and also senso y cha ac e is ics esul ed om ol ac o y and gus a o y
examina ions. Thei es ima ion shows ha he Bo deaux ma ke p ice is essen ially de e mined by he objec i e
cha ac e is ics. Comb is e al. (2000) made he same s udy o Bu gundy wine and compa ed he esul s: hey
ob ained he same esul s ha con i m ha objec i e cha ac e is ics a e he mos signi ican ac o s and ha e a
s ong e ec on p ice.
Wade (1999) es ima ed wo hedonic p ice equa ions ega ding Aus alian able wine. In he i s equa ion he
consume s knew a subs an ial amoun o in o ma ion including cu en quali y, g ape a ie y, and egion and
cella ing po en ial o he wine. In he second equa ion, only epu a ion de e mines he p ice ha consume is
willing o pay o he wine. The esul s ound ha epu a ion has a la ge impac on he p ice o he wine. Angulo
e al. (2000) es ima ed a hedonic p ice unc ion o Spanish ed quali y wine. In hei model, he independen s
a iables we e egion o p oduc ion, he yea o he in age, g ape a ie y, alcoholic con en and expe quali y
a ings. The esul s indica ed ha he egion o p oduc ion and he yea o in age we e he main ma ke p ice
de e minan s and ha g ape a ie y and alcoholic con en we e no signi ican co ela ed wi h ed wine p ices.
Schamel (2000) es ima ed a hedonic p icing model using U.S da a o p emium wines om se e al coun ies
including U.S, Aus alia. Schamel ocus on he alue o wine quali y o consume s and how quali y indica o s
in luence hei willingness o pay o p emium wines. Wine a ibu es included a ie y, senso y quali y a ings,
and indi idual and collec i e epu a ion indica o s. Regional o igin was used as a collec i e epu a ion indica o .
Indi idual epu a ion indica o included "cella selec ion" (po en ial o imp o e wi h age), " ecommended" (wine
especially no ewo hy) and "bes buy" (ou s anding alues). Highly signi ican es ima o s o senso y wine
quali y, indi idual and collec i e epu a ion indica o s explained p ice di e ences o p emium wines.
ACCOUNTING TRENDS
641
Foga y (2002) s udied he Aus alian wine e ail ma ke and conside ed wo pe spec i es: he consume s' and
he p oduce 's. F om he p oduce 's pe spec i e he age o wine a ime o elease, egion and a ie y we e
candida es o inclusion wi hin he se o a ibu es likely o in luence p ice. F om he iewpoin o he consume
he age o wine a ime o elease, egion, a ie y, in es men quali y and quali y guides we e likely o be ac o s
impo an in aluing wine. The empi ical esul s ounded sugges ha objec i e cha ac e is ics and long- e m
quali y a ings explain p ice and annual quali y a ings do no in luence p ice. Noe (2002) es ima ed a hedonic
p ice unc ion o he Bulga ian wine ma ke . He de eloped a hedonic p ice model o es ima ion o wine
quali y, egional and a ie al epu a ion e ec s in Bulga ian wine ma ke . He showed ha wine quali y has
signi ican and posi i e impac on wine p ices.
Ali and Nauges (2003) s udied he ela i e impac o epu a ion and quali y on he p ice o Bo deaux wine. The
pape made a compa ison be ween en p imeu and bo led wine. In hei pape , epu a ion is de ined h ough i s
appella ion and i s ank. Thei wo k was pionee on en p imeu p ice and con i med se e al o he p e ious
s udies ha epu a ion is a majo de e minan o he p ice o bo led wine and ha p esen quali y as measu ed by
expe s' a ings has a much smalle impac . They showed ha he ank o he Bo deaux wine (da es back o 1855)
was s ill highly alue by consume s. Bomb un and Summe (2003) ocused on how g ape cha ac e is ics a ec
wine p ices. The explana o y a iable we e as ing sco e, age a elease, label designa ion, g ape a ie y, g ape
in age and g ape appella ions. Region o o igin, g ape cha ac e is ics and in age a e he key a ibu es di ec ly
ela ed o g apes ha ha e a signi ican impac on wine p ices.
3. RESEARCH METHOD AND DATA
Ou Model
Rosen (1974) sa es ha a pa icula class o commodi ies can be desc ibed by a ec o o n objec i ely measu ed
cha ac e is ics as
(1) ),,.........,( 21 n
zzzz =
whe e zi is he amoun o he i h cha ac e is ic con ained in each good.
In ou pape , he class o commodi ies e e s o he se o special ca ego ies unde conside a ion o a pa icula
Po wine. The n cha ac e is ics, which di e be ween he special ca ego ies, include he age o he wine a he
ime o sale, Po Wine ypes, b and ype, cella ing po en ial and company epu a ion. The ma ke p ice o a
pa icula wine e lec s he "p ice" o each o he cha ac e is ics embodied in hose special ca ego ies. Tha is,
(2) ),........,,()( 21 n
zzzpzp =
whe e p is an inc easing unc ion o all he cha ac e is ics.
The aim in his pape is o es ima e a Po wine hedonic p ice unc ion in o de o ela e he p ice o a Po wine
bo le o i s quali y, objec i ely measu ed by age, Po ype, b and ype, cella ing po en ial and companies’
epu a ion.

CITIES IN COMPETITION
642
Explana o y a iables
Age o wine a ime o sale
Age is ela ed o p ice because o he sca ci y. As a wine ages, mo e o he gi en po wine is consumed so ha
sca ci y, and hence p ice, inc eases. Age is also ela ed o p ice because o he wine's abili y o de elop and
imp o e wi h age. Po wine, wi hin special ca ego ies, is cha ac e ized by a high con en o annins in hei
you h, making hem no so pleasan o ea ly d inking. The quali y g adually imp o es wi h he dec ease o he
annins. A he same ime, he longe a i m s o es wine in ba els, o cella s i , he g ea e he cos . So he age o
he wine a he ime o elease is likely o make a posi i e con ibu ion owa d he cos o p oduc ion and
he e o e a posi i e con ibu ion owa d he p ice o sale.
In he po wine indus y, due o p oduc speci ica ion, he companies elease se e al wines each yea , all o
di e en age. Fo example, a company may elease se e al Vin age po , om he same b and, om se e al
in ages. So, in he 2002, he buye expec s a bo le o an olde Vin age po o be highe han ano he Vin age
po (wi h he same in age a ing).
The Po Wine indus y is much egula ed so each ype o Po is ca e ully egula ed in e ms o winemaking
p ocess. Fo each wine he e is a minimum age o ma u ing be o e bo ling and o be a ailable o sale. In ou
s udy, we calcula e he age by h ee p ocesses: i s , when a wine has in he label he wine in age (like he
Vin age, La e Bo led Vin age o Colhei a – o Da ed po ) he age was calcula ed by he di e ence be ween he
yea o he sale and he yea o he in age; second, when he wine is an Indica ing o Age Po (10, 20, 30 and
40 yea s old) we adop ed ha indica ion and hi d, he age o he o he ypes is calcula ed h ough he minimum
ma u ing age and winemaking p ocess p ac ices.
Po wine ypes
The e a e, acco ding o he way hey a e aged, wo majo Po Wine ca ego ies: Ruby s yle, Tawny s yle. The e
is also he Whi e po . The Ruby s yle is he ype o wine ha can be ound in special ca ego ies like, in
ascending o de o quali y: Ruby, Rese e, La e Bo led Vin age (LBV) and Vin age. The ines ca ego y wines,
especially Vin age, ollowed by LBV, a e good o s o ing as hey age well in bo le. The Vin age Po is a wine
o excep ional quali y made in a single yea . I mus be bo led be ween he second and hi d yea a e he
ha es ; i is deep pu ple in colou and ull-bodied. Due o i s quali y, he is app o ed as a Vin age (when he
wine is wo yea s old). As i ages in bo le i becomes smoo h and elegan , and i g adually loses i s ini ial
as ingency and acqui es a balanced, complex and highly dis inc i e bouque .
The La e Bo led Vin age (L.B.V.) is a wine o a supe io quali y om a single yea ha is bo led be ween he
ou h and six h yea a e i was made. Al hough equally well balanced, his Po ype, is usually less as ingen
and less ull bodied ha a Vin age om he same yea . Rese e Po Wines a e e y good quali y wines ob ained
by blending wines wi h di e en ages. In his ca ego y, he mos no able a e he Tawny Rese e Po s. They a e
medium awny in colou esul ing om hei compulso y minimum 7 yea s' aging in wood. In u n, Ruby
Rese e Po s, a blend o younge wines ha gi es hem hei uby colou and in ense and ui y bouque , a e
ull-bodied and annic in he mou h, bu less so han Vin age and LBV. The awny s yle wine is ob ained om
lo s o di e en wines ha ha e aged o di e en leng hs o ime in casks. The p esen ca ego ies in his s yle
a e: Tawny, Tawny Rese e, Tawny wi h an Indica ion o Age (10, 20, 30, 40 yea s old o mo e han 40 yea s)
and Colhei a (Da ed po ). These wines a e eady o d ink when hey a e bo led. Colhei a o Da ed Po a e
ACCOUNTING TRENDS
643
supe io quali y wines om a single yea ha a e aged in wood du ing a ying pe iods o ime, bu ne e less
han 7 yea s, be o e bo ling.
Po wi h an Indica ion o Age: 10 yea s old; 20 yea s old; 30 yea s old; 40 yea s old a e supe io quali y Po s
a e ob ained by blending wines om di e en yea s in o de o ob ain complemen a y o ganolep ic
cha ac e is ics. Aged in wood o a ying pe iods o ime, he age ha is indica ed on he label co esponds o he
app oxima e a e age age o he di e en wines in he blend. F om ha age, one ob ains an indica ion o he
cha ac e is ics he wine acqui ed as i aged in wood.
Whi e Po a ies in s yle acco ding o whe he i has aged o a sho e o longe pe iod o ime, and di e en
deg ees o swee ness acco ding o he manne by which i is made.
Type o B and
All Po wine is sold unde a b and. This b and can be a Shippe s Own B and – a b and own by a Po wine
company – o a Buye s Own B and. We hink ha he owne ship o he b and is an impo an explana o y
a iable.
Al hough a bo le we e sold unde a buye s own b and, like a supe ma ke b and, he name o he Po i m is
always disclosu e in he bo le label.
Company Repu a ion
On Landon and Smi h (1997) pape ega ding Bo deaux wine, he au ho assumed a epu a ion model based on
he ac ha in o ma ion on cu en quali y is no a ailable o is oo cos ly o ob ain and, as a esul , he p ice o a
i m's p oduc depends on i s epu a ion o quali y, no i s cu en quali y. Thei esul s indica ed ha he e ec
on p ice o sho - e m changes in quali y is ela i ely small. Acco ding wi h he au ho s " his implies ha
consume s p ima ily base hei pu chase decisions on pe sis en , a he han sho un, mo emen s on quali y"
(Landon and Smi h, pp. 313).
E e y Vin age Po mus be au ho ized and disclosu e by he Po and Dou o Wines Ins i u e. The Po Wine
Ins i u e has, by companies and yea s, all Vin age decla a ion and we use his in o ma ion o build a p oxy o
disclosu e epu a ion. We g ade each wine as "excellen ", " e y good", "good" and " egula " epu a ion based on
i m's decla a ion and indus y egula o au ho iza ion o "Vin age". The Dou o and Po wine Ins i u e
disclosu e all "Vin age" decla a ion by i m since 1900. Based on ha da a we buil ou p oxy o Repu a ion.
In o de o build ha p oxy, we s a o coun he numbe o Vin age decla a ion o each i m since 1900 o
2000. Then we ound he i y, se en y- i e and nine y and nine y- i e pe cen ile o ha collec ion o da a –
numbe o in age decla a ions o each i m om 1900 o 2000. A e ha we g ade as "excellen " i ms whose
numbe o decla a ion a e highe han 95 pe cen ile; g ade as " e y good" i ms ha a e be ween 90 o 95
pe cen ile; g ade as "good" i ms ha a e be ween 90 o 75 pe cen ile and as " egula " he o he s – see Appendix
II.
Table 1 - Numbe o Vin age decla a ion om 1900-2000
A e age 50 Pe cen ile 75 Pe cen ile 90 Pe cen ile 95 Pe cen ile
14.73 11 23 38.6 41.9
CITIES IN COMPETITION
644
Table 2 – Fi m epu a ion classi ica ion
Numbe o Fi ms
Excellen 4
Ve y Good 4
Good 12
Regula 54
We could es ablish, based on p e ious li e a u e and heo e ical backg ound, he "expe opinion" as an
explana o y a iable bu he lack o da a - he e is no Po Wine edi ion ega ding in e na ional magazine o
expe s - s opped us.
Cella ing po en ial
As we s a ed be o e, he e a e se e al "Po Wines" - he Po ypes. The indus y egula o (IVDP) s a es on his
web si e ha "The ines ca ego y wines, especially Vin age, ollowed by LBV, a e good o s o ing as hey age
well in bo le"350. A e consul ing se e al Po wine expe om he companies and om he indus y egula o ,
we buil his a iable by “me ging” hei opinions ega ding he cella ing po en ial o he po wines. We ook
his p oceeding in o de o educe he subjec i i y o he a iable.
So, we classi ied he wines as:
• "high cella ing po en ial", i we a e dealing wi h young Vin ages (less han 15 yea s old);
• "medium cella ing po en ial", i we a e dealing wi h young L.B.V. (less han 5 yea s old), young Da ed
Po (less han 10 yea s old) o C us ed ype
• "low cella ing po en ial" i we a e dealing wi h Vin age Cha ac e , Ruby Rese e, Tawny Rese e,
Indica ion o Age, old Da ed Po (mo e han 10 yea s old) and 5 o 10 yea s old L.B.V.;
• “null cella ing po en ial” ega ding he Ruby, Tawny, Whi e, old L.B.V. (mo e han 10 yea s old).
Vin age o ha es
This explana o y a iable can only be used when s udying he speci ic ypes o po : La e Bo led Vin age,
Vin age o Colhei a Po ypes. The eason is simple: he o he Po ypes a e blend wines so hey a e no di ec ly
connec ed wi h a in age. As any o he ui , he quali y o he g apes depends la gely o he wea he condi ions
du ing he g owing season ha p oduced he ui . Sma (2001) a gues ha all clima e pa ame e s can be
impo an in in luencing g ape quali y. So, as we ea ly s a ed, he po en ial quali y o a wine is a esul o he
quali y o he inpu s (mainly g apes) and o he winemaking echniques, and he in age351 (o ha es ) o he
wine cap u es he in luence o he wea he condi ions and i s in luence on he quali y o he g apes. In his
explana o y a iable, he s udy ies o cap u e he in age e ec , associa ed wi h wea he and g ape condi ions,
and no wi h ime.
The da a
All he da a was gi en o us by he IVDP bu , as classi ied da a, he access was gi en due o a clea au ho iza ion
om each i m. E e y Po wine i m and e e y ime ha makes a wine ansac ion (e en a single bo le) mus
ill a documen (RCDO) which has se e al in o ma ion: name o he i m, des ina ion coun y, b and code,
350 h p://www.i p.p /uk/TiposVinho/ - 04/06/2004
351 The in age is di e en om he Po 's special ca ego y named "Vin age".
ACCOUNTING TRENDS
645
quan i y and alue o he ansac ion/sell. B and code has i sel se e al o he s in o ma ion like he ype o Po
and i s in age. This documen is s o ed in IVDP. The ansac ions ega d sales o wholesale s, e aile s o o
inal consume s. The da a collec ed e e s o sales made o he majo Po wine wo ld ma ke s352 and o he
bigges Po Wine Companies du ing 2002. Ou o he o iginal sample o 14 209 sales, 14 122 we e selec ed. The
obse a ions we e elimina ed due o he lack o one o mo e in o ma ion like p ice. This si ua ion occu s due o
mis iled o he RCDO.
P ice Da a
The P ice da a was ound by di iding he o al amoun o i wine by he o al amoun o li e o ha i wine. The
p ice collec ed is he p ice o he ansac ion, he ma ke p ice; so i is no he o e p ice o he demand p ice.
Es ima ion me hodology
In his s udy, he unc ional o m used is he log-linea and we speci y he ollowing hedonic p ice unc ion:
Ln P
i = F(Agei, B andi, Po S yi, Fi Repi, CelPo ei)
Ou aim is o es whe he he e ec o Age on Po wine ma ke p ice a e con olling o some o he main
e ec ha de e minan p ice. The con ol a iable included in he model, like ype o Po (Po S y), ype o
b and (B and), i m epu a ion (Fi mRep) and cella ing po en ial (CelPo ). The Coun y a iable is also a con ol
one and was in oduced in o de o cap u e he ma ke e ec : i he po wine de e minan s emain he same e en
when we conside ed each coun y indi idually.
Ln P
i = F(Agei, B andi, Po S yi, Fi Repi, CelPo ei, Coun yi)
Whe e Pi = he sale p ice o he wine i pe li e, Agei = he age a he ime o sale o he wine i, B andi = he ype
o b and o he wine i, Po S yi = he S yle o he wine i, Fi Repi = he epu a ion o he i m ha sales he wine i,
CelPo ei = he cella ing po en ial o he winei , Coun yi = he coun y o whe e he winei is shipped.
Mo e speci ically, he explana o y a iables a e p esen in Appendix 1.
4. ESTIMATION RESULTS
Bo h Linea and Log-linea (only in he dependen a iable) o m indica ed signi ican deg ees o
he e oscedas ici y and was di icul o iden i y he speci ic o m o he e ocedas ici y. So, we used Whi e’s (1980)
he e ocedas ici y consis en co a iance ma ix o u he analysis. By doing his, he pa ame e es ima es and
summa y s a is ics a e s ill app op ia ely based on leas squa es es ima es. We adop he mos used unc ion o m
in wine hedonic s udies: he Log-linea model. One hedonic p ice unc ion was es ima ed- Table 6. The
in oduc ion o coun y dummies is jus i ied as a con ol a iable. Some in o ma ion ega ding he ansac ions:
352 Po ugal, F ance, Holland, Ge many, Uni ed Kingdom, Denma k, Belgium, Uni ed Sa es o Ame ica and Canada
CITIES IN COMPETITION
652
(a) Whi e He e oskedas ici y-Consis en S anda d E o s & Co a iance
(b) Exp 1) a (5.0 −












−
∧∧∧ ββ i, Kennedy (1981) and De ick (1984)
* he coe icien associa ed wi h he explana o y a iable is signi ican a 99%
** he coe icien associa ed wi h he explana o y a iable is signi ican a 90%