Nunzia a, Luca; Rocco, Lo enzo
A icle — Published Ve sion
The p o es an e hic and en ep eneu ship: inside he
black box
Small Business Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Nunzia a, Luca; Rocco, Lo enzo (2024) : The p o es an e hic and
en ep eneu ship: inside he black box, Small Business Economics, ISSN 1573-0913, Sp inge US,
New Yo k, NY, Vol. 62, Iss. 4, pp. 1285-1313,
h ps://doi.o g/10.1007/s11187-023-00838-3
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/317720
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 p://c ea i ecommons.o g/licenses/by/4.0/
Small Bus Econ (2024) 62:1285–1313
h ps://doi.o g/10.1007/s11187-023-00838-3
RESEARCH ARTICLE
The p o es an e hic and en ep eneu ship: inside
he black box
Luca Nunzia a ·Lo enzo Rocco
© The Au ho (s) 2023
Published online: 9 Janua y 2024
Abs ac We p o ide one o he i s in es iga ions
o he cul u al and alue dimensions h ough which
P o es an ism a ec s en ep eneu ship by ocusing
on he his o ically p ede e mined eligious mino i ies
esiding in he egions o he o me Holy Roman
Empi e. By exploi ing he mino i ies’ s ong a ach-
men o eligious e hic, we ind ha P o es an ism sig-
ni ican ly a ec s he p obabili y o be an en ep eneu
and ha he e ec is ela i ely la ge when we conside
la ge en e p ises. A o mal media ion analysis sug-
ges s ha among he ich se o indi idual cha ac e is-
ics and alues ypically associa ed wi h en ep eneu -
ship a ailable in ou da a, he dimensions ha media e
he e ec o P o es an ism a e educa ion and a as e
o indi idualism. In pa icula , such media ing ac o s
explain a ound 26% o he o al e ec o P o es an ism
on en ep eneu ship.
Plain English Summa y P o es an ism boos s he
likelihood o en ep eneu ship h ough highe educa-
ion, indi idualism, less ega d o ules, and a zes
o li e. This esea ch e eals ha P o es an belie s
inc ease he chance o becoming an en ep eneu , espe-
cially o la ge businesses. The key ac o s linking
L. Nunzia a (B)·L. Rocco
Uni e si y o Padua, Padua, I aly
e-mail: [email p o ec ed]
L. Rocco
e-mail: [email p o ec ed]
L. Nunzia a ·L. Rocco
IZA, Bonn, Ge many
P o es an ism o en ep eneu ship a e highe educa ion
and a s ong sense o indi idualism, which includes
an a e sion o ollowing ules and humili y and a as e
o an exci ing li e. Toge he hese accoun o a ound
26% o he impac o P o es an ism on en ep eneu -
ship. In essence, cul u e and alues play a key ole in
shaping success ul en ep eneu s. Unde s anding cul-
u al in luences is c ucial o hose seeking o p omo e
new businesses and o hose s udying he de e minan s
o new and success ul en u es.
Keywo ds En ep eneu ship ·Religion ·
P o es an ism ·Ca holicism ·Cul u e ·Value ·
Media ion
JEL Classi ica ion L26 ·Z12 ·Z13
1 In oduc ion
The ela ionship be ween eligion and en ep eneu ship
has a ac ed much in e es since he classic “P o es an
E hics and he Spi i o Capi alism” by Max Webe
(Webe , 1904) whe e P o es an ism was seen as ins u-
men al o he de elopmen o a mode n capi alis ic
economy mo e han Ca holicism was.
P o es an ism and Ca holicism a e bo h Ch is ian
denomina ions, bu hey di e in a leas h ee aspec s,
which a e ela ed o indi idual economic ac ions and
decisions. Fi s , P o es an ism assigns mo e esponsi-
bili y, bu also eedom, o he indi idual in e ms o
123
1286 L. Nunzia a and L. Rocco
a pe sonal in e p e a ion o he sc ip u es and a di ec
connec ion wi h God, while Ca holicism a ibu es o
he cle gy bo h he oles o being he only legi ima e
in e p e e o he Wo d o God and he necessa y medi-
a o be ween God and he people (He zog and Scha ,
1908). Second, P o es an ism has a mo e posi i e a i-
ude owa ds weal h and ma e ial possessions com-
pa ed o Ca holicism. I main ains ha he achie emen
o ma e ial possession, s a us, and wo ldly success a e
signs o sal a ion and o God’s a ou (Maye and
Sha p, 1967). On he con a y, Ca holicism gene ally
gi es li le impo ance o wo ldly success, while humili y
and po e y a e ypically p aised. Fo example, Pope
F ancis de i ed his name om Sain F ancis, who
“lo ed no only he poo bu po e y i sel ” as he
ga eway o sal a ion.1Thi d, P o es an ism conside s
people’s labou as hei own oca ion (Ea on, 2013).
Vice e sa, Ca holicism conside s wo k as mainly ins u-
men al o li e a he han a li e’s goal (Tilghe , 1958).
Gi en hese di e ences, Max Webe hypo hesised
ha P o es an ism should be mo e suppo i e o sel -
employmen and en ep eneu ship han Ca holicism,
since i pushes belie e s o coun mo e on hemsel es,
o alue and seek achie emen and ind in hei wo k
he pu pose o hei li e. These alues suppo bo h he
decision o become an en ep eneu and he likelihood
o succeed and emain so.
The causal chain unde laying Webe ’s hypo hesis
is indi ec : P o es an e hics suppo s he de elopmen
o ce ain e hical alues and pe sonali y ai s, and in
u n, alues and pe sonali y in luence he likelihood
o becoming en ep eneu o op ing o independen
wo k. In o he wo ds, he e ec o P o es an ism is
media ed by he alues and he pe sonali y ai s ha
a e de eloped by P o es an s. In eali y, he media o s
o P o es an ism a e unlikely o be limi ed o alues
and pe sonali y only. Fo ins ance, Becke and Woess-
mann (2009) ha e emphasised he ole o P o es an ism
in os e ing educa ional a ainmen . McCullough and
Wolloughby (2009) poin ed ou ha sel -con ol and
sel - egula ion, which a e in luenced by eligion, a e
associa ed wi h lowe mo ali y, sa e heal h beha io s,
and less c iminal ac ions, all ou comes which migh
hemsel es in luence he likelihood o becoming and
h i ing as an en ep eneu . I is li le con o e sial ha
a deeply in e nalised eligion may se e as a powe ul
1Pope F ancis’ add ess o he e en “ he Economy o
F ancesco,” Assisi Sep 24 h 2022.
social o ce ha in luences many domains o an indi-
idual li e. Howe e , his ac makes i di icul o d aw
a comple e map o he pa hways om P o es an ism o
en ep eneu ship, and mo e gene ally o any economic
ou come.
In his pape , we in es iga e a po ion o his map,
he limi s o which a e dic a ed by da a a ailabili y.
Mo e speci ically, we in es iga e (i) he di e en ial
e ec o P o es an ism (compa ed o Ca holicism) on
a se o alues and pe sonali y ai s, and on educa-
ional a ainmen ; and (ii) he e ec o hese media o s
on en ep eneu ship. Gi en he o e all e ec o P o es-
an ism on en ep eneu ship, ha we also es ima e, we
can assess wha p opo ion o his e ec is accoun ed
o by his se o media o s.
The li e a u e on he e ec s o eligion on alues
and pe sonali y is a he small and mos ly examines
he e ec o eligiosi y in gene al a he han ha o
speci ic eligious denomina ions. We add o his li e -
a u e by compa ing P o es an ism and Ca holicism as
o hei in luences on, i s , a ba e y o alues and
pe sonali y ai s ha he li e a u e sugges s a e plausi-
bly ela ed o en ep eneu ship, and second, on college
educa ion, adap ing (Becke and Woessmann, 2009)’s
analysis o ou con ex and empi ical s a egy. To ou
knowledge, we p o ide he i s ull media ion analysis
o he impac o P o es an ism on en ep eneu ship.
The business and psychological li e a u es ha e
analysed he psychology o en ep eneu s and he
impo ance o indi idual a i udes in pushing owa ds
en ep eneu ship. In pa icula , sel -e icacy, achie e-
men mo i a ion, p oac i e pe sonali y, inno a i eness,
and au onomy a e he pe sonali y ai s ha ha e been
ypically ound associa ed wi h bo h business c ea ion
and business pe o mance. We con ibu e o his li -
e a u e by p o iding one o he i s a emp a es i-
ma ing he associa ions be ween alues, pe sonali y
ai s, and educa ional a ainmen on he one hand and
en ep eneu ship on he o he , by con olling o ele-
an con ounde s, abo e all whe he he indi idual has
an en ep eneu ial amily backg ound (a a iable a ely
a ailable in he da a).
To deal wi h endogenei y, we adop he s a egy
p oposed in Nunzia a and Rocco (2016,2018), which
consis s in compa ing P o es an and Ca holic mino i-
ies in he as a ea o Cen al Eu ope unde he ule
o he Holy Roman Empi e, whe e he geog aphical
dis ibu ion o eligion is he esul o he cuius egio
eius eligio p inciple es ablished a he end o he eli-
123
The p o es an e hic and en ep eneu ship:... 1287
gious wa s o he se en een h cen u y. The a ionale
o his app oach is ha eligious mino i ies a e mo e
a ached o eligious p inciples and alues, ha in e -
gene a ional ansmission is s onge among hem, and
ha he geog aphical dis ibu ion o eligious mino i-
ies is his o ically de e mined and no he esul o sub-
sequen mig a ion lows.
We analyze indi idual-le el da a om he Eu opean
Social Su ey (ESS hence o h) collec ed in he a ea
o he o me Holy Roman Empi e and we expand ou
o iginal da ase by including mo e ecen da a, ha now
spans om 2002 o 2018. The much la ge sample size
allows o es ima e he e ec o P o es an ism on he
p opensi y o sel -employmen and en ep eneu ship
o e al e na i e de ini ions o an en ep eneu . Mo e
speci ically, we a e now able o dis inguish be ween
sel -employed indi iduals wi h no dependen employ-
ees, and hose who employ a small o a la ge num-
be o wo ke s. Ou analysis shows ha no only
P o es an ism is conduc i e o en ep eneu ship, in line
wi h ou p e ious indings in Nunzia a and Rocco
(2016,2018), bu also ha P o es an ism inc eases he
p obabili y o be a success ul en ep eneu , wi h he
e ec being mono onically inc easing he la ge is he
en ep eneu ’s business.
Ou indings e eal ha P o es an ism a o s
en ep eneu ship mos ly h ough wo dis inc channels,
i.e., educa ion a ainmen and a se o indi idualis-
ic pe sonali y ai s. The la e a e ep esen ed by a
lowe p e e ence o ollowing ules and doing wha
one’s old, o being humble and modes and no d aw-
ing a en ion, and a s onge p e e ence o seeking
ad en u es and ha ing an exci ing li e. All hese ac-
o s a e signi ican ly associa ed wi h en ep eneu ship.
The associa ion be ween he app ecia ion o wo ldly
success and a luence and en ep eneu ship has he
expec ed sign, bu i is no s a is ically signi ican in
ou da a.
Ou media ion analysis a ibu es o educa ional
a ainmen a ound 10% o he o e all e ec o P o es-
an ism on en ep eneu ship and an o e all 26% o
he comple e se o alues conside ed in he analy-
sis. These indings indica e ha al hough ou media-
o s accoun o a sizeable p opo ion o he e ec o
P o es an ism, he ela ionship be ween P o es an ism
and en ep eneu ship migh be mo e complex han p e-
iously hypo hesised. Addi ional media o s o he han
educa ion and he alues obse ed in ou da a a e likely
o be a wo k. On he one hand, o he alues, no
a ailable in ou da a, could d i e pa o he e ec o
P o es an ism. On he o he hand, measu emen e o s
o delibe a e mis- epo ing o alues, i he di e en-
ial be ween P o es an s and Ca holics, could lead us
o unde es ima e he con ibu ion o ou media o s.
Bo h conside a ions p o ide some di ec ion o u u e
esea ch.
The pape is o ganised as ollows: Sec .2p o ides
a gene al e iew o he li e a u e; Sec . 3p esen s
he esea ch design; Sec .4desc ibes he da a and
Sec .5p esen s he empi ical indings. Finally, Sec . 6
concludes.
2 Li e a u e e iew
We discuss he li e a u e ollowing he s uc u e o
he media ion analysis, i s conside ing s udies ha
examine he o e all e ec o eligion on en ep eneu -
ship, hen hose ha ocus on he e ec s o eli-
gion on he media o s (educa ion, alues and pe son-
ali y), and inally hose ha ela e he media o s o
en ep eneu ship.
2.1 Religion and en ep eneu ship
Se e al s udies in es iga e he classical Webe ’s hesis,
acco ding o which P o es an alues os e s he de el-
opmen o a capi alis economy, o which en ep eneu -
ship is a c ucial componen , mo e han Ca holicism
does. In pa icula , A unada (2010) a gues ha P o es-
an ism is conduci e o capi alism by p omo ing a
social e hic ha a o s impe sonal ade. Acco ding
o Bas en and Be z (2013) P o es an ism educes p e -
e ences o leisu e, edis ibu ion, and go e nmen
in e en ion in Swi ze land, acili a ing an inc ease in
income pe capi a and income inequali y. Spenkuch
(2017) inds ha P o es an s in con empo a y Ge many
end o wo k longe hou s han Ca holics.2
A ew s udies challenge Webe ’s hypo hesis, by
wa ning abou possible ins i u ions p e-da ing he Re -
o ma ion ha migh con ound he e ec o P o es-
an ism. Ande sen e al. (2017) show ha he P o es an
alues ha Webe iewed as conduc i e o economic
de elopmen , such as he app ecia ion o ha d wo k and
2Di e ences in indi iduals’ alues accoun s o mos o he
e ec , whe eas ins i u ional ac o s o di e ences in human cap-
i al acquisi ion do no seem o play a ele an ole.
123
1288 L. Nunzia a and L. Rocco
h i , had a p e-Re o ma ion o igin due o he local
p esence o Cis e cian monas e ies ha is associa ed
wi h a as e p oduc i i y g ow h om he 13 h cen u y
onwa ds in England, and whose e ec on con empo-
a y alues in Eu ope a e s ill p esen o his day. F i sch
e al. (2020,2021) show ha he e is a legacy o ancien
Roman ule in mode n Ge many. In pa icula , in he
a eas ha once belonged o he Roman Empi e, sou h o
he Limes, we obse e a highe a e o en ep eneu ship
oday. This has o do wi h he legacy o Roman oads,
Roman ma ke s, and mines and is obus o la e majo
ins i u ional changes. F i sch and Wy wich (2018) doc-
umen s he impo ance o ancien uni e si ies o he
p esence o inno a i e en ep eneu s.
The con ibu ions ha speci ically in es iga e he
associa ion be ween eligion and en ep eneu ship a e
he e ogeneous along se e al dimensions and hei
indings a e no always di ec ly compa able. Fi s ,
he ou come o in e es is ei he sel -employmen o
en ep eneu ship. The o me measu e encompasses
en ep eneu ship and i is ypically mo e eadily a ail-
able. I usually includes en ep eneu s, high-skilled lib-
e al p o essions and pe y se ices p o ide s, and sel -
employed indi iduals wi h no dependen employees.
Second, when en ep eneu ship is he ocus o he anal-
ysis, d awing a p ecise de ini ion o he ca ego y is
o en p oblema ic. Fo ins ance, does wo king in a am-
ily business o being sel -employed ou o necessi y
quali y he wo ke as an ac ual en ep eneu ? Thi d,
some pape s analyze only new en u es while o he s
combine bo h new and exis ing en u es. An upda ed
and de ailed accoun o his li e a u e is p o ided by
Block e al. (2020), Dana (2009) and in he ecen pape
by Rie eld and Hoogendoo n (2022).
While p o iding a comple e accoun o he li e a-
u e would exceed he scope o his pape , we discuss
in some de ail a ew wo ks which a e mo e ele an
o ou analysis. Zelekha e al. (2014) pe o m a la ge
c oss-coun y analysis using an inno a i e indica o o
en ep eneu ship de i ed om Linkedin. They es ima e
he associa ion be ween he sha e o en ep eneu s and
(i) he sha es o inhabi an s ollowing each eligion,
(ii) a se o dummies cap u ing he majo i y eligion a
he coun y le el, o (iii) bo h, con olling o a num-
be o coun y cha ac e is ics. Resul s indica e ha he
Jewish eligion has he highes associa ion wi h he
sha e o en ep eneu s, ollowed by Hindu and P o es-
an ism. Ca holicism and Islam a e he leas associ-
a ed. An impo an ques ion is whe he eligion should
be conside ed as an indi idual-le el a iable ha only
in luences he beha iou o eligious indi iduals ( he
so-called mic o app oach), o whe he eligious al-
ues “ loa in he ai ” and in luence all inhabi an s o a
gi en egion ( his is he mac o app oach, which con-
side s eligion as a local-le el a iable). Zelekha e al.
(2014)’s indings suppo he iew ha a coun y’s main
eligion p o ides a con ex ual e ec , i.e., a mac o-le el
ea u e ha in luences he beha iou o bo h eligious
and non- eligious people li ing in he coun y.
Se e al pape s es he hypo hesis o a s onge i
be ween he P o es an e hic (compa ed o he Ca holic
e hic) and en ep eneu ship. Ca oll and Mosakowski
(1987) exploi su ey da a om Wes Ge many and
ind ha P o es an s a e mo e likely o mo e o sel -
employmen (including amily business). Simila ind-
ings a e ob ained by Minns and Rizo (2005) o
Canada and Henley (2017) using a c oss-sec ion o
coun ies pa icipa ing o he Global En ep eneu ship
Moni o . Nunzia a and Rocco (2016,2018) ind ha
mino i y P o es an s a e mo e likely han mino i y
Ca holics o be sel -employed, and he e ec is s onge
o high-skilled occupa ions. Wy wich (2018) con i ms
ha P o es an s a e mo e likely o plan o become sel -
employed, and o ac ually en e sel -employmen .
2.2 Religion and alues and pe sonali y ai s
A ew pape s in es iga e he mechanisms behind he
link be ween eligion and sel -employmen . Recen ly,
Rie eld and Hoogendoo n (2022) analyze he medi-
a ing ole o alues.3They exploi ESS da a be ween
2002 and 2016 and compa e, i s , he alues o eli-
gious and non- eligious people and, second, he al-
ues o en ep eneu s and dependen wo ke s. They
adop Schwa z (1992,2012)’s model o alues and
ind ha indi iduals who belong o any eligion p i-
3We quo e Rie eld and Hoogendoo n (2022)’s de ini ion o
alues: “Values a e he c i e ia o b oad li e goals guiding an
indi idual’s judgmen s, ac ions, and beha io s […]. They en ail
concep ions o he desi able, and, as such, hey cons i u e human
beha io […]. Values pe ain o desi able end s a es, guide beha -
io , anscend speci ic si ua ions and-impo an ly-can be o de ed
by ela i e impo ance […]. Values a e co e elemen s o a pe -
son’s sense o sel , and indi iduals y o a oid beha io s ha
con lic wi h hei alues while unde aking ac ions ha ein-
o ce hei sense o sel . Hence, indi iduals may choose di e -
en ly when con on ed wi h appa en ly simila choices because
o di e en alue p io i ies.”
123
The p o es an e hic and en ep eneu ship:... 1289
o i ise alues ela ed o conse a ion a he han al-
ues ela ed o openness o change, whe eas he oppo-
si e is ue o en ep eneu s. Howe e , bo h eligious
people and en ep eneu s p io i ise alues ela ed o
sel - anscendence (i.e., a p o-social alues) o e hose
ela ed o sel -enhancemen (i.e., indi idualism and
sel - ealisa ion).4
Roccas (2005) documen s ha eligiosi y is ela ed
o p e e ence o conse a i e alues among bo h
P o es an s and Ca holics. Howe e , Ingleha and
Bake (2000) in hei map o 59 coun ies acco ding
o hei alues, inds ha P o es an Eu opean coun-
ies a ibu e much impo ance bo h o sel -exp ession
and o secula - a ional alues, while Ca holic Eu opean
coun ies a ibu e only mode a e impo ance o bo h
ypes o alues.
The gene al li e a u e on he e ec s o eligion on
pe sonali y is a he limi ed. McCullough and Wol-
loughby (2009) e iews he in luence o eligion on
sel -con ol and sel - egula ion. Koenig e al. (2012)
conduc s a comp ehensi e analysis sugges ing ha eli-
gion can p o ide a sense o pu pose, social suppo , and
coping ha bene i o e all psychological unc ioning.
A ecen la ge longi udinal s udy in New Zealand con-
cludes ha con e sion o Ch is iani y appea s o change
a pe son’s cha ac e , speci ically inducing an inc ease
in aspec s o hones y-humili y, p ima ily ela ed o
modes y and a oidance o g eed, conscien iousness,
and neu o icism (S onge e al., 2021). In e es ingly,
Sa oglou (2010) poin s ou ha he e ec could also un
in he opposi e di ec ion, om pe sonali y ai o eli-
giousness, as mo e ag eeable and conscien ious indi-
iduals would be mo e eligious.
2.3 Religion and educa ion
Becke and Woessmann (2009) p opose ha i was no
he P o es an wo k e hic ha p omo ed capi alism, bu
he accumula ion o human capi al gene a ed by he
P o es an emphasis on he impo ance o educa ion
4In Schwa z (1992,2012)’s model he e a e wo dimen-
sions. The i s dimension con as s sel -enhancemen wi h sel -
anscendence and cap u es he con lic be ween a conce n o
he wel a e and in e es s o o he s and he pu sui o one’s own
in e es . The second dimension dis inguishes openness o change
om conse a ion and cap u es he con lic o independence and
eadiness o change wi h a desi e o p ese e he pas and a
esis ance o change.
and he pe sonal eading o he Bible. Thei indings
sugges ha he esul ing highe li e acy o P o es an s
compa ed o Ca holics explains mos o he di e ences
in local economic de elopmen ac oss P ussia in he
nine een h cen u y. In a subsequen pape , Becke and
Woessmann (2010) show ha educa ional a ainmen
was al eady highe in P o es an P ussia in 1816, be o e
he ise o a capi alis and indus ial economy, sugges -
ing ha highe educa ion was no he esul o indus-
ialisa ion bu one o i s de e minan s. Boppa e al.
(2014) documen ha in Swi ze land P o es an ism is
associa ed no only wi h be e eading skills bu also
wi h be e skills in ma hema ics, w i ing, and his o y.
This ac con i ms ha P o es an ism a ou ed edu-
ca ion as a means o human, economic, and social
de elopmen (O ns ein e al., 2016), and sugges s ha
he P o es an wo k e hic has expanded in o an educa-
ion e hic (Boppa e al., 2014), inc easing indi idual
willingness o engage and succeed in educa ion. Bo h
P o es an and Ca holic chu ches encou aged he de el-
opmen o schools and educa ional ins i u ions. P o es-
an ule s did so qui e ex ensi ely a he beginning o
he Re o ma ion in o de o sp ead li e acy (Di ma and
Meisenzahl, 2020, S auss, 1988). In addi ion o his
supply-side dimension ha cha ac e ised bo h denom-
ina ions, P o es an ism had a demand-side dimension
in ha i encou aged each indi idual belie e o acqui e
mo e educa ion, bo h o be able o ead he Bible and
o succeed in he wo ldly li e.5
Such a i udes owa d educa ional a ainmen may be
subjec o in e gene a ional ansmission and pe sis o
his day. Indeed, some less ecen con ibu ions sugges
ha ce e ibus pa ibus, P o es an s a e ypically mo e
inclined o in es in highe educa ion han Ca holics.
Fo example, Fox and Jackson (1973) documen ha in
he 1950s P o es an s in he USA we e mo e likely han
Ca holics o ob ain a college deg ee and o pe sis in
educa ion once en olled. Simila indings o he same
pe iod ha e been epo ed by Mo gan e al. (1962) and
Wa en (1970), among o he s. Howe e , he li e a u e
5In XIX cen u y Swi ze land, Ca holic dis ic s we e cha ac-
e ised by lowe spending in p ima y schooling and a lowe edu-
ca ional pe o mance han P o es an dis ic s, al hough only in
conse a i e a eas (Boppa e al., 2013) In XVI cen u y Ge -
many, he p o ision o mass public educa ion in P o es an ci ies
inc eased he p oduc ion and a ac ion o high-skill human cap-
i al (Di ma and Meisenzahl, 2020). An ex ensi e su ey o he
li e a u e on he causes and consequences o he P o es an e -
o ma ion is p o ided by Becke e al. (2016).
123
1290 L. Nunzia a and L. Rocco
lacks mo e ecen da a analysis on his speci ic poin .
Ou pape con ibu es o illing his gap by p o iding
some new e idence on he e ec o P o es an ism on
indi idual-le el educa ional a ainmen in he o me
Holy Roman Empi e egions o Eu ope.
2.4 The ole o alues and pe sonali y
in he en ep eneu ial spi i
The business and psychological li e a u es ha e anal-
ysed he psychology o en ep eneu s and he impo -
ance o indi idual a i udes in d i ing en ep eneu -
ship. En ep eneu s a e indi iduals who ake isks on
hei own, who wan o be independen and succeed
h ough hei own skills and e o s (Be glann e al.,
2011, Hamil on, 2000), and he e o e end o be asso-
cia ed wi h ce ain indi idual a i udes and alues.
A e he seminal con ibu ion o McClelland
(1961), who pu o wa d he ole o psychology as
an impo an d i e o en ep eneu ship, i is only in
he las wo decades ha a enewed in e es in psycho-
logical ac o s has eme ged. F ese and Gielnik (2014)
e iew he psychological li e a u e and ind ha pe -
sonali y ai s a e he ac o s mo e s ongly associa ed
wi h bo h business c ea ion and business pe o mance.
The pe sonali y ai s ha a e mo e s ongly associ-
a ed wi h en ep eneu ial ou comes a e sel -e icacy,
achie emen mo i a ion, p oac i e pe sonali y, inno a-
i eness, and au onomy. In a mo e ecen e iew, F ese
and Gielnik (2023) con i m he impo ance o pe son-
ali y in explaining en ep eneu ship.
Achie emen mo i a ion eme ges as he s onges
p edic o o business pe o mance among all he ac-
o s conside ed in F ese and Gielnik (2014)’s me a-
analysis, and i is one o he s onges p edic o s o busi-
ness c ea ion. In ui i ely, achie emen -o ien ed indi-
iduals a e a ac ed o independen wo k, end o seek
challenges, se ambi ious goals, ake calcula ed isks,
and pe se e e in he ace o obs acles (Collins e al.,
2004, McClelland, 1961). In e es ingly, he Big Fi e
pe sonali y ai s u n ou o be somewha less co -
ela ed wi h bo h en ep eneu ship and pe o mance
because hey a e complex cons uc s ha combine sub-
dimensions ha a e di e en ly ela ed o en ep eneu -
ship. Fo example, du i ulness and achie emen s i -
ing a e wo sub-dimensions o conscien iousness, he
o me being nega i ely co ela ed wi h en ep eneu -
ship and he la e being posi i ely co ela ed (Rauch
and F ese, 2007).
2.5 Educa ion and en ep eneu ship
The e is some deba e abou he ole o educa ion o
en ep eneu ship. A p io i, i is no clea ha a highe
le el o educa ion inc eases he likelihood o es ablish-
ing a business en u e. Fo ins ance, spino s c ea ed
by o me dependen wo ke s bene i mo e om hei
ounde s’ speci ic know-how han om hei o mal
educa ion. Baumol (2004) ac ually sugges s ha o -
mal educa ion migh hampe c ea i i y and inno a i e
hinking. Se e al s udies show ha e en g adua es in
en ep eneu ship p og ams a e less likely han o he s
o be en ep eneu s (Oos e beek e al., 2010). On he
o he hand, highe educa ion is a necessa y condi ion
o en e ing libe al p o essions, ha a e ypical occupa-
ions o he sel -employed. Van de Sluis e al. (2008)
e iew he li e a u e and p o ide a me a analysis. Thei
esul s indica e ha educa ion is gene ally un ela ed
wi h he p obabili y o en y, bu i is posi i ely co -
ela ed wi h he pe o mance, and hus wi h he p ob-
abili y o su i al and emaining en ep eneu . In e -
es ingly, he majo i y o s udies analyzed by Van de
Sluis e al. (2008) ind a s a is ically signi ican pos-
i i e e ec o pos g adua e s udies on he p obabili y
o en e ing in en ep eneu ship. Ha og e al. (2010)
documen ha echnical and ma hema ical abili ies
ha e highe e u ns among en ep eneu s han among
dependen wo ke s. Mo e ecen ly, Du a and Sobel
(2018) documen ha he e ec o educa ion depends
on he con ex , and sugges ha e ia y educa ion os-
e s en ep eneu ship mos when he le el o inancial
de elopmen is low. Fo highe le els o inancial de el-
opmen , he impac o e ia y educa ion is s ill posi i e
bu much smalle .
2.6 Causali y
F om a me hodological pe spec i e, mos con ibu ions
ind associa ions be ween eligion and a ious ou -
comes. As eligion is deeply oo ed in he cul u e o a
place, he e a e po en ially many con ounding ac o s
a play, and ocusing on associa ions could be mislead-
ing. Rela i ely ew pape s ocus on iden i ying causal
e ec s. To he bes o ou knowledge, only Nunzia a
and Rocco (2016,2018) and Wy wich (2018)in he
123
The p o es an e hic and en ep eneu ship:... 1291
li e a u e on eligion and en ep eneu ship pay a en-
ion o he iden i ica ion o causal e ec s.
Es ima ing he e ec o eligion on any ou come is
challenging because he adop ion and in e nalisa ion o
eligious p inciples is an indi idual choice.6I is he e-
o e possible ha people who a e mo e indi idualis-
ic and who end o seek success and sel - ealisa ion
a e also mo e likely o adhe e o and ollow P o es-
an p inciples because hey a e mo e in line wi h hei
p esupposed alues. I so, eligious a ilia ion would
e lec inna e indi idual cha ac e is ics and could no
play an au onomous ole in shaping en ep eneu -
ship. Mo eo e , pe sonali y could in luence eligios-
i y, bo h ex insically and in insically, i.e., he ex e nal
aspec s o eligious a ilia ion, such as a ending ce -
emonies, and he mo e in ima e aspec s, such as he
in e nalisa ion o eligious p inciples and alues. In ou
analysis, hese conce ns a e pa icula ly p oblema ic
when es ima ing he o e all e ec o P o es an ism on
en ep eneu ship and he e ec o P o es an ism on he
media o s.
In wo ecen con ibu ions (Nunzia a and Rocco,
2016,2018), we p opose a s a egy o alle ia e his
p oblem based on he compa ison o membe s o
P o es an and Ca holic mino i ies in Swi ze land and
in he a ea o he o me Holy Roman Empi e. These
eligious mino i ies a e his o ically de e mined as a
esul o cuius egio eius eligio-like ules adop ed a e
he eligion wa s o he 16 h and 17 h cen u ies. The
compa ison o P o es an and Ca holic mino i ies is,
he e o e, a compa ison o indi iduals who, on a e age,
in e nalise he espec i e eligious e hics s ongly, i e-
spec i e o hei inna e cha ac e is ics. The addi ional
ea u es ha cha ac e ise a mino i y eligion (e.g., di -
e en dep h o social ne wo ks, o possible disc imi-
na ion) a e common o P o es an s and Ca holics and
a e di e enced ou in he compa ison.7
Wy wich (2018) adop s a simila s a egy by com-
pa ing he P o es an and Ca holic mino i ies ha su -
6Eu ope is cha ac e ised by a s ong in e -gene a ional eli-
gious pe sis ence. Religious denomina ions a e acqui ed om
pa en s and indi iduals ypically do no con e om one Ch is-
ian denomina ion o ano he , excep in a e cases (e.g., because
he pa ne belongs o a di e en denomina ion). Howe e , eli-
gious a ilia ion may be endogenous mainly because in some
cases indi iduals abandon hei pa en s’ denomina ion and u n
o secula ism.
7We e e o Sec . 3o his pape and Nunzia a and Rocco (2016,
2018) o a mo e de ailed discussion o he iden i ica ion s a egy.
i ed in eas e n Ge many a e 40 yea s o communis
ule, a he e e o he Ge man euni ica ion.
Rega dless o he empi ical s a egy adop ed, all
hese s udies a e “ educed o ms,” i.e., hey es ima e
he o e all e ec o being P o es an ( e sus Ca holic)
on he likelihood o being an en ep eneu , wi h-
ou es ima ing and es ing he a ious mechanisms
ha lay behind his empi ical ela ionship. In ac ,
a educed- o m model does no in es iga e and dis-
inguish be ween he possible mechanisms bu a he
ea s eligious a ilia ion as a black box.
Recen ly, Rie eld and Hoogendoo n (2022)s a ed
o open he black box by es ima ing he ela ionship
be ween eligious a ilia ion and alues. In his pape ,
we go a s ep u he in his di ec ion. On he one hand,
we exploi he compa ison o mino i ies o es ima e he
causal e ec o P o es an ism on a ba e y o alues
and indi idual cha ac e is ics ha ha e been ela ed o
en ep eneu ship in he li e a u e. On he o he hand,
we conduc a ull media ion analysis o assess how
much o he o e all e ec o P o es an ism is medi-
a ed h ough he se o media o s unde conside a ion.
Ou analysis is one o he i s a emp s o iden i y he
ela ionship be ween en ep eneu ship and he e hical
and alue con en ha di e en ia es P o es an ism om
Ca holicism.
3 Resea ch design
3.1 The e ec o P o es an ism on en ep eneu ship
and pe sonal ai s
Ou objec i e is o es ima e he e ec o P o es an ism
on en ep eneu ship and shedding some ligh on he
cul u al and alue mechanisms h ough which P o es-
an ism a ec s en ep eneu ship. Compa ed o Nun-
zia a and Rocco (2018), we exploi a much la ge
da ase , collec ed be ween 2002 and 2018 om he
Eu opean Social Su ey. As a consequence, we a e
able o iden i y he e ec o P o es an ism on sel -
employmen and en ep eneu ship cha ac e ised by
a ious deg ees o success, i.e., dis inguishing be ween
small and la ge en e p ises. In addi ion, we p o ide
wha o ou knowledge is he i s ull- ledged medi-
a ion analysis designed o e alua e wha is he p o-
po ion o he o e all e ec o P o es an ism which is
channelled ia a se o obse ed media o s.
123
1292 L. Nunzia a and L. Rocco
A majo issue in in es iga ing he implica ions o
he e hical con en o eligions is ha sel -iden i ica ion
wi h a ce ain eligious c eed does no necessa ily imply
he in e nalisa ion o i s e hical p inciples. Mo eo e ,
such in e nalisa ion is usually mo e likely i hese p in-
ciples ma ch wi h indi iduals’ alues and p e e ences.
As a esul , he conclusions de i ed om a simple com-
pa ison o epo ed a ilia ions in indi idual-le el da a
may be misleading.8
To o e come his p oblem, in Nunzia a and Rocco
(2016,2018), we p opose an indi ec measu e o
a achmen o eligious p inciples: he condi ion o
belonging o a eligious mino i y. The a ionale o
his app oach is ha membe s o mino i y eligions a e
mo e e en belie e s han a e membe s o majo i y
eligions. This is because eligion is an impo an ele-
men o people’s iden i y ha mino i ies seek o de end
(Bisin and Ve die , 2000, Bisin and Ve die , 2001, and
Bisin e al., 2004)9and because a mino i y eligion’s
cle gy wo ks ha d o p ese e i s ollowe s om he
cons an p essu e o he su ounding dominan eligion
(S a k e al., 1995, Finke and S a k, 1998, and S a k,
1998). Table A3 in he online appendix o Nunzia a
and Rocco (2016) es ima es and es s he di e en ial
e ec o belonging o a P o es an mino i y ( s no )
and o belonging o a Ca holic Mino i y ( s no ) on
a numbe o ou comes, including eligious pa icipa-
ion, equency o p aying, and a numbe o ai h a i-
cles which a e speci ic o each eligious denomina ion.
The able p o ides sys ema ic e idence ha mino i-
ies do pa icipa e mo e o eligious ac i i ies and ha e
s onge belie s. Fo ins ance, mino i y P o es an s a e
21% mo e likely o epo p aying egula ly. Simila ly,
mino i y Ca holics a e 25% mo e likely o epo p ay-
ing egula ly.
Mo eo e , being bo n in o a gi en eligious mino i y
can be conside ed p ede e mined because he ise and
geog aphic dis ibu ion o mino i ies o bo h denom-
ina ions in he egions o he o me Holy Roman
8A numbe o eligiousness indica o s ha e been sugges ed by
he li e a u e o measu e he in ensi y o he indi idual a achmen
o eligious belie s (McClea y and Ba o, 2006) Some examples
a e he equency o a endance o eligious se ices, weekly
p aye s, and dona ion o money and ime o eligious o ganisa-
ions. Howe e , such indica o s a e likely o be endogenous o
economic beha iou and a i udes.
9Membe s o a mino i y eligion exp ess hei own iden i y by
pa icipa ing mo e o eligious ac i i ies (ex insic eligiosi y)
and by aking eligious alues and p inciples mo e se iously
(in insic eligiosi y).
Empi e (HRE hence o h) ollow he equilib ium ound
a he end o he Religious Wa s o he 16 h and 17 h
cen u ies ha depended mainly on con ingen his o -
ical condi ions (Can oni, 2012, Can oni, 2015). The
esul ing dis ibu ion has pe sis ed o cen u ies so ha
he condi ion o being pa o a mino i y and he co -
esponding deepe in e nalisa ion o eligious p inci-
ples can be conside ed p ede e mined wi h espec o
cu en indi idual labou ma ke choices. I is wo h
no ing ha ou iden i ica ion hinges on he compa i-
son o mino i ies loca ed on ei he side o he Roman
Limes. Hence, ou indings do no e lec he his o i-
cal ole a ou able o en ep eneu ship played by he
ancien Roman domina ion and discussed in F i sch
e al. (2020,2021).
Pe sis ence is he consequence o c oss-gene a ion
ansmission (Guiso e al., 2008). As documen ed by
ISSP da a, we obse e indeed a ce ain pe sis ence
o eligious a ilia ion ac oss gene a ions in Eu ope.
Acco ding o ISSP Religion III da a, 96 (94) pe cen
o esponden s who had wo Ca holic (P o es an ) pa -
en s we e aised Ca holic (P o es an ). O hose aised
Ca holic (P o es an ), 83 (79) pe cen con inue o ol-
low hei denomina ion when hey each adul hood,
while 16 (20) pe cen become a heis s, and only abou
1 (1) pe cen con e o P o es an ism (Ca holicism).10
Doepke and Zilibo i (2005) poin ou ha cul u e is no
simply abso bed passi ely by child en li ing wi h and
obse ing hei pa en s, a he i is delibe a ely augh
by pa en s o hei o sp ings (such as he p e e ence o
delaying ewa ds and pa ience, which is unc ional o
business, o example). In e gene a ional ansmission
also explains why eligious alues a e slow o change,
despi e ecen heological de elopmen s and a ia ions
in he doc ine (such as he Second Va ican Council).
Hence, as eligious a ilia ion is ansmi ed ac oss gen-
e a ion and pe sis s o e ime, also he cul u al and eli-
gious alues a e ansmi ed and end o pe sis .
To es ima e he di e en ial e ec o P o es an e hic
compa ed o Ca holic e hic on sel -employmen and
en ep eneu ship we con as indi iduals belonging o
P o es an and Ca holic mino i ies. We conside a sam-
ple o sel -decla ed P o es an and Ca holic indi iduals
10 See he online appendix o Nunzia a and Rocco (2018) o
u he de ails.
123
The p o es an e hic and en ep eneu ship:... 1299
Table 2 The e ec o P o es an ism on en ep eneu ship
(1) (2) (3) (4) (5) (6)
Sel -employed Supe iso No dep. empl ≥4dep.empl. ≥8 dep. empl. Es ab. size ≥10
Min P o - Min Ca h 0.0203** 0.0146** 0.0099 0.0079** 0.0085** 0.0087***
(0.0100) (0.0059) (0.0065) (0.0035) (0.0035) (0.0031)
Obse a ions 23,436 23,436 23,436 23,436 23,436 23,436
R-squa ed 0.048 0.032 0.022 0.019 0.013 0.011
Demog aphic con ols YES YES YES YES YES YES
Coun y FE YES YES YES YES YES YES
Yea FE YES YES YES YES YES YES
Sample mean 0.121 0.055 0.062 0.023 0.011 0.020
S anda d e o s clus e ed a he egional le el in pa en heses. ***p<0.01, **p<0.05, *p<0.1. The able p esen s he es ima ed di e en ial e ec o being a mino i y P o es an
compa ed o being a mino i y Ca holic (Min P o - Min Ca h) on a dummy a iable equal o 1 i esponden is sel -employed (wi h and wi hou dependen employees,col.1),
en ep eneu esponsible o supe ising o he employees (col. 2), en ep eneu wi h no dependen employees (col. 3), en ep eneu wi h 4 o mo e dependen employees (col. 4),
en ep eneu wi h 8 o mo e dependen employees (col. 5), en ep eneu wi h an es ablishmen size wi h 10 o mo e employees (col. 6). All columns a e es ima ed by a linea
p obabili y model. Indi idual le el con ols include whe he he esponden is a o eign na ional, age, age squa ed, gende and whe he he esponden ’s a he was an en ep eneu
123
1300 L. Nunzia a and L. Rocco
Fig. 4 Ma ginal e ec o
P o es an ism on
en ep eneu ship, by
numbe o en ep eneu ’s
dependen employees
0
50
100
150
200
Pe cen e ec
0 2 4 6 8 10
Minimum numbe o en ep eneu ’s dependen employees
Pe cen e ec +2SE −2SE
5.2 Robus ness checks
In Table 3, we check he obus ness o ou indings
by pe o ming a numbe o sensi i i y checks. Fi s ,
in column 1, we expand ou sample by including non-
eligious indi iduals. Ou esea ch design iden i ies he
e ec o P o es an ism on en ep eneu ship by com-
pa ing indi iduals who sha e a simila cul u al iden i y
ela ed o hei Ch is ian eligion a ilia ion, bu di e
o speci ic elemen s ha a e peculia o, espec i ely,
P o es an ism and Ca holicism. In addi ion, by explic-
i ly compa ing mino i y P o es an s and Ca holics we
make su e ha hey s ongly adhe e o hei denomina-
ion’s e hical p inciples and cul u al alues. Howe e ,
one may a gue ha by doing so, we lea e ou an impo -
an pa o he popula ion, i.e., non- eligious espon-
den s, who in p inciple may be simila o he majo i y
o P o es an s and Ca holics o hei weak in e nal-
isa ion o eligious e hical p inciples. This exclusion
should no ma e much in ou es ima es, since ou
pa ame e o in e es is iden i ied by mino i y Ch is-
ians only. Indeed, we show ha ou es ima ed e ec
is s a is ically signi ican and ema kably s able when
we include non- eligious indi iduals, despi e he sam-
ple mo e han doubling om 23,436 obse a ions o
49,474.
In column 2, we include egion- ixed e ec s ins ead
o coun y- ixed e ec s, o con ol o egion-speci ic
ins ead o coun y-speci ic unobse ables. Again, he
es ima es a e ema kably s able. In column 3, we
include a se o po en ially endogenous con ols o
check whe he ou es ima es a e a ec ed. Ou esea ch
design explici ly lea es ou any po en ial endogenous
a iable ha may gene a e a ypical “bad con ol” bias
(Ang is and Pischke, 2009). Ne e heless, when we
con ol o whe he he esponden is ma ied, whe he
he /his main sou ce o income is inancial, and o wo
social capi al dimensions,20 We ind ha ou poin es i-
ma e o in e es is sligh ly educed, bu he implica ions
o ou model a e unchanged.
Ano he po en ial limi a ion o ou model is ha he
ESS de ines egional indica o s a di e en NUTS le -
els o di e en coun ies. In pa icula , we conside
NUTS 2 egions whe e a ailable, bu o Belgium, Ge -
many, F ance, and Luxembou g only NUTS 1 egions
a e conside ed. This means ha he de ini ion o mino -
i y P o es an s and Ca holics o hese coun ies may be
imp ecise. In he case o Ge many, o example, only 16
la ge egions (i.e., s a es o “Land”) a e obse ed. As a
esul , o example, in Lowe Saxony in he no h-wes
o Ge many, Ca holics end o be clus e ed in an a ea
close o he Du ch bo de , whe e hey a e no necessa -
ily a local mino i y, al hough hey a e a mino i y a he
egional (“Land”) le el. To check whe he and how his
complica ion migh a ec ou esul s, in column 4, we
exclude om ou sample all NUTS 1 egion coun ies,
i.e., Belgium, Ge many, F ance, and Luxembou g. Ou
20 These a e (i) o ed in he las elec ion and (ii) a se o dummies
o whe he “mos people can be us ed o you can’ be oo
ca e ul,” a ca ego ical a iable wi h 10 possible answe s, om
“you can’ be oo ca e ul” o “mos people can be us ed.”
123
The p o es an e hic and en ep eneu ship:... 1301
sample is g ea ly educed, d opping o 11,648 obse a-
ions, bu ou poin es ima e is la ge (3.3% e ec ) and
s ill s a is ically signi ican .
In column 5, we exclude all Eas e n Eu opean
egions ha we e unde a communis egime be o e
he all o he Be lin Wall. This includes Eas Ge -
many, he Czech Republic, Poland and Slo enia. These
a eas expe ienced a s ong impac o communism in
pushing eligion ou o socie y (Nunzia a and To o-
lu i, 2019) and may he e o e be cha ac e ised by a
peculia ela ionship wi h eligious no ms and alues
a he indi idual le el compa ed o Wes e n Eu ope. In
addi ion, he Polish egions included in ou baseline
sample may ha e been la gely a ec ed by pos -WWII
mig a ion pa e ns and, as a esul , eligious mino i ies
may no be as his o ically de e mined as in he es o
ou sample. In his case, ou poin es ima e is sligh ly
educed o 1.6% bu is s ill s a is ically signi ican a
he 10% le el.
Finally, in column 6, we exclude hose coun ies
whe e a chu ch ax is manda o y bu indi iduals ha e
he op ion o no paying i by de egis e ing om hei
chu ch a ilia ion, i.e., Aus ia and Ge many. In p inci-
ple, his should no ma e in ou su ey da a, since he
in o ma ion on eligious a ilia ion is no de i ed om
adminis a i e sou ces. In addi ion, p e ious e idence
om Pew Resea ch Cen e da a shows ha manda-
o y chu ch axes a e no associa ed wi h a decline in
epo ed eligiosi y in Eu ope.21 Ne e heless, we can-
no exclude ha in Aus ia and Ge many, indi iduals
may be less inclined o epo a eligious a ilia ion,
aising issues o selec ion bias. When hese coun ies
a e emo ed om ou sample, ou poin es ima es a e
ac ually la ge (4.2%) and s a is ically signi ican .
5.3 Educa ion, alues, cul u al ai s
and en ep eneu ship
Tu ning o he channels h ough which P o es an ism
a ec s en ep eneu ship, Table 4 epo s a se o associ-
a ions be ween al e na i e de ini ions o en ep eneu ,
and educa ion plus a se o alues and cul u al ai s.
The la e a e selec ed om he ull se o alues lis ed in
Sec .4because hey esul ed signi ican in a eg ession
including all dimensions (no epo ed, a ailable upon
eques ). The included eg esso s a e dummy a iables
21 See h ps://pew s .ch/2J UuVh.
equal o one i he esponden has e ia y educa ion
(column 1) o epo s ha she o he in e nalises a ce -
ain alue a leas o some deg ee. The included alues
indica e whe he o he esponden i is impo an o:
ake own decisions and be ee (column 2); do wha is
old and ollow ules (column 3); hink new ideas and
being c ea i e (column 4); be success ul and ha peo-
ple ecognise achie emen s (column 5); be humble and
modes , do no d aw a en ion (column 6); be ich, ha e
money and expensi e hings (column 7); ha people a e
ea ed equally and ha e equal oppo uni ies (column
8); ha go e nmen is s ong and ensu es sa e y (col-
umn 9), and o seek ad en u es and ha e an exci ing
li e (column 10). Desc ip i e s a is ics abou hese al-
ues, ac oss eligious denomina ions, a e p esen ed in
Table 1.
The dependen a iable in he eg essions is, espec-
i ely, a dummy equal o 1 i he esponden is sel -
employed (column 1), an en ep eneu esponsible o
supe ising and moni o ing o he employees (column
2), an en ep eneu wi h dependen employees (column
3), and an en ep eneu wi h an es ablishmen size wi h
10 o mo e employees (column 4).
We ind ha almos all alues included in he able
ha e a signi ican associa ion wi h en ep eneu ship
ac oss al e na i e de ini ions o en ep eneu s. Mo e-
o e , he sign o he associa ion is consis en wi h wha
he li e a u e sugges s. In pa icula , en ep eneu s a e
mo e likely o be highly educa ed, o alue eedom and
dislike ollowing ules, o be inno a i e, o alue suc-
cess and weal h, o a ac a en ion, o seek an exci ing
li e, and o dislike equali y and a s ong go e nmen .
O e all, en ep eneu s appea o be be e educa ed and
o alue indi idualism, inno a ion, weal h, and success.
5.4 The es ima ed e ec o p o es an ism
on educa ion, alues, and cul u al ai s
Ou nex ques ion is whe he hese alues, which a e
ypically associa ed wi h en ep eneu ship, a e in lu-
enced by he P o es an e hic. In Table 5,weuse he
same s a egy as in Table 2 o es ima e he di e en-
ial e ec o mino i y P o es an ism e sus mino i y
Ca holicism on hese alues. Ou es ima es show ha
among he se o a i udes obse ed, mino i y P o es-
an s a e mo e likely o ha e e ia y educa ion and o
ha e a mo e indi idualis ic pe sonali y ai han mino -
i y Ca holics. The la e is cap u ed by a lowe p e e -
123
1302 L. Nunzia a and L. Rocco
Table 3 Robus ness checks
(1) (2) (3) (4) (5) (6)
Incl. No Rel. Reg. FE End. Con . NUTS 2 only Wes EU No Rel. Tax
MinP-MinC 0.0203** 0.0221** 0.0195* 0.0329** 0.0157* 0.0420**
(0.0092) (0.0099) (0.0098) (0.0132) (0.0092) (0.0192)
Obse a ions 49,474 23,436 23,225 11,648 16,064 10,306
R-squa ed 0.035 0.051 0.053 0.045 0.054 0.055
Demog aphic con ols YES YES YES YES YES YES
Endogenous Con ols NO NO YES NO NO NO
Coun y FE YES NO YES YES YES YES
Region FE NO YES NO NO NO NO
Yea FE YES YES YES YES YES YES
Sample mean 0.1217 0.1206 0.1207 0.1077 0.1266 0.1178
S anda d e o s clus e ed a he egional le el in pa en heses. ***p<0.01, **p<0.05, *p<0.1. The able p esen s he es ima ed di e en ial e ec o being a mino i y P o es an
compa ed o being a mino i y Ca holic (Min P o - Min Ca h) on a dummy a iable equal o 1 i he esponden is sel -employed, unde a numbe o obus ness checks: including
non- eligious indi iduals in he sample (col. 1), using egion ins ead o coun y ixed e ec s (col. 2), including po en ially endogenous con ols (col. 3), excluding coun ies whe e
egions a e de ined a he NUTS 1 le el (Belgium, Ge many, F ance and Luxembou g, col. 4), excluding ex-Communis Eas e n Eu opean coun ies only (col. 5), excluding coun ies
whe e a chu ch ax is manda o y bu indi iduals ha e he op ion o no paying i by de egis e ing om hei chu ch a ilia ion (Aus ia and Ge many, col. 6). All columns a e es ima ed
by a linea p obabili y model. Indi idual le el con ols include whe he he esponden is a o eign na ional, age, age squa ed, gende and whe he he esponden ’s a he was an
en ep eneu
123
The p o es an e hic and en ep eneu ship:... 1303
Table 4 The associa ion be ween e ia y educa ion, cul u al ai s and alues, and en ep eneu ship
(1) (2) (3) (4)
Sel -employed Supe iso Wi h dep. empl Es ab. size ≥10
Te ia y Educa ion 0.035*** 0.034*** 0.029*** 0.011***
(0.008) (0.005) (0.005) (0.003)
F eedom 0.030*** 0.014** 0.011 −0.001
(0.010) (0.006) (0.007) (0.005)
Rules −0.030*** −0.008** −0.006 −0.005**
(0.005) (0.003) (0.004) (0.002)
New ideas 0.036*** 0.017*** 0.019*** 0.007**
(0.007) (0.005) (0.004) (0.003)
Success 0.034*** 0.014*** 0.010*** 0.006**
(0.006) (0.004) (0.003) (0.002)
Humili y −0.038*** −0.015*** −0.012** −0.009**
(0.009) (0.005) (0.005) (0.004)
A luence 0.015*** 0.013*** 0.012*** 0.001
(0.004) (0.003) (0.003) (0.002)
Equali y −0.032*** −0.026*** −0.019** −0.011*
(0.011) (0.009) (0.008) (0.006)
S ong go e nmen −0.031*** −0.005 −0.001 −0.002
(0.009) (0.006) (0.006) (0.005)
Ad en u e 0.017*** 0.011*** 0.012*** 0.005**
(0.004) (0.003) (0.003) (0.002)
Obse a ions 23,436 23,436 23,436 23,436
R-squa ed 0.059 0.040 0.036 0.014
Demog aphic con ols YES YES YES YES
Coun y FE YES YES YES YES
Yea FE YES YES YES YES
S anda d e o s clus e ed a he egional le el in pa en heses. ***p<0.01, **p<0.05, *p<0.1. Dependen a iable is a dummy
equal o 1 i esponden is sel -employed (wi h and wi hou dependen employees, col. 1), en ep eneu esponsible o supe ising
o he employees (col. 2), en ep eneu wi h dependen employees (col. 3), en ep eneu wi h an es ablishmen size wi h 10 o mo e
employees (col. 4). Reg esso s a e dummy a iables equal o one i he esponden has e ia y educa ion (col. 1) o epo s a ce ain
alue o be: “ e y much like me,” “like me,” “somewha like me,” “a li le like me,” and ze o i “no like me” o “no like me a all.”
Cul u al ai s and alues indica e whe he o he esponden i is impo an o: make own decisions and be ee; do wha is old and
ollow ules; hink new ideas and being c ea i e; be success ul and ha people ecognise achie emen s; be humble and modes , no
d aw a en ion; be ich, ha e money and expensi e hings; ha people a e ea ed equally and ha e equal oppo uni ies; ha go e nmen
is s ong and ensu es sa e y, and o seek ad en u es and ha e an exci ing li e. All columns a e es ima ed by a linea p obabili y model.
Indi idual le el con ols include whe he he esponden is a o eign na ional, age, age squa ed, gende and whe he he esponden ’s
a he was an en ep eneu
ence o ollowing ules and doing wha one is old,
being humble and modes and no a ac ing a en ion,
and a s onge p e e ence o seeking ad en u e and
ha ing an exci ing li e. The emaining e ec s on all
o he alues ha e he expec ed sign bu a e no s a is i-
cally signi ican . Simila esul s a e ob ained in all ou
eg essions, in all ables, and when we include egional
a he han coun y- ixed e ec s in ou model speci i-
ca ions (no epo ed, a ailable on eques ).
5.5 Media ion analysis’ indings
Ou es ima es indica e ha P o es an ism a ec s (i)
en ep eneu ship and (ii) a numbe o obse ed indi-
idual cha ac e is ics, alues and a i udes ha (iii) a e
associa ed wi h en ep eneu ship. These h ee pieces
o e idence combined, displayed in Tables 2,4, and 5,
a e pa o he media ion analysis ou lined in Sec .3.2.
In his sec ion, we p o ide u he e idence on how
123
1304 L. Nunzia a and L. Rocco
Table 5 The e ec o P o es an ism on educa ion and cul u al ai s and alues
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Te . Educ. F eedom Rules New ideas Success Humili y A luence Equali y S ong go . Ad en u e
Min P o - Min Ca h 0.0617*** 0.0021 −0.0370** 0.0034 0.0077 −0.0210** 0.0174 −0.0079 −0.0137 0.0436***
(0.0143) (0.0054) (0.0144) (0.0110) (0.0136) (0.0095) (0.0162) (0.0062) (0.0093) (0.0137)
Obse a ions 23,436 23,436 23,436 23,436 23,436 23,436 23,436 23,436 23,436 23,436
R-squa ed 0.091 0.169 0.052 0.096 0.080 0.095 0.072 0.152 0.107 0.077
Demog aphic con ols YES YES YES YES YES YES YES YES YES YES
Coun y FE YES YES YES YES YES YES YES YES YES YES
Yea FE YES YES YES YES YES YES YES YES YES YES
S anda d e o s clus e ed a he egional le el in pa en heses. ***p<0.01, **p<0.05, *p<0.1. Dependen a iable: dummy a iable equal o one i he esponden has e ia y
educa ion o epo s a ce ain alue o be: “ e y much like me,” “like me,” “somewha like me,” “a li le like me,” and ze o i “no like me” o “no like me a all.” The able p esen s
he es ima ed di e en ial e ec o being a mino i y P o es an compa ed o being a mino i y Ca holic (Min P o - Min Ca h) on e ia y educa ion (col. 1) and on a se o cul u al
ai s and alues. Cul u al ai s and alues indica e whe he o he esponden i is impo an o: make own decisions and be ee (col. 2); do wha is old and ollow ules (col. 3);
hink new ideas and being c ea i e (col. 4); be success ul and ha people ecognise achie emen s (col. 5); be humble and modes , no d aw a en ion (col. 6); be ich, ha e money and
expensi e hings (col. 7); ha people a e ea ed equally and ha e equal oppo uni ies (col. 8); ha go e nmen is s ong and ensu es sa e y (col. 9), and o seek ad en u es and ha e
an exci ing li e (col. 10). All columns a e es ima ed by a linea p obabili y model. Indi idual le el con ols include whe he he esponden is a o eign na ional, age, age squa ed,
gende and whe he he esponden ’s a he was an en ep eneu
123
The p o es an e hic and en ep eneu ship:... 1305
P o es an ism a ec s en ep eneu ship using he mos
common es s used in he media ion analysis li e a u e.
Ou analysis dis inguishes be ween an indi ec e ec
ha is media ed by he obse ed alues and a i udes
and a di ec e ec ha is he esidual e ec o P o es-
an ism on en ep eneu ship once he obse ed chan-
nels a e ac o ed ou . The la e is likely o be media ed
by o he ac o s ha a e no obse ed in he da a a
hand. Ou discussion should p o ide some di ec ions
o u u e esea ch on his opic.
Fi s , we es whe he he a iables measu ing alues
and a i udes ac as mode a ing a iables, i.e., whe he
hey impac en ep eneu ship h ough hei in e ac ion
wi h P o es an ism. We pe o m his es o all po en ial
channels conside ed in ou analysis and we ne e ind
he in e ac ion o be s a is ically signi ican . As a esul ,
we do no need o include such in e ac ion in he model.
Once mode a ion is excluded, we ocus on he medi-
a ion e ec , ollowing he ou -s ep p ocedu e ou lined
in Sec .3.2. As a i s s ep, we show ha he causal a i-
able and he ou come a e indeed co ela ed, as indica ed
by he s a is ically signi ican e ec o P o es an ism
on en ep eneu ship ha we ind in he eg essions p e-
sen ed in Table 2.
As a second s ep, Table 5shows ha P o es an ism
a ec s some o he obse ed cul u al ai s and alues
ha we conside as po en ial channel a iables. We ind
a posi i e e ec on e ia y educa ion and he as e o
ad en u e and a nega i e e ec on he p e e ence o
ollowing ules and doing wha one’s old, and o being
humble and modes and no d awing a en ion. This
consis s in he second s ep o he media ion assessmen
p ocedu e, whe e he media o is ea ed as an ou come
a iable.
As a hi d s ep, we al eady show he associa ion
be ween ou cul u al ai s and alue a iables and
en ep eneu ship in Table 4. Table 6shows ha he
e ec s ound in Table 4a e s ill p esen once P o es-
an ism is con olled o , i.e., excluding ha he medi-
a o a ec s he ou come h ough he e ec o he ea -
men on bo h.
Finally, as a ou h s ep, we need o show whe he
he media ion is comple e o pa ial. Table 6shows
ha he e ec o P o es an ism on en ep eneu ship is
signi ican ly educed once we con ol o he media-
o a iables. We obse e a 44% educ ion in he e ec
o P o es an ism on en ep eneu ship, wi h an almos
iden ical educ ion o en ep eneu s wi h dependen
employees. The e ec is educed by 36% and 20%
o en ep eneu s esponsible o supe ising o he
employees and wi h an es ablishmen size wi h a leas
10 employees, espec i ely. Ou empi ical e idence
is he e o e consis en wi h he p esence o pa ial
media ion.
As a u he s ep, we aim a es ima ing: (i) he size
o he media ion e ec , also called he indi ec e ec ,
in ela ion o he o al e ec o he ea men on he
ou come, and (ii) he di ec e ec o he ea men on
he ou come once he indi ec e ec is accoun ed o .
We pe o m a numbe o es s ha a e epo ed in
Table 7, whe e he single media ing a iables a e lis ed
by column and he es ima ed s a is ics by ow. The i s
ow epo s he associa ions be ween each single medi-
a ing a iable and he ou come, i.e., he p obabili y o be
an en ep eneu , es ima ed by single eg essions. These
coe icien s a e he ones epo ed in he i s column o
Table 4, and a e included o e e ence. The second
ow epo s he e ec s o P o es an ism on he medi-
a ing a iables ha a e es ima ed in Table 5,again o
e e ence.
In he hi d ow, we epo he P- alue o he es
p oposed by F i z and MacKinnon (2007) and F i z
e al. (2012). This is es ed o each media ing a i-
able h ough a join es ima ion o he eg essions o he
e ec o P o es an ism on each a iable oge he wi h a
model o he p obabili y o be an en ep eneu includ-
ing bo h he media ing a iables and he P o es an ism
dummies. The es is always s a is ically signi ican ,
sugges ing he p esence o media ion.
Nex , we pe o m he es p oposed by Sobel (1982),
illus a ed in Sec .3.2, ha p o ides a speci ic es i-
ma e o he indi ec e ec and i s s anda d e o o
each media ing a iable and o all a iables conside ed
oge he . The es ’s esul s a e epo ed in he ou h
ow o he able, whe e we p esen he indi ec e ec o
P o es an ism on en ep eneu ship h ough he media-
ion o each single a iable lis ed by column, and i s
s anda d e o , es ima ed h ough simul aneous eg es-
sions. We no ice ha he Sobel es sugges s he la ges
media ion e ec s o he ou dimensions ha we e
ound signi ican in he p e ious ables.
In he i h ow, we epo he p opo ion o he
o al e ec o P o es an ism on en ep eneu ship ha
is media ed by each single a iable. We no ice, again
ha he media ion akes place mos ly h ough he a i-
ables highligh ed abo e, wi h he media ion o educa-
ion alone explaining a ound 10% o he o al e ec o
P o es an ism, and he a i ude owa d ules a ound 5%.
123
1306 L. Nunzia a and L. Rocco
Table 6 The associa ion be ween e ia y educa ion, cul u al ai s, and alues and en ep eneu ship, a e con olling o P o es an ism
(1) (2) (3) (4)
Sel -employed Supe iso Wi h dep. empl Es ab. size ≥10
Te ia y Educa ion 0.035*** 0.034*** 0.029*** 0.011***
(0.008) (0.005) (0.005) (0.003)
F eedom 0.030*** 0.014** 0.011 −0.001
(0.010) (0.006) (0.007) (0.005)
Rules −0.030*** −0.008** −0.006 −0.005**
(0.005) (0.003) (0.004) (0.002)
New ideas 0.036*** 0.017*** 0.019*** 0.007*
(0.007) (0.005) (0.004) (0.003)
Success 0.034*** 0.013*** 0.010*** 0.006**
(0.006) (0.004) (0.003) (0.002)
Humili y −0.037*** −0.016*** −0.013** −0.009**
(0.009) (0.005) (0.005) (0.004)
A luence 0.015*** 0.013*** 0.012*** 0.001
(0.004) (0.003) (0.003) (0.002)
Equali y −0.032*** −0.026*** −0.019** −0.011*
(0.011) (0.009) (0.008) (0.006)
S ong go e nmen −0.031*** −0.005 −0.001 −0.002
(0.010) (0.006) (0.006) (0.005)
Ad en u e 0.017*** 0.011*** 0.012*** 0.005**
(0.004) (0.003) (0.003) (0.002)
MinP-MinC 0.0141 0.0107* 0.00758 0.00725**
(0.0094) (0.0056) (0.0059) (0.0031)
% educ ion MinP-MinC 43.8 36.0 43.7 20.5
Obse a ions 23,436 23,436 23,436 23,436
R-squa ed 0.059 0.041 0.036 0.014
Demog aphic con ols YES YES YES YES
Coun y FE YES YES YES YES
Yea FE YES YES YES YES
Sample mean 0.121 0.0553 0.0538 0.0195
S anda d e o s clus e ed a he egional le el in pa en heses. ***p<0.01, **p<0.05, *p<0.1. Dependen a iable is a dummy
equal o 1 i esponden is sel -employed (wi h and wi hou dependen employees, col. 1), en ep eneu esponsible o supe ising o he
employees (col. 2), en ep eneu wi h dependen employees (col. 3), en ep eneu wi h an es ablishmen size wi h 10 o mo e employees
(col. 4). Reg esso s a e dummy a iables equal o one i he esponden has e ia y educa ion (col. 1) o epo s a ce ain alue o be:
“ e y much like me,” “like me,” “somewha like me,” “a li le like me,” and ze o i “no like me” o “no like me a all.” Cul u al ai s
and alues indica e whe he o he esponden i is impo an o: make own decisions and be ee; do wha is old and ollow ules;
hink new ideas and being c ea i e; be success ul and ha people ecognise achie emen s; be humble and modes , no d aw a en ion;
be ich, ha e money and expensi e hings; ha people a e ea ed equally and ha e equal oppo uni ies; ha go e nmen is s ong and
ensu es sa e y, and o seek ad en u es and ha e an exci ing li e. Di e en ly om Table 4, he e, we con ol o he di e en ial e ec
o being a mino i y P o es an compa ed o being a mino i y Ca holic (Min P o - Min Ca h). We also epo he pe cen educ ion in
he poin es ima e o (Min P o - Min Ca h) ela i e o wha is ound in Table 2whe e we do no include he media ing a iables. All
columns a e es ima ed by a linea p obabili y model. Indi idual le el con ols include whe he he esponden is a o eign na ional, age,
age squa ed, gende and whe he he esponden ’s a he was an en ep eneu
123
The p o es an e hic and en ep eneu ship:... 1307
Table 7 Media ion analysis
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Te . Educ. F eedom Rules New ideas Success Humili y A luence Equali y S ong go . Ad en u e
E ec o M on En . .035*** .03*** −.03*** .036*** .034*** −.038*** .015*** −.032*** −.031*** .017***
(.008) (.01) (.005) (.007) (.006) (.009) (.004) (.011) (.009) (.004)
E ec o P o . on M 0.062*** 0.002. −0.037** 0.003. 0.008. −0.021** 0.017. −0.008. −0.014. 0.044***
(0.014) (0.005) (0.014) (0.011) (0.014) (0.009) (0.016) (0.006) (0.009) (0.014)
MacKinnon es P- al 0.0001*** .0115** 0.0001*** 0.0001*** 0.0001*** 0.0001*** .0042*** .007*** .0003*** .0001***
Sobel Tes , single M .00217*** .00006. .00108** .00012. .00026. .00078* .00026. .00026. .00043. .00074**
(.0006) (.00021) (.00052) (.00037) (.00044) (.00044) (.00028) (.00026) (.00029) (.00038)
P op To e ec .094 .003 .047 .005 .011 .034 .011 .011 .019 .032
media ed by single M
Sobel Tes , all M . .006**
(.001)
P op To e ec .26
media ed by all M
S anda d e o s clus e ed a he egional le el in pa en heses. ***p<0.01, **p<0.05, *p<0.1. The single media ing a iables a e lis ed by column and he es ima ed s a is ics by
ow. The media ing a iables a e a dummy a iable equal o one i he esponden has e ia y educa ion (col. 1) and whe he o he esponden i is impo an o: make own decisions
and be ee (col. 2); do wha is old and ollow ules (col. 3); hink new ideas and being c ea i e (col. 4); be success ul and ha people ecognise achie emen s (col. 5); be humble and
modes , no d aw a en ion (col. 6); be ich, ha e money and expensi e hings (col. 7); ha people a e ea ed equally and ha e equal oppo uni ies (col. 8); ha go e nmen is s ong
and ensu es sa e y (col. 9), and o seek ad en u es and ha e an exci ing li e (col. 10). The i s ow epo s he associa ions be ween each single media ing a iable and he ou come,
epo ed in he i s column o Table 4, o e e ence. The second ow epo s he e ec s o P o es an ism on he media ing a iables ha a e es ima ed in Table 5, o e e ence
The hi d ow epo s he P- alue o he es p oposed by F i z and MacKinnon (2007)andF i ze al.(2012) o check o he join signi icance o pa hs Aand B. The ou h ow
epo s he indi ec e ec o P o es an ism on en ep eneu ship h ough he media ion o each single a iable lis ed by column, es ima ed h ough a boo s apped (100 eps) Sobel
es by simul aneous eg essions (P eache and Hayes, 2008, Sobel, 1982)
The i h ow epo s he p opo ion o he o al e ec o P o es an ism on en ep eneu ship ha is media ed by each single a iable. The six h ow epo s he Sobel es o he o al
media ing e ec o all he obse ed cul u al and alue channels conside ed in he analysis, whe eas he se en h ow epo s he p opo ion o he o al e ec ha is media ed by all
ou obse ed media ing a iables. Indi idual le el con ols include whe he he esponden is a o eign na ional, age, age squa ed, gende , and whe he he esponden ’s a he was
an en ep eneu
123
1308 L. Nunzia a and L. Rocco
Finally, he six h ow epo s he Sobel es o he
o al media ing e ec o all he obse ed cul u al and
alue channels conside ed in ou analysis, whe eas he
se en h ow epo s he p opo ion o he o al e ec
ha is media ed by all ou obse ed media ing a i-
ables. We ind ha ou cul u al and alue dimensions
explain a ound 26% o he o al e ec o P o es an ism
on en ep eneu ship. The mos impo an media ing
a iables a e, again, e ia y educa ion and he se o
indi idualis ic pe sonali y ai s ep esen ed by a lowe
p e e ence o ollowing ules and doing wha one’s
old, o being humble and modes and no d awing
a en ion, and a s onge p e e ence o seeking ad en-
u es and ha e an exci ing li e.
5.6 Discussion o he empi ical indings, limi a ions,
and di ec ions o u u e esea ch
Ou empi ical indings indica e ha mino i y P o es-
an s a e mo e likely o engage in en ep eneu ship
han mino i y Ca holic, wi h he e ec mono oni-
cally inc easing in i s ela i e size he la ge is he
en ep eneu ’s business. These esul s a e obus o a
numbe o checks, and do no seem o depend on he
inclusion o some speci ic egions in he sample.
In addi ion, we ind ha P o es an ism a ec s a num-
be o pe sonal ai s and alues ha a e associa ed wi h
en ep eneu ship. O all ai s obse ed in ou da a ha
cha ac e ise en ep eneu s—i.e., being high skilled, o
alue eedom and dislike ollowing ules, o be inno a-
i e, o app ecia e success and a luence, o d aw a en-
ion o hemsel es, o seek an exci ing li e and o dis-
like equali y and a s ong go e nmen —P o es an ism
is ound o ha e a posi i e e ec on he p obabili y o
ha e a e ia y educa ion and seeking ad en u es and
ha e an exci ing li e, and a nega i e e ec on he p e -
e ence o ollowing ules and doing wha one’s old,
and o being humble and modes and no d awing
a en ion.
The media ion analysis sugges s ha hese alues
and pe sonal ai s ac as media o s o he e ec o
P o es an ism on en ep eneu ship. In pa icula , he
media ion o e ia y educa ion alone accoun s o
a ound 10% o he o al e ec o P o es an ism whe eas
he a i ude owa d ules o a ound 5%. The whole se
o cul u al and alue dimensions explain a ound 26%
o he o al e ec .
These esul s sugges ha , al hough ou obse ed
media o s accoun o a subs an ial pa o he e ec
o P o es an ism, he e is s ill a la ge pa o he chan-
nels behind he ela ionship be ween P o es an ism and
en ep eneu ship o be ully unco e ed. The e may be
addi ional media o s a wo k o he han educa ional
a ainmen and he se o indi idualis ic pe sonali y
ai s highligh ed in ou analysis. Indeed, one pos-
sible explana ion o he inabili y o iden i y mo e
o he exis ing media ion be ween P o es an ism and
en ep eneu ship is ha , al hough ou da a a e qui e
ich in p o iding a la ge se o indi idual-le el pe -
sonal ai s and alues, hey a e unlikely o con ain all
possible in o ma ion on he a i udes and alues ha
migh d i e he e ec o P o es an ism. In addi ion, he
epo ed in o ma ion on indi idual alues may be mis-
epo ed and/o measu ed wi h e o . In he la e case,
i measu emen e o is he e ogeneous ac oss P o es-
an s and Ca holics, he con ibu ion o ou media o s
may be unde es ima ed.
These conside a ions p o ide some guidance o
u u e esea ch. In pa icula , he collec ion o mo e
comp ehensi e da a on pe sonali y ai s and cul u al
alues and a i udes may help iden i ying a la ge se
o ele an media o s. In addi ion, alida ing such a i-
ables in e ms o hei ue adhe ence o he ac ual
belie s and alues o esponden s, may help o educe
measu emen e o s and possibly a bias in he es ima es.
Fo example, wo k e hic, a alue adi ionally associ-
a ed wi h P o es an ism and likely o be conduci e o
en ep eneu ship, may be measu ed wi h some e o
in ou da a, p e en ing us om p ope ly es ima ing
i s ole in he media ion. In addi ion, sel -con ol and
sel - egula ion (indica ed as d i e s o heal hy and p o-
social beha io ) may only be pa ially cap u ed by he
lis o alues we obse e. The Big Fi e pe sonali y ai s
(conscien iousness, ag eeableness, openness, ex a e -
sion, neu o icism) o hei componen s also appea o
be only pa ially co e ed by ou da a. Fo example,
ag eeableness and ex a e sion, which a e impo an
o es ablishing and main aining social ela ionships,
a e missing. Religious alues ela ed o amily, ma -
iage, and e ili y a e also missing. In ac , Ca holics
a e mo e commi ed o ma iage han P o es an s, wi h
a cons an emphasis on he pe manence o ma iage
and he p oc ea i e pu pose o ma iage (Ri chey and
Die z, 1990). These belie s in luence amily o ma ion
and he numbe o child en, which in u n may in lu-
ence he decision o become an en ep eneu .
123
