The Effect of Cultural Values On Ec - Dev, - Granato

The Effect of Cultural Values on Economic Development: Theory, Hypotheses, and Some
Empirical Tests
Author(s): Jim Granato, Ronald Inglehart and David Leblang
Source: American Journal of Political Science, Vol. 40, No. 3 (Aug., 1996), pp. 607-631
Published by: Midwest Political Science Association
Stable URL: http://www.jstor.org/stable/2111786 .
Accessed: 17/11/2013 18:13
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Midwest Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to
American Journal of Political Science.
http://www.jstor.org
This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

All use subject to JSTOR Terms and Conditions
The Effectof CulturalValues on
EconomicDevelopment:Theory,
Hypotheses,and Some EmpiricalTests*
JimGranato,MichiganState University
of Michigan
Ronald Inglehart,University
of NorthTexas
David Leblang, University
Theory:Cultural variablesareincorporatedintoa baselineendogenous economic

growth model.
Hypotheses: Cultural attitudestowardachievement andthrift
havea positiveeffect
oneconomic growth. Cultural attitudes
concerning postmaterialism havea negative
effecton economicgrowth.
Methods:Ordinary leastsquaresregression is usedto testeconomicandcultural
modelsofgrowth on a crosssectionof25 countries. Theencompassing principle
is usedtoresolvecompeting theoretical andtogenerate
specifications a finalparsi-
moniousmodel.A variant of Leamer'sExtreme BoundsAnalysis(EBA) is used
toevaluatethesensitivity ofparameter estimates.Theconclusions arefurther sup-
ported bynonparametric methods includingrobust andbootstrap
regression resam-
pling.Thedatafortheanalysisarefrom theWorldValuesSurvey(1990)andfrom
LevineandRenelt(1992).
Results:Anempirical modelthatincorporates bothculturalandeconomic variables
toanexplanation
is superior emphasizing onesetofthesevariables. Thefinalmodel
is robustto: (1) alterations
in theconditioning setof variables;(2) elimination of
cases; and(3) variations
influential in estimationprocedures.
Introduction
Do culturalfactorsinfluenceeconomic development?If so, can they
be measuredand theireffectcomparedwiththatof standardeconomicfac-
torssuch as savingsand investment?This articleexaminestheexplanatory
powerof thestandardendogenousgrowthmodel and comparesit withthat
of twotypesof culturalvariablescapturingmotivationalfactors-achieve-
mentmotivationand postmaterialist values. We believe thatit is not an
either/orproposition:culturaland economic factorsplay complementary
roles. This belief is borne out empirically;we use recentlydeveloped
econometrictechniquesto assess the relativemeritsof these alternative
explanations.
MaryBange,MikeBratton,
*We wouldliketothankPaulAbramson, DarrenDavis,Mark
Jones,KenMeier,andsomeanonymous
reviewers. withnamesinalpha-
Equal authorship,
beticalorder.
AmericanJournalofPoliticalScience,Vol. 40, No. 3, August1996,Pp. 607-31
? 1996bytheBoardof RegentsoftheUniversity ofWisconsin System

608 JimGranato,Ronald Inglehart,and David Leblang
Culturalfactorsalone do notexplainall of thecross-national variation

in economicgrowthrates.Everyeconomyexperiencessignificant fluctua-
tionsin growthratesfromyearto yearas a resultof short-term factorssuch
as technologicalshocks or unforeseencircumstancesthataffectoutput.
These could notbe attributed to culturalfactors,whichchangegradually.
A society'seconomicand politicalinstitutions also make a difference. For
example,priorto 1945, NorthKorea and SouthKorea had a commoncul-
ture,but South Korea's economicperformance has been farsuperior.
On theotherhand,theevidence suggeststhatculturaldifferences are
an important partof the story.Over thepast fivedecades, the Confucian-
influencedeconomiesof East Asia outperformed therestof the worldby
a wide margin.This holds truedespitethe factthattheyare shaped by a
wide varietyof economicand politicalinstitutions. Conversely,duringthe
same periodmostAfricaneconomiesexperiencedlow growthrates.Both
societal-leveland individual-levelevidence suggeststhata society'seco-
nomic and politicalinstitutions are not the only factorsdetermining eco-
nomicdevelopment;culturalfactorsare also important.
Traditionally,theliteraturepresentscultureand economicdeterminants
of growthas distinct.Politicaleconomistsand politicalsociologistsview
theirrespectiveapproachesas mutuallyexclusive. One reason lies in the
level of analysisemployedand withthistheunderlying assumptionsabout
humanbehavior.Anotherreasonis thatwe have had inadequatemeasures
of culturalfactors.Previousattemptsto establishtherole of cultureeither
inferculturefromeconomicperformance or estimateculturalfactorsfrom
impressionistic historicalevidence. Both factorscould be important, but
untilculturalfactorsare enteredintoa quantitative analysis,thispossibility
could not be tested.
By culture, we referto a systemof basic commonvalues thathelp
shape thebehaviorof thepeople in a given society.In mostpreindustrial
societies,thisvalue systemtakestheformof a religionand changesvery
slowly;butwithindustrialization and accompanyingprocessesof modern-
ization,theseworldviewstendto become moresecular,rational,and open
to change.
For reasonsdiscussedbelow, theculturesof virtuallyall preindustrial
societiesare hostileto social mobilityand individualeconomicaccumula-
tion. Thus, both medieval Christianity and traditionalConfucianculture
stigmatizedprofit-making and entrepreneurship.But, (as Weber argues) a
Protestantversionof Christianity played a key role in the rise of capital-
ism-and muchlater-a modernizedversionof Confuciansocietyencour-
ages economicgrowth,throughits supportof educationand achievement.
The theoryand evidencepresentedin thispaperis organizedas follows:
sectionone discusses theoriesthatdeal withthe effectof cultureon eco-

CULTURALVALUES AND ECONOMIC DEVELOPMENT 609
nomicdevelopment.This literature emphasizestheimportanceof motiva-

tionalfactorsin thegrowthprocess.Sectiontwo introducesthedata. This
data, based on representativenationalsurveysof basic values, enable us
toconstruct twomeasuresofculture-achievementmotivation andpostma-
terialistvalues. Section threediscusses the baseline endogenousgrowth
model.We drawupona recentpaperbyLevine andRenelt(1992) to specify
thismodel,and we augmentit withculturalvariables.Section fouris the
multivariate analysis.Economicand culturalvariableseach explainunique
aspects of thecross-nationalvariationin economicgrowth.Using the en-
compassingprinciplewe findthatan improvedand parsimoniousexplana-
tionforeconomicgrowthcomes froma modelthatincludesbotheconomic
and culturalvariables.Section fouralso examinesthe robustnessof this
economic-cultural model and findsthatthe specificationis robustto alter-
ationsin theconditioningset of information, theeliminationof influential
cases, and variationsin estimationprocedure.Section fiveconcludes.
Culture,MotivationalFactors,and EconomicGrowth
We firstdiscusstheliterature thatviews achievementmotivationas an
essentialcomponentin theprocessof economicdevelopment,and thenwe
explorehow culturalmeasuresfromtheWorldValues Surveycan be used
to examinethe effectof motivationon growth.
The motivationalliteraturestressesthe role of culturalemphasison
economic achievement.It grows out of Weber's (1904-1905) Protestant
Ethic thesis.This school of thoughtgave rise to thehistoricalresearchof
Tawney(1926, 1955), case studiesby Harrison(1992), and empiricalwork
by McClelland et al. (1953) and McClelland (1961) on achievementmoti-
vation.Inglehart(1971, 1977, 1990) extendsthisworkby examiningthe
shiftfrommaterialist to postmaterialist
value priorities.Althoughprevious
workmainlyfocuses on the politicalconsequencesof these values, their
emergencerepresentsa shiftaway fromemphasison economicaccumula-
tionand growth.These "new" values could be viewedas theerosionofthe
Protestant Ethicamongpopulationsthatexperiencehighlevels ofeconomic
security.
We suggestthatWeberis correctin arguingthattheriseof Protestant-
ism is a crucialeventin modernizingEurope. He emphasizesthattheCal-
vinistversionof Protestantism encouragesnormsfavorableto economic
achievement.But we view therise of Protestantism as one case of a more
generalphenomenon.It is important, notonlybecause of thespecificcon-
tentof earlyProtestant beliefs,butbecause thisbeliefsystemundermines
a setofreligiousnormsthatinhibiteconomicachievementand arecommon
to mostpreindustrial societies.
Preindustrialeconomiesare zero-sumsystems:theyare characterized

by littleor no economicgrowthwhichimpliesthatupwardsocial mobility

only comes at the expense of someone else. A society's culturalsystem
generallyreflectsthisfact.Social statusis hereditaryratherthanachieved,
and social normsencourageone to acceptone's social positionin thislife.
Aspirationstowardsocial mobilityare sternlyrepressed.Such value sys-
temshelp to maintainsocial solidaritybutdiscourageeconomicaccumula-
tion.
Weber's emphasison the role of Protestantism seems to capturean
important partof reality.The ProtestantReformationcombinedwiththe
emergenceof scientificlogic broke the grip of the medieval Christian
Worldviewon a significant partofEurope.Priorto theReformation, South-
ernEurope was economicallymoreadvancedthanNorthern Europe.Dur-
ing the threecenturiesafterthe Reformation, capitalismemerged,mainly
among the Protestantregionsof Europe and the Protestantminoritiesin
Catholiccountries.Withinthisculturalcontext,individualeconomicaccu-
mulationwas no longerrejected.
ProtestantEurope manifesteda subsequenteconomic dynamismthat
moved it farahead of Catholic Europe. Shiftingtradepatterns,declining
food productionin SouthernEurope and otherfactorsalso contributed to
this shift,but the evidence suggeststhatculturalfactorsplayed a major
role.Throughout thefirst150 yearsof theIndustrialRevolution,industrial
developmenttook place almost entirelywithinthe Protestantregionsof
Europe,and theProtestant portionsof theNew World.It was onlyduring
the second half of the twentiethcenturythatan entrepreneurial outlook
emergedin Catholic Europe and in the Far East. Both now show higher
ratesof economicgrowththanProtestant Europe. In short,theconceptof
the ProtestantEthic would be outdatedif we take it to mean something
thatexists in historicallyProtestantcountries.But Weber's moregeneral
concept,thatcertainculturalfactorsinfluenceeconomicgrowth,is an im-
portantand valid insight.
McClelland et al. (1953) and McClelland's (1961) workon achieve-
mentmotivationbuilds on theWeberianthesisbut focuseson the values
thatwereencouragedin childrenby theirparents,schools,and otheragen-
cies of socialization.He hypothesizesthatsome societiesemphasizeeco-
nomic achievementas a positivegoal while othersgive it littleemphasis.
Since it was notfeasibleforhimto measuredirectlythevalues emphasized
in given societiesthroughrepresentative nationalsurveys,McClelland at-
temptsto measurethemindirectly, throughcontentanalysisof the stories
and school books used to educate children.He findsthatsome cultures
emphasizeachievementin theirschool books moreheavilythanothers
and thattheformershowedconsiderablyhigherratesof economicgrowth
thandid the latter.

McClelland's work is criticizedon variousgrounds.It is questioned

whetherhis approachreallymeasuresthevalues taughtto children,or sim-
ply those of textbookwriters.Subsequently,writersof the dependency
school arguethatanyattemptto tracedifferences in economicgrowthrates
to factorswithina givenculture,ratherthanto globalcapitalistexploitation,
is simplya means of justifyingexploitationof the peripheraleconomies.
Such criticismtendsto discreditthistypeofresearchbutis hardlyan empir-
ical refutation.
Surveyresearchby Lenski(1963) and Alwin(1986) findthatCatholics
and Protestantsin theUnitedStatesshow significant differences
in theval-
ues theyemphasizeas the mostimportant thingsto teach children.These
differencesare moreor less along the lines of the ProtestantEthic thesis.
Alwinalso demonstrates thatthesedifferenceserodeovertime,withProtes-
tantsand Catholicsgraduallyconvergingtowarda commonbeliefsystem.
The Data
The WorldValues Surveyasks representative nationalsamplesof the
publicsin a numberof societies,"Here is a listof qualitieswhichchildren
can be encouragedto learnat home. Which,if any,do you considerto be
especiallyimportant?"This listincludesqualitiesthatreflectemphasison
autonomyand economic achievement,such as "thrift,""saving money
and things,"and "determination."Otheritemson thelistreflectemphasis
on conformity to traditional
social norms,suchas "obedience," and "reli-
gious faith."
We constructan index of achievementmotivationthatsums up the
percentagein each countryemphasizingthefirsttwo goals minustheper-
centageemphasizingthelattertwogoals. This methodofindexconstruction
controlsforthetendencyofrespondents in some societiestoplace relatively
heavyemphasison all of thesegoals, whilerespondentsin othercountries
mentionrelativelyfew of them.
Figure 1 shows the simple bivariaterelationshipbetweenthis index
and ratesofpercapitaeconomicgrowthbetween1960 and 1989.1 The zero-
pointon theachievementmotivationindexreflectsthepointwhereexactly
as manypeople emphasizeobedienceand religion,as emphasizethrift and
determination.As we move to theright,thelattervalues are givenincreas-
ing emphasis.A givensociety'semphasison thrift and determination over
obedienceand religiousfaithhas a strongbivariatelinkagewithits rateof
economic growthover thepast threedecades (r = .66; p = .001).
Though often stereotypedas having authoritariancultures,Japan,
China, and South Korea emergenearthepole thatemphasizesthrift more
'Data sourcesandvariabledescriptions
arecontained
in Appendix
Table 1.

Figure1. Economicgrowthrate by achievement

motivation
scoresof
publics.
EastAsia
Japanl
5
.0
0 4
C)
0 pi
d)
San Nrway Fini2
Brazil Italy o 0 utl

cx3_L Canada *
o ireland France Belgium Sweden
z & New World

\ ~~~~~Europe Gerany I
United0 * Denmark
CL
~~~~~~~~Statea
Mexico * Nether-
o 2_ Great
d) Indla Britain
(U
asco~~~~~~~~~ot S
,z 1
Co Africa A
Afic
-100 -75 -50 -25 0 + 25 + 50 + 75 + 100

Public's score on 4-ItemAchievement Motivation Index
Note: AchievementMotivationindex is based on the percentagein each
society who emphasized "Thrift" and "Determination" as important
thingsfora childto learn,minus thepercentageemphasizing"Obedience"
and "Religious Faith."
heavilythanobedience.The threeEast Asian societiesrankhigheston that

dimension,while the two Africansocieties includedin this surveyrank
near the oppositeend of thecontinuum,emphasizingobedience and reli-
gious faith.
The publics of India and the UnitedStates also fall towardthe latter
dimension.It reflectsthe
end of the scale. This is not an authoritarianism
balance betweenemphasison two typesof values. One set of values-
thriftand determination-supporteconomic achievement; while the

other-obedience and religiousfaith-tend to discourageit,emphasizing

conformity to traditionalauthority and groupnorms.These two typesof
values are notnecessarilyincompatible:some societiesrankrelativelyhigh
on both,whileothersrankrelativelylow on both.But,therelativepriority
given to themis stronglyrelatedto its growthrate.
Do culturalfactorslead to economicgrowth,or does economicgrowth
lead to culturalchange?We believe thatthecausal flowcan workin both
directions.For example,thereis strongevidencethatpostmaterialist values
emergeswhena societyattainsrelativelyhighlevels of economicsecurity.
In thiscase, economic change reshapesculture.On the otherhand,once
thesevalues becomewidespread,theyare linkedwithrelativelylow subse-
quentratesofeconomicgrowth.Here,cultureseemsto be shapingeconom-
ics-a parallelto theWeberianthesis,exceptthatwhatis happeninghere
is, in a sense, therise of theProtestant Ethic in reverse.
Demonstrating causal connectionsis always difficult.In connection
withour achievementmotivationindex,the obvious interpretation would
be thatemphasison thrift and hard work,ratherthanon obedience and
respectis conduciveto economicgrowth.The twomostsensitiveindicators
of thisdimensionare thrift, on theone hand,and obedienceon theother.
For some time,economistshave been aware thata nation'srate of gross
domesticinvestment is a majorinfluenceon its long termgrowthrate.In-
vestment,in turn,depends on savings. Thus, a societythatemphasizes
thrift,producessavings,whichleads to investment, and laterto economic
growth.We provideevidence below thatthis is probablythe case. This
does notruleout thepossibilitythateconomicgrowthmightbe conducive
to thrift but thislinkageis less obvious.
Emphasison obedienceis negativelylinkedwitheconomicgrowth,for
a conversereason.In preindustrial societies,obediencemeans conformity
to traditionalnorms,whichde-emphasizeand even stigmatizeeconomic
accumulation.Obedience,respectforothers,and religiousfaithall empha-
size obligationsto sharewithand supportone's relatives,friendsandneigh-
bors.Such communalobligationsare strongly feltin preindustrial
societies.
But fromthe perspectiveof a bureaucratizedrational-legalsociety,these
normsare antithetical to capital accumulationand conduciveto nepotism.
Furthermore, conformity to authorityinhibitsinnovationand entrepreneur-
ship.
The motivational componentis also tappedby materialist/postmaterial-
ist values, withpostmaterialism havinga negativerelationshipwitheco-
nomic growth.The achievementmotivationvariable is only modestly
correlatedwith the materialist/postmaterialist dimension (r = -.39;
p = .0581). Thoughbothdimensionshave significant linkageswitheco-
nomicgrowth,theyaffectitin different ways.The achievementmotivation

614 and David Leblang
JimGranato,RonaldInglehart,
dimensionseemsto tapthetransition frompreindustrialto industrialvalues

systems,linkedwiththe modernization process.
dimensionreflects
The materialist/postmaterialist thetransitionto post-
industrialsociety,linked witha shiftaway fromemphasison economic
growth,towardincreasingemphasison protectionof theenvironment and
on thequalityof lifemoregenerally.Previousresearchdemonstrates that:
(1) a gradualshiftfrommaterialisttowardpostmaterialist goals has been
takingplace throughout advanced industrialsociety;(2) thatthis shiftis
stronglyrelatedto the emergenceof democracy(r = .71); but (3) thatit
has a tendencyto be negativelylinkedwitheconomicgrowth(Abramson
and Inglehart1995).
Baseline EndogenousGrowthModel
Neoclassical growthmodels today owe much to the work of Solow
(1956) and Swan (1956). The essentialfeatureofthesemodelsis theirfocus
on savings,populationgrowth,and shiftsin technology.Productionfunc-
tionsdependon shiftsin these "exogenous" variables.For example,one
could tracetheeconomicgrowthconsequencesresultingfroma shiftin the
rateof saving,thepopulationgrowthrate,or technology.The weaknessin
thesemodels,however,is thattheyshow a paradoxicalsteadystateresult.
In thesemodelsaggregatesavingsproducea level ofcapitalformation such
thatgross investment exceeds depreciation,and therebyincreasescapital
per worker.Consequently,at thelimit,themarginalproductof capitalde-
clines to the pointwhere the savings (revenue)generatedby the capital
falls to a level just large enough to replace old equipmentand provide
machinesfornew workers.The steadystateresultis an unchangingstan-
dard of living.2
This latterresultis clearlynot supportedby evidence fromthe real
world.In timeeconomistsbegan searchingforways to augmenttheneo-
classical model thatwould allow sustainablegrowthand increasesin the
standardof living. These models have been termedendogenousgrowth
models. At the heartof the endogenousgrowthliteratureis an emphasis
on the productivity of the population(Lucas 1988; Romer 1990). Unlike
the "old" neoclassical models,endogenousgrowthmodels show thatre-
produciblecapitalneed nothave decreasingreturns to scale. Growthcan be
sustainedin endogenousgrowthmodels.In particular, theyassumeconstant
returnsto scale to a broad rangeof reproducibleinputs,includinghuman
capital.
The two leadingschools of thought, however,differin theiremphasis.
2Thisresultwas basedon an assumption

of constant to scale andfixedtech-
returns
nology.

CULTURAL VALUES AND ECONOMIC DEVELOPMENT 615
Romer(1990), arguesthatResearchand Development(R&D) spendingis

thekey to new technologicaldevelopments,whichresultin increasingso-
cial returnsto social knowledge.Alternatively,
Lucas (1988) argues that
expansionof humancapital in termsof botheducationand "learningby
doing," also plays a pivotalrole in economicgrowth.
Empiricalendogenousgrowthmodels invariablyare of thefollowing
form:
Yj=I 1,O + HX + [1]
whereYiis outputgrowth(per capita) forcountryi, Ii,0is a setof economic

variablesmeasuredat thebeginningof thetimeperiodforcountryi. These
variablesincludeinitiallevels of wealthand investment in humancapital,
and are includedbecause studiesby Barro(1991), Helliwell(1994), Levine
and Renelt(1992), and Mankiw,Romer,and Weil (1992) all findthatthey
have a robustand positivepartialcorrelationwitheconomicgrowth.Xi is
a setof "othervariables" includinga constant,physicalcapitalinvestment
rates(as a percentof GDP usually),and whateverothervariablestheinves-
tigatoris interestedin exploring.3
Obviously,giventhediscussionin section
one, our X variableswill includeachievementmotivationand postmateri-
alism.
We have a greatdeal of confidencein theselectionof theseeconomic
variables.Levine and Renelt(1992) findthattheinitiallevel of per capita
income,theinitiallevel of humancapitalinvestment, and theperiodshare
ofinvestment to GDP haverobustcorrelations witheconomicgrowth.Their
investigation uses a variantof Leamer's (1983) ExtremeBounds Analysis
(EBA) wheretheemphasisis on the"stability"of various"focus" param-
eterswhenvariablesare removedor added. Theyfindthatmostotherexog-
enous variablesare fragileto alterationsin theconditioningsetof informa-
tion. Thus, the conclusions of most empiricalwork rest on parameter
estimatesthatfluctuate at a magnitudelargeenoughto makescholarswary.
Levine and Renelt's (1992) workis also informative in thattheyprovide
a straightforward way to evaluate the sensitivityof theculturalvariables.
We implementthisprocedurebelow.
MultivariateAnalysis
Our empiricalapproachis straightforward:
we beginby estimating
(via
OLS) a baselineendogenous modelthatincludesvariablesidenti-
growth
3Theindicator
forhumancapitalinvestment is thenumber ofstudents
enrolledat pri-
maryandsecondaryeducationinstitutions
relativetothetotalpopulation
ofthatage group.
Theindicator
forphysical
capitalinvestment
is theratioofrealdomestic
investmenttoGDP.
Thedefinitions
andsourcesforthisdata,whichwe uselater,arefoundinAppendix Table 1.

fiedby Levine and Renelt(1992) as havingrobustpartialcorrelations with

economicgrowth.Using data for25 countries4 we firsttesttheendogenous
growthspecification(Model 1 in Table 1). FollowingEquation [1], a na-
tion's rate of per capita economic growthis regressedon its initiallevel
of per capita income and humancapital investment (educationspending)
as well as on its rate of physicalcapital accumulation.As expected,the
resultsare quite compatiblewiththe expectationsof endogenousgrowth
theory.The resultsof Model 1 are summarizedas follows:(1) thesignifi-
cantnegativecoefficient on theinitiallevel of per capita incomeindicates
thatthereis evidence of "conditionalconvergence."That is, controlling
forhumanand physicalcapitalinvestment, poorernationsgrowfasterthan
richernations;(2) investment in humancapital(educationspending)has a
positiveand statisticallysignificanteffecton subsequenteconomicgrowth;
and (3) increasingthe rate of physicalcapital accumulationincreases a
nation'srateof economicgrowth.
Overall thisbaseline economic model performswell: it accountsfor
55% of thevariationin cross-nationalgrowthratesand is consistentwith
priorcross-nationaltestsof the conditionalconvergencehypothesis(e.g.,
Barro 1991; Mankiw, Romer,and Weil 1992). Model 1 also passes all
diagnostictests,indicatingthattheresidualsare notseriallycorrelated5 (LM
test),arenormallydistributed (Jarque-Bera test),andhomoskedastic(White
test).
Model 2 in Table 1 regressestherateof per capita economic growth
on a constantand thetwoculturalvariables.As expected,bothachievement
motivationand postmaterialismare significantpredictorsof economic
growthand have theexpectedsign.Thus,thearguments of bothProtestant
Ethicand postmaterialist typetheoriescannotbe rejectedby thisevidence.
In addition,thesevariables,takenby themselves,do fairlywell,accounting
for59% of the variancein growthrates.A glance at the diagnosticsalso
indicatesthattheresidualsare well behaved.
EmpiricalModels: Encompassing
ComparingCompeting Results
Both the economic and culturalmodels give similargoodness-of-fit
performance.Each model's regressorsare statisticallysignificant.Yet,
whichmodel is superior?Or do bothmodels possess explanatoryfactors
thatare missingin theother?In Table 1 theSchwarzcriterion(SC) favors
4Thenations included
inthemultivariate
analysis
are:Austria,Belgium, Brazil,Canada,
China,Denmark, Finland,
France,Germany, GreatBritain,
India,Ireland,
Italy,Japan,Korea,
Mexico,Netherlands, Nigeria,
Norway, SouthAfrica,Spain,Sweden,Switzerland, Turkey,
UnitedStates.
'This is a checkforspatialcorrelation
betweentheerrors ofthecases.

Table 1. OLS EstimationofEconomicGrowthModels

DependentVariable: Mean Rate ofPer Capita EconomicGrowth
(1960-89)
Model Variable Model 1 Model 2 Model 3 Model 4
Constant -0.70 7.29* 3.16 2.40*
(1.08) (1.49) (1.94) (0.77)
Per CapitaGDP in 1960 -0.63* -0.42* -0.43*
(0.14) (0.14) (0.10)
Educationin 1960
Primary 2.69* 2.19* 2.09*
(1.22) (1.06) (0.96)
SecondaryEducationin 1960 3.27* 1.21
(1.01) (1.08)
Investment 8.69* 3.09
(4.90) (4.40)
Achievement
Motivation 2.07* 1.44* 1.88*
(0.37) (0.48) (0.35)
Postmaterialism -2.24* -1.07
(0.77) (1.03)
R2 Adjusted .55 .59 .69 .70
SEE .86 .83 .72 .71
LM (X2(1)) .42 .65 .68 .87
Jarque-Bera (X2(2)) .05 .30 .18 .57
White(X2(1)) .28 .24 .37 .18
SC .119 -.117 -.095 -.352
variable:3.04;N is 25 forall models;Standard

Notes:Meanofdependent in paren-
errors
theses.
*t test:p < .05.
theculturalmodel (Model 2), but it is moredesirableto implementa re-

searchstrategythatallows us to eliminatevariablesand explanationsthat
are empiricallyunsupportable. Mizon and Richard(1986) devised theen-
compassingprinciple-a set of statisticalprocedures-consistentwitha
progressiveresearchstrategy. In thissubsectionwe use theencompassing
principleto guide us in buildinga theoretically
parsimoniousand statisti-
model of economicgrowth.6
cally efficient
6Ifmodelsunderconsideration orifa newmodelsimplyaddsoneormore

arenested,
variablesto theoriginalmodel,conventionalsignificancetests(e.g.,Waldand likelihood
ratiotests)foradditionalparametersare sufficient
formodelrefinement. a
Traditionally,
modelis saidto be nestedifitcan be obtainedby imposing restrictions
on an alternative
model.Analternative nesting is usedhere.Following
conceptualization Hendry andRichard
(1989),sinceall modelsarenecessarilyreductionsofa datagenerationprocess(DGP), they
mustnecessarily be relatedor minimally
nested.

A progressiveresearchstrategyrequiresthatthe test statisticsem-

ployedbe comparableacross samples.We knowthatthecoefficient of de-
termination-theR2-does not pass muster(Achen 1982). On the other
hand,theresidualvariance(SEE) is comparableacross samplesand,there-
fore,is an appropriateencompassingteststatistic.In fact,Hendryand Rich-
ard (1989) arguethata necessaryconditionforone model to encompassa
rival is variancedominance.The superiormodel mustbe a moreaccurate
(smallerSEE) characterization of thedata generationprocess(DGP). Vari-
ance dominancealso has meaningforsuperiorout of sampleperformance.
Thus, a model thatencompassesa rival also has superiorforecastperfor-
mance.7
Empiricalmodelsare de factoabstractions of theDGP based on certain
theoretical The encompassingprincipleinvestigates
constructs. thevalidity
of a modelrelativeto an alternative by determining whethera modelstatis-
ticallyaccountsforthemainfeaturesof a rival.Encompassingenablesana-
lyststo choose one model over anotherand assess therelativecredibility
of theoreticaland empiricalmodels.
The encompassingprinciplehas attractivefunctions:it aids the re-
searchprocessat twodifferent levels. Encompassingassistsin thebuilding
of a parsimoniousmodel. Because an encompassingmodel (denoted t)
predictsa rival model's parameters, it is possible to determinewhichpa-
rameters(variables)shouldbe eliminated, orreplacedbybetteralternatives.
Second, encompassinghelps findmisspecification. Since an encompassed
modelis merelyan erroneousreparameterization of a morecrediblemodel,
one can, given the latter'sparameters, determinewherethe encompassed
model wentawry.Therefore,repeatedapplicationsof encompassingto a
large set of models facilitatesthediscoveryof usefulmodels thatare ap-
proximations to theDGP, butcloserto theDGP thanothermodelsconsid-
ered.
Encompassing'sstatisticalanalogue centerson thecompetingempiri-
cal modelsparametersand residualvariances.Formaldefinitions and deri-
vationsof parameterand varianceencompassingare foundin Granatoand
Suzuki (N.d.).
In the formercase, the concernis whetherthe substitution of a rival
model's parameters forthosein thecurrent modelare statistically
(in)distin-
'This is notto saythatgoodness-of-fit andattendant

statistics testsdo not
diagnostic
havevalue.But,theirfunction is to indicatea specificmodel'saccuracyand whether the
parametersareconsistentandefficient. Theproper useofencompassing testsandtheencom-
passingprincipledependson modelswhicharevalidapproximations oftheDGP (i.e.,the
modelpassesanyand all diagnostic tests).Encompassing testsare nota replacement for
conventional testing rather,
practices; encompassing testsaugment existing This
practice.
augmentation putsempirical workwithin a progressive researchframework.

Table 2. EncompassingTests
Model 1 4 Model 2 Form Test Form Model 2 4 Model 1
3.34 t(l) JA-Test t(l) 2.44
5.42 F(2,17) JointModel F(4,17) 2.87
Notes: * = p > .05.
guishable(parameterencompassing).To testforparameterencompassing
we employa jointF-test.This testcombinesall instruments fromtwocom-
petingmodelsintoa largegeneralmodel(artificial nesting).Commonvari-
ables are removedfromthegeneralmodel to avoid multicollinearity. The
testimposeszero restrictions on theinstruments of each (sub)modelto de-
termineifeitherset of instruments altersthesum of squarederrorssignifi-
cantlyfromthegeneralmodel(see Appendix).In a testof Model 1 4 Model
2 acceptingthe null indicatesthe zero restrictions on Model 2 have no
statisticaleffecton thejoint model. Model 1 4 Model 2 in thiscase. In a
progressiveresearchstrategycontext,thisparametersubstitution testsif
one theoryexplains morethana rival's explanation.
Variance encompassing,on the otherhand,requiresthatthe "new"
parameterrestrictions be at least as efficient-intermsof theresidualvari-
ance-as theoriginalmodel.8In short,a progressiveresearchstrategy not
onlynecessitatesthatnew and novel factsbe putforthby thesuperiorthe-
ory,but thatthistheoryis also a moreaccurateexplanation.
Returning to themodelsin Table 1,culturalvalues clearlymatter.With
theencompassingprinciplein ourarsenal,we compareendogenousgrowth
and culturalexplanationsforeconomicgrowth.The encompassingresults
presentedin Table 2 are definitive: The JA-testforvarianceencompassing
is significant
and indicatesthatbothmodelsencompasseach other.Neither
model is an "efficient"substitute fortheother.In addition,bothmodels'
parameter encompasseach otheras indicatedbysignificant F-tests.In short,
bothmodelsexplainaspectsofgrowththattherivalcannot.The implication
is straightforward: growthratesare best understoodas a consequence of
botheconomicand culturalfactors.
Whathappenswhenwe combinetheeconomicmodelwiththecultural
model?The resultsof thisexperiment are containedin Model 3. Beginning
withtheendogenousgrowthvariables,addingthevariablesfromModel 2
8Theencompassing principle
is mostcommonly usedforlinearmodels.To assessvari-
ance encompassingwe use theJA-test(Davidsonand MacKinnon1981) as modified by
Godfrey(1984).Thistestinvolvesconstructing
Y-hat'sfortwocompeting models,andthen
sideof itsrival.
addingtherivalY-hat'sto theright-hand

significantlyalterstheparameter estimates andstandard errors on second-

aryeducation spending andphysicalcapitalinvestment. In fact,thecoeffi-
cienton thephysicalcapitalinvestment variablechangesdramatically. It
decreasesfrom8.69 in Model 1 to3.09 in Model3. Whilethiscoefficient
stillhas theexpectedsign,itis nowfarfromsignificant.
Whyis physicalcapitalinvestment, a variable"robustly"correlated
witheconomicgrowthin a numberof otherstudies,now insignificant?
Achievement motivation quitepossiblyis conduciveto economicgrowth
at leastpartlybecauseit encourages relativelyhighratesof investment.
Achievement motivation also has an important directeffect on economic
growth rates,quiteapartfromitstendency toincreaseinvestment. Presum-
ablythedirectpathfromculture to economicgrowth reflects theeffectof
motivational factors on entrepreneurship andeffort.
Returning to theanalysisof Model 3 in Table 1, we now examine
thedirecteffect ofculturalvalues,particularly achievement motivation,on
economicgrowth. As inModel2, achievement motivation is positively
and
significantlyrelatedto economicgrowth. Combining Model2 andModel
3 resultsinpostmaterialism nowbeinginsignificant, however. Thisis prob-
ablydue to thefactthatcountries withpostmaterialist valuesare already
fairlyrich;thebivariate correlationbetweentheinitiallevelofwealthand
postmaterialism is .75 andis significantat the.0000level.Combining the
regressors ofthesemodels(Model3) we againhavea modelthatdoesnot
violateanydiagnostic test.In addition,
thefitis moreaccurate(SEE).
SensitivityAnalysis
Table 1 contains an additional In Model4 we eliminate
specification.
thethreeinsignificantvariablesfromModel3-those forpostmaterialism,
investment, andsecondary schoolenrollment-to checkthestabilityofthe
remaining parameters. Model 4 is themostparsimonious and efficient
model,explaining 70% ofthevariance inpercapitagrowth rateswithonly
threevariablesandgenerating a Schwarzcriterionvalueof -.352. In addi-
tion,theresidualsare wellbehavedand themodelpassestestsforserial
(spatial)correlation,
normality, andheteroskedasticity.
Aretheresults inModel4 theconsequence ofeitherhighly influential
observations or theproductof specificvariablesselected?We ask these
questionsbecauseJackman (1987) demonstratesthatremovalof evena
singleinfluentialcase mayreduceparameter estimatesto insignificance.
Levineand Renelt(1992) takea different approachand showhowalter-
ationsinthesetofvariablesincludedina modelnotonlychangethestan-
darderrorof a variableof interest butalso cause theparameter estimate
tochangesigns.We usebothoftheseapproaches toevaluatetherobustness
of Model4.

Figure2. PartialRegressionPlot ofAchievement

Motivationon
EconomicGrowth.
aM+2.4-
+2.4 Korea
0) China
0, +2.0
Japan
-+1 6
0
0) +1.2 Canada
-
._E Finland
o +0.8
C - Norway
o United 0Sweden
a) +0.4 - States
Great
Denmark
,o0
S
co Britaln.Belgium Austria
Switzer - *
0. 0 Italy land /
co 0
o - Spain Turkey Netherlands
a- ~~0 0 0
d) -0.4 -
. Mexico
_ Bra
O _ Germany
0 -0.8 *-
Ireland India
I- 1.2
. 0 Nigeria
France
a) SouthAfrica
E -1.6 l IlI l I I I I I I I I I I I I I
-60 -40 -20 0 + 20 + 40 +60 +80 +100
Public's score on 4-itemAchievement MotivationIndex
The firstrobustness testsexaminetheinfluence ofindividualcases on

theparameters of interest.Fox (1991,21) suggeststhatinfluence can be
thought ofas a productofleverageanddependency. Leverage-thepoten-
tialforthemodelas a wholetobe influenced bya fewlarge"X" values
is measured byCook'sDistance(D) andDFFITS. Discrepancy is measured
bystandardized andstudentized residuals
andindicatewherelargeroutliers
generate largeresiduals.Calculatedvaluesforthesequantities based on
Model4 arecontained in Appendix Table 2.
Figure2 is a partial
regression plotoftheeffect
ofachievement motiva-
tionon economicgrowth.9 It appearsKoreais quiteinfluential.
The diag-
9We also examinepartialregression plotsfortheeffectof percapitagrossnational

productandprimary schoolenrollment. Theseplotsareavailablefromtheauthors. Fora
discussionof partialregressionplotssee Bollenand Jackman (1985). For an illustrative
see Jackman
application (1987).

Table 3. Diagnosticson Model 4

DependentVariable:Mean Rate ofPer Capita EconomicGrowth
(1960-89)
4a 4b 4c 4d 4e
Korea/U.S. Korea/U.S. Robust Bounded Bootstrapped
Model Dummy Omitted Regression Influence 1000 reps
Variable
Constant 2.42* 2.36* 2.29* 1.98* 2.42*
(0.63) (0.64) (0.79) (0.63) (0.78)
PerCapitaGDP in 1960 -0.44* -0.39* -0.41* -0.44* -0.43*
(0.08) (0.09) (0.10) (0.08) (0.09)
Educationin 1960
Primary 1.98* 1.88* 2.09* 2.53* 2.10*
(0.79) (.80) (0.98) (0.80) (0.96)
Achievement
Motivation 1.87* 1.78* 1.81* 1.79* 1.88*
(0.28) (.30) (0.35) (0.27) (0.33)
Korea/United
StatesDummy 1.43*
(0.43)
Notes: Meanofdependent
variable:3.04.;N is 25 forall modelsexceptformodel4b (N = 23); Standard
errorsin parentheses.
*t test:p < .05.
4a estimated witha dummy variablecoded 1 forKoreaandtheUnitedStates.
4b estimated withoutKoreaandtheUnitedStates.
4c estimated usingrobustregression.
4d estimated usingWelsch'sone stepboundedinfluence estimator
(Welsch1980).
4e estimated usingboostrapresamplingoftheresidualswith1,000replications.
nosticsprovidemoreconcrete evidencethatKoreais an outlier.It has a

standardized residualof2.64whichis higher thantheusualcutoff
of ?2.0.
Othercases ofnoteareGermany (-1.85), Canada(1.66),andtheUnited
States(1.5); however, thesethreecasesdo notexceedthecutoff.Do these
cases radicallyinfluencetheparameter estimates?
The Cook's Distance(D) diagnostic measures influence
on themodel
as a whole(Cook andWeisberg1982).A case is considered if
influential
Cook's Distance(Di) > 4/n.10Two casesexertinfluence accordingtothis
criterion:Korea (D = .42) and the UnitedStates (D = .18).11
We deal withthisproblemin a number of ways.12Table 3 compares
'?Chatterjee
andHadi(1988) suggest thecutoff forCook'sDistanceis defined as Di >
41(n - k - 1). To be more cautious (due to the small sample size) we use Di > 41n.
"A relateddiagnostic,
DFFITS,confirms theinfluenceofthesetwoobservations. The
DFFITS valueforKoreais 1.54andfortheUnitedStatesis .87.Thecutoff pointforDFFITS,
as suggested byBollenandJackman (1985) is 2* (k/n)"2.
'2Alongwiththeseoverallmeasures therearediagnostics thatexaminetheinfluence
ofindividualcasesonspecificparameter estimates.Themostpopularis theDFBETA which
looksat theeffecton eachcoefficient ofdeleting theobservations one at a time.We found
thatno case exertsundueinfluence on thevariablewe areprimarily interestedin,thefour
itemachievement motivationindex.

OLS results from Model4 inTable1 witha variety ofestimation strategies.

The firststrategyis theonewe liketheleast.Thisstrategy involveseither
including a dummy variablefortheinfluentialcases (4a) or droppingthe
offending cases fromthesample(4b). Whilebothoftheseequationscon-
firm theresultsinModel4, however, theyarebothad hocandatheoretical.
Thereis no a priori theoretical reasonforaddingadditional variablesto
Model4. The strategy ofremoving influential indefensi-
cases is similarly
ble. Not onlydo thesecases providevaluableinformation, butalso one
imagines a situationwhereafter deletingobservations andreestimatingthe
modelotherinfluential andareremoved.
cases areidentified Thisprocess
continues untilfewinteresting observationsremain."3
Sincewe arenotsatisfied withstrategiesthateitheradd variablesor
removeobservations, we reestimate Model4 usingthreealternative nonpa-
rameteric techniques. Equation4c is estimated usinga variantof robust
regression.Robustregression usesestimatorsthatperform wellevenwhen
thereareminor violationsofassumptions regarding theunderlying popula-
tion.'4
We stillhavea problem whencaseshavehighleveragebecauselever-
age affectsrobust
regressioninthesamewaythatitaffects OLS (Hamilton
theinfluence
1992). In thesecases, we wantto constrain of suchcases
withincertainbounds.In equation4d, we use a simpleone-stepbounded
influenceestimator suggestedby Welsch(1980). The bounded-influence
estimatoralso uses weightedleastsquareswiththeweightsbeingdeter-
minedas follows:(1) perform OLS regressionandcalculateDFFITS; (2)
use theDFFITS valuesto construct a weightequal to one forIDFFITSI
'3Evenifonesubscribes tothisstrategy thereis evidencethatModel4 performs quite

well.Figures1 and 2 indicatethatthecountries of East Asia (China,Japan,and Korea)
havecultural and growth experiences quitedifferent fromtherestof thesample.Thisis
reasonenoughforus to argueagainstremovalofthesecases.Whenwe reestimate Model
4 without China,Japan,andKorea,however, theresults stillsupportourgeneralconclusion:
thevariablesareall significant downward)
(albeitattenuated andwiththeexpectedsign.
'4Therobustregressionprocedure weusecomesfrom thefamily ofM-estimators.Esti-
mationproceedsas follows:
(1) Use OLS toobtaininitialregression parameter estimates values)
(tobe usedas starting
andcalculatetheresiduals. In generalcircumstances thefirst stepwouldbe touse OLS to
estimate theparameters andcompute Cook's Distance.Cases Di > 1 areeliminated prior
starting
to calculating values(as Appendix Table2 indicates, oursampledoes notcontain
anycases whereDi > 1).
(2) Use theresidualsto calculatea setof case weights.
(3) Applyweighted leastsquaresto obtaina newsetof parameter estimatesandcalculate
newresiduals.
(4) Go backtostep2 andrepeattheprocessuntilthemaximum inweights
difference drops
below.01 (Hamilton1992).

< .34 andequal to .34/1DFFITSI forIDFFITSI > .34. The cutoff of .34
is usedbecauseWelschsuggeststhatit providesforapproximately 95%
asymptoticefficiency.Theresultsofusingthisbounded-influence estimator
are in Equation4d. Again,we findthattheparameter estimatesare not
muchdifferent fromthoseobtainedusingOLS. Note,however, thatthe
weightsassignedby thebounded-influence estimatorare a resultof the
valueforDFFITS.
The finalestimation technique we use is a nonparametric approach:
thebootstrap.Bootstrap resampling treatsthesampleas a population and
resamplestheresidualswithreplacement a specifiednumberof times15
(Mooneyand Duval 1993;Stine1990).Equation4e is basedon residual
resampling using1,000replications.16 The parameter estimates and stan-
darderrorsbasedontheresidual resampling areveryclosetothoseobtained
withOLS.
Thesecondrobustness checkis concerned withtheeffect ofotherpos-
sibleexplanatoryvariablesonourparameter estimates.
Inordertoascertain
whether theseparameter estimatesare"robust"toalterations inthecondi-
tioningsetofinformation, we followLevineandRenelt(1992) andinclude
a setof variablesin Model4 anddetermine whetherthese"conditioning
alterthecoefficients
variables"significantly orstandard errorsonourvari-
ablesofinterest.
Theconditioning variablesweuse,as suggested byLevine
andRenelt(1992),arethegrowth rateofdomestic thestandard
credit, devi-
ationof domesticcreditgrowth, theaverageinflation rate,thestandard
deviationoftheinflation rate,thegrowth in government consumption ex-
theaveragenumber
penditure, ofrevolutions andcoups,anda dummy vari-
able indicating
exportorientation."7
The findings,notreported here,indicatethatwhilethecoefficient on
thefouritemindexofachievement motivation doesdecreaseto 1.73,itis
(t = 2.8).18 In short,we have a greatdeal of
still statisticallysignificant
in theparameter
confidence estimates
andstandard in Model4.
errors
Conclusion
Theideathateconomicgrowthis partly
shapedbyculturalfactorshas
encountered resistance.
considerable is be-
One reasonforthisresistance
"Notethatthereis a significant
difference
between resamplingwithrandom regressors
andresampling withfixedregressors.
We resample becausewe haveassumedthat
residuals
regressorsin ourmodelarefixedin repeatedsamples.See Stine(1990) fora discussion.
'6Duetothesmallsamplesize,we "fattened" bydividing
theresiduals by((1 -k)ln)
We also used 10,000replicationsandfoundalmostidentical results.
"We couldnotincludeChinainthisexercisesincedataformostoftheseconditioning
variablesarenotavailable.
'8Thespecificsofthissensitivity as wellas theobtained
analysis, coefficient
estimates
andstandard areavailablefromtheauthors
errors, uponrequest.

causecultural valueshavebeenwidelyperceived as diffuse andpermanent

featuresofgivensocieties:ifcultural valuesdetermine economicgrowth,
thentheoutlookforeconomicdevelopment seemshopeless,becausecul-
turecannotbe changed.Another reasonforopposition is thatstandardeco-
nomicarguments supposedly suffice forinternational differences insavings
andgrowth rates.Forexample,thestandard lifecyclemodelandnotcul-
turalarguments explainsthedifference in savingsratesand growth rates
between, say,Germany, Japan,andtheUnitedStates.19
Whenwe approach culture as something tobe measured ona quantita-
tiveempirical basis,theillusionofdiffuseness andpermanence disappears.
We no longerdealwithgrossstereotypes, suchas theideathat"Germans
havealwaysbeenmilitaristic," or "Hispanicculture is unfavorable to de-
velopment." Wecanmovetotheanalysisofspecific components ofa given
cultureat a giventimeandplace.Thus,we findthat,from1945to 1975,
WestGermanpoliticalcultureunderwent a striking transformation from
beingrelatively authoritariantobecoming increasingly democratic andpar-
ticipant(Baker,Dalton,and Hildebrandt 1981). And we findthat,from
1970to 1993,theUnitedStatesanda number ofWestEuropeansocieties
experienced a gradualintergenerational shiftfromhavingpredominantly
materialisttowardincreasingly postmaterialist valuepriorities (Abramson
andInglehart 1995).Thoughthesechangeshavebeengradual, theydemon-
stratethatcentralelements ofculture can anddo change.
Furthermore, empirical research canhelpidentify specificcomponents
ofculturethatare relevant to economicdevelopment. One neednotseek
to changea society'sentirewayoflife.The present findings suggestthat
one specificdimension-achievement motivation-ishighlyrelevant to
economicgrowth rates.In theshort run,tochangeevena relatively narrow
andwell-defined culturalcomponent suchas thisis noteasy,butitshould
be fareasierthanattempting tochangean entire culture. Furthermore, em-
piricalresearchdemonstrates thatculturecan and does change.Simply
makingparents, schoolsand otherorganizations awareof thepotentially
relevantfactors, maybe a stepin therightdirection.
We findthateconomictheoryalreadyis augmented with"social
'9Inthepost-World WarII period,thelifecyclemodelarguesthatsinceJapanand

Germany hada substantial
portion oftheir
capitalstockdestroyed,the"permanent income"
ofthepopulation wasgoingtobe lessthanwasexpectedattheonsetofthewar.Thelower
ratiocontributes
capital-labor to lowerrealwagesandhigher rates.In response
interest the
publicraiseditssavingsrateto "smooth"itspostretirement income.TheUnitedStates,on
theotherhand,saw a significant increasein itscapitalstock-as a resultofthewar.This
hadtheoppositeeffectsincethehigher capital-labor
ratiodepressesinterestratesandraises
realwages.Thepublic'ssavingsratefallsinthiscase since"permanent income"increases,
whilecurrent consumption rises.

626 JimGranato,RonaldInglehart,
and David Leblang
norms"and"cultural"factors (Cole,Malaith,andPostlewaite 1992;Elster

1989;Fershtman andWeiss1993).Wherewouldcultural valuesfittheoreti-
callyin growth models?The economicsliterature is repletewithmodels
of savingsbehavior thatfocuson the"lifecycle" and,morespecifically,
thebequestmotive.Culturalvariablesmatter here.Sincesavingsand in-
vestment behavior holdsanimportant placeingrowth models,a determina-
tionofhowcultural andmotivational factorscanbe usedtoaugment these
existingeconomicmodels,it seemstous,is thenextsteptouncovering a
betterunderstanding ofeconomicgrowth.20
In theend,however, thesearguments canonlybe resolvedon theem-
piricalbattlefield.
We useordinary leastsquaresregressiontotesteconomic
andcultural modelsofgrowth on a crosssectionof25 countries. We find
thateconomicandcultural factors
affectgrowth.Theencompassing princi-
pleis usedtoresolvethesecompeting theoreticalspecifications
andtogen-
eratea finalparsimonious model.Theencompassing resultsshowthatboth
modelsexplainaspectsofgrowth thattheothercannot.Therobustness of
theseresultswerefurther validatedusinga variantof Leamer's Extreme
BoundsAnalysis(EBA) andnonparametric methods including robustre-
gressionandbootstrap resampling.
Theresultsinthisarticledemonstrate thatbothcultural andeconomic
arguments matter.Neither supplantstheother.
Future theoreticalandempir-
icalworkis betterservedbytreating these"separate"explanations as com-
plementary.
Manuscriptsubmitted3 March 1995.

Final manuscriptreceived25 August1995.
20Institutional
factors
suchas regime
typeandproperty rights
havealsobeensuggested
as importantdeterminants
ofeconomicgrowth (Helliwell1994;Leblang1996).

APPENDIX
We outlinethe F-testforparameterencompassingbelow. Proceduresshould be

reversedto show Model 2 4 Model 1.
The F-test
To show thatModel 1 4 Model 2 do the following:
Al) Estimatethe joint specificationof Model 1 and Model 2 below. Save the
"unrestricted" residual sum of squares (RSSu):
Y = aX + FZ + j*
A2) Estimatea "restricted"regressionthatsets F = 0. Save the "restricted"

residuals(RSSr).
A3) Calculate the F-testbelow.
(RSSr - RSSu)lk2F(k2, n - k)
RSSul(n - k)
A null findingindicatesthatModel 1 4 Model 2.

APPENDIX
Table 1. Data Used in EconomicGrowthRegressions
Country Growtha GDPb PrimarycSecondarydInvestmenteFourItem' Postmaterialg
Austria 3.141 3.908 1.05 0.5 0.24373 0.46 2.11
Belgium 3.0639 4.379 1.09 0.69 0.19595 0.22 2.02
Brazil 3.2383 1.313 0.95 0.11 0.20599 -0.32 1.67
Canada 3.0608 6.069 1.04 0.52 0.201 0 2.14
China 5.5 0.567 0.75 0.41 0.20163 0.9 1.36
Denmark 2.4935 5.49 1.03 0.65 0.21627 0.2 1.99
Finland 3.5184 4.073 0.97 0.74 0.25217 0.38 2.23
France 2.9729 4.473 1.44 0.46 0.2224 0.09 2.04
Germany 2.7082 5.217 1.33 0.53 0.20923 0.52 2.14
GreatBritain 2.1637 4.97 0.95 0.67 0.15317 -0.01 2
India 1.9398 0.533 0.61 0.2 0.19982 -0.46 1.58
Ireland 2.9652 2.545 1.1 0.35 0.22252 -0.44 1.96
Italy 3.5253 3.233 1.11 0.34 0.22909 -0.1 2.07
Japan 5.5539 2.239 1.03 0.74 0.31723 0.82 1.81
Korea 6.6378 0.69 0.94 0.27 0.2493 0.47 1.66
Mexico 2.26 2.157 0.8 0.11 0.20675 -0.15 1.86
Netherlands 2.3531 4.69 1.05 0.58 0.19853 0.13 2.26
Nigeria .7517 0.552 0.36 0.03 0.147 -1.24 1.67
Norway 3.551 5.001 1.18 0.53 0.29782 0.1 1.81
SouthAfrica 1.428 2.627 0.89 0.15 0.2555 -0.46 1.73
Spain 3.6954 2.425 1.1 0.23 0.22484 -0.24 1.94
Sweden 2.542 5.149 0.98 0.55 0.21237 0.5 2.09
Switzerland 1.9991 6.834 1.18 0.26 0.25747 -0.03 2.1
Turkey 2.8506 1.255 0.75 0.14 0.19792 -0.19 1.95
UnitedStates 2.0976 7.38 1.18 0.86 0.13906 -0.28 2.06
aGrowth:Growth rateofrealpercapitaGDP from1960to 1989.Source:LevineandRenelt(1992).
bGDP:The 1960valueofrealpercapitaGDP (1980 base year).Source:LevineandRenelt(1992).
cPrimary:Thenumber ofstudentsenrolled
inprimary schoolgradelevelrelative
tothetotalpopulation
ofthatage groupin 1960.Source:LevineandRenelt(1992).
dSecondary:Thenumber ofstudents
enrolledinsecondaryschoolgradelevelrelativetothetotalpopula-
tionofthatage groupin 1960.Source:LevineandRenelt(1992).
'lnvestment:
Averagefrom1960to 1989oftheratioofrealdomestic investment (private
pluspublic)
to realGDP. Source:LevineandRenelt(1992).
'Fourltem:FourItemAchievement MotivationIndexcomprisedof(Thrift+ Determination) - (Obedi-
ence + ReligiousFaith).Source:WorldValuesSurvey(1990).
Meanscoreofpostmaterialism.
8Postmatefialism: Source:WorldValuesSurvey(1990).

APPENDIX
Table 2. Diagnosticsand Case Weights
Country Rstandarda Rstudentb Cooks Distc DFFITSd RobustWe BoundWI
Austria -0.9495 -0.947182 0.019 -0.2736804 0.9344327 1
Belgium -0.2202 -0.2152011 0.001 -0.0504492 0.9968488 1
Brazil 0.0193 0.0188692 0.000 0.0078119 0.9990563 1
Canada 1.6618 1.740317 0.110 0.6955425 0.7999728 0.48882705
China 0.1369 0.1336649 0.003 0.1037808 0.98891 1
Denmark -0.1206 -0.1177412 0.000 -0.0409734 0.9985626 1
Finland 0.1797 0.1755344 0.001 0.0541456 0.9949721 1
France -1.1715 -1.182587 0.150 -0.7829324 0.9146342 0.43426483
Germany -1.8565 -1.981786 0.146 -0.8162843 0.7501034 0.41652156
Great Britain -0.1063 -0.1037699 0.000 -0.0342944 0.998759 1
India -1.0073 -1.007698 0.055 -0.4691486 0.9372047 0.72471705
Ireland 0.2852 0.278949 0.005 0.1360241 0.9923928 1
Italy 0.5576 0.5483339 0.007 0.1696072 0.9692861 1
Japan 0.6540 0.6448598 0.027 0.3214996 0.946776 1
Korea 2.6414 3.154506 0.417 1.54147 0.4755642 0.22056864
Mexico -0.8926 -0.8881056 0.015 -0.2475407 0.9416614 1
Netherlands -0.6922 -0.6834447 0.007 -0.1679137 0.9590508 1
Nigeria 0.3351 0.3279361 0.029 0.3349275 0.994396 1
Norway 0.9402 0.9375238 0.017 0.2619067 0.9240916 1
South Africa -1.2503 -1.268319 0.039 -.4015923 0.8742813 0.84662985
Spain 0.7417 0.7335156 0.024 .3084299 0.9493236 1
Sweden -0.9887 -0.9882439 0.049 -0.4408704 0.9296764 0.77120165
Switzerland 0.1883 0.1839838 0.002 0.0804552 0.9991782 1
Turkey -0.3363 -0.3291797 0.003 -0.1098775 0.9953059 1
UnitedStates 1.5000 1.549266 0.176 0.8659872 0.8714244 0.39261552
aStandardizedResiduals
IStudentizedResiduals
CCook's Distance
dDFFITS
eRobust RegressionWeights
'Bounded-InfluenceWeights
REFERENCES
Abramson,Paul, and Ronald Inglehart.1995. Value Change in Global Perspective.Ann
of MichiganPress.
Arbor:University
and Using Regression.BeverlyHills: Sage.
Achen, Christopher.1982. Interpreting
Alwin,DuaneF. 1986."ReligionandParental
Child-Rearing Evidenceof a
Orientations:
Catholic-Protestant
Convergence." AmericanJournalof Sociology 92:412-40.
Baker,KendallL., RussellDalton,andKai Hildebrandt.
1981.GermanyTransformed.
Cam-
bridge:Harvard UniversityPress.
Barro,Robert.1991."EconomicGrowth ina CrossSectionofCountries."
Quarterly
Jour-
nal of Economics 106:407-44.
andRobert
Bollen,Kenneth, Jackman.
1985."Regression AnExpository
Diagnostics: Treat-
mentof Outliersand InfluentialCases." Sociological Methodsand Research 13:510-
42.

Chatterjee,S., and A. S. Hadi. 1988. Sensitivity

Analysisin Linear Regression.New York:
JohnWiley.
andAndrew
Cole,Harold,GeorgeMalaith, 1992."Social Norms,SavingsBe-
Postlewaite.
havior,and Growth." Journalof Political Economy 100:1092-125.
Cook, R. D., and S. Weisberg.1982. Residuals and Influencein Regression.London: Chap-
manandHall.
Davidson,Ronald,andJamesMacKinnon.1991."SeveralTestsforModelSpecification
in
thePresenceof Alternative Econometrica
Hypotheses." 49:781-93.
Jon.1989."Social NormsandEconomic
Elster, Theory."JournalofEconomicPerspectives
3:99-117.
Fershtman,
Chaim,andYoramWeiss.1993."Social Status,Culture,
andEconomicPerfor-
mance." The Economic Journal 103:946-59.
Fox, John.1991. RegressionDiagnostics:An Introduction.
Sage University
Paper on Quanti-
tativeApplicationsin theSocial Sciences,07-079.Newbury
Park,CA: Sage.
Godfrey,Leslie. 1984. "On the Uses of Misspecification
Checksand Tests of Non-
nestedHypotheses inEmpirical Econometrics."EconomicJournal
Supplement 96:69-
81.
Granato,
Jim,andMotoshiSuzuki.N.d."The Use oftheEncompassing toResolve
Principle
Empirical Controversiesin VotingBehavior:An Application
to VoterRationality
in
Congressional Elections."ElectoralStudies.Forthcoming.
Hamilton,Lawrence. 1992. RegressionwithGraphics: A Second Course in AppliedStatis-
tics.PacificGrove,CA: Brooks/Cole
Publishing.
Harrison,Lawrence E. 1992. Who Prospers? How Cultural Values Shape Economic and
Political Success. New York: Basic Books.
J.F. 1994."Empirical
Helliwell, LinkagesBetweenDemocracy
andGrowth."
British
Jour-
nal of Political Science 24:225-48.
David,andJean-Francois
Hendry, Richard.1989."RecentDevelopments
intheTheoryof
to OperationsResearchand Econometrics:TheXXth
Encompassing." In Contributions
ofCORE,ed. B. CornetandH. Tulkens.Boston:MIT Press.
Anniversary
Ronald.1971. "The SilentRevolution
Inglehart, in Europe."American
PoliticalScience
Review4:991-1017.
Inglehart,Ronald. 1977. The Silent Revolution: Changing Values and Political Styles.
Princeton:
Princeton Press.
University
Inglehart,Ronald. 1990. CultureShiftin AdvancedIndustrialSociety.Princeton:Princeton
Press.
University
Robert.1987."The PoliticsofEconomicGrowth
Jackman, inIndustrialized
Democracies,
1974-1980." Journalof Politics 49:242-56.
Leamer,Edward.1983."Let's Takethe'Con' OutofEconometrics."
American
Economic
Review 73:31-43.
Leblang,David. 1996. "Property and EconomicGrowth."Political
Rights,Democracy,
Research Quarterly.49:5-26.
Lenski,Gerhard.
1963.TheReligiousFactor.New York:Anchor-Doubleday.
Levine,Ross,andDavid Renelt.1992."A Sensitivity
Analysisof Cross-Country
Growth
Regressions." AmericanEconomic Review 82:942-63.
Lucas,Robert.1988."On theMechanics
ofEconomicDevelopment."
Journal
ofMonetary
Economics 1:3-32.
N. Gregory,
Mankiw, andDavidWeil.1992."A Contribution
DavidRomer, totheEmpirics
of Economic Growth." QuarterlyJournalof Economics 152:407-37.
David. 1961.TheAchieving
McClelland, Society.Princeton:
Van Nostrand.

McClelland,David.et al. 1953.TheAchievementMotive.New York:Appleton-Century-

Crofts.
Mizon,Grayham, andJean-Francois Richard.1986."The Encompassing andits
Principle
to Non-nested
Application Tests."Econometrica
Hypothesis 54:657-78.
Mooney, Christoper,and RobertDuval. 1993. Bootstrapping:A NonparametricApproach
Inference.
to Statistical PaperSerieson Quantitative
Sage University Applications
in
theSocial Sciences,07-095.NewburyPark,CA: Sage.
Romer,Paul. 1990. "EndogenousTechnologicalChange."JournalofPoliticalEconomy
98:71-102.
Solow,Robert.1956."A Contribution
totheTheoryofEconomicGrowth." Jour-
Quarterly
nal of Economics 70:65-94.
Stine,Robert.1990."An Introduction Methods."SociologicalMethodsand
to Bootstrap
Research 18:243-91.
andCapitalAccumulation."
Swan,Trevor.1956."EconomicGrowth Record22:
Economic
334-61.
Tawney, Richard. 1926. Religion and the Rise of Capitalism:A History.Gloucester,MA:
P. Smith.
Tawney, [1922]1955.TheAcquisitive
Richard. Society. NewYork:Harcourt,
Reprint. Brace.
Welsch,Roy.1980."Regression AnalysisandBounded-Influence
Sensitivity Estimation."
In Evaluationof EconometricModels, ed. JanKmentaand JamesRamsey.New York:
AcademicPress.


The Effect of Cultural Values On Ec - Dev, - Granato

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Effect of Cultural Values On Ec - Dev, - Granato

Uploaded by

Copyright:

Available Formats

The Effect of Cultural Values on Economic Development: Theory, Hypotheses, and Some

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Theory:Cultural variablesareincorporatedintoa baselineendogenous economic

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Culturalfactorsalone do notexplainall of thecross-national variation

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

nomicdevelopment.This literature emphasizestheimportanceof motiva-

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

by littleor no economicgrowthwhichimpliesthatupwardsocial mobility

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

McClelland's work is criticizedon variousgrounds.It is questioned

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Figure1. Economicgrowthrate by achievement

Brazil Italy o 0 utl

z & New World

-100 -75 -50 -25 0 + 25 + 50 + 75 + 100

heavilythanobedience.The threeEast Asian societiesrankhigheston that

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

other-obedience and religiousfaith-tend to discourageit,emphasizing

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

dimensionseemsto tapthetransition frompreindustrialto industrialvalues

2Thisresultwas basedon an assumption

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Romer(1990), arguesthatResearchand Development(R&D) spendingis

Yj=I 1,O + HX + [1]

whereYiis outputgrowth(per capita) forcountryi, Ii,0is a setof economic

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

fiedby Levine and Renelt(1992) as havingrobustpartialcorrelations with

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Table 1. OLS EstimationofEconomicGrowthModels

variable:3.04;N is 25 forall models;Standard

theculturalmodel (Model 2), but it is moredesirableto implementa re-

6Ifmodelsunderconsideration orifa newmodelsimplyaddsoneormore

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

A progressiveresearchstrategyrequiresthatthe test statisticsem-

'This is notto saythatgoodness-of-fit andattendant

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

significantlyalterstheparameter estimates andstandard errors on second-

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Figure2. PartialRegressionPlot ofAchievement

The firstrobustness testsexaminetheinfluence ofindividualcases on

9We also examinepartialregression plotsfortheeffectof percapitagrossnational

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

Table 3. Diagnosticson Model 4

nosticsprovidemoreconcrete evidencethatKoreais an outlier.It has a

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

OLS results from Model4 inTable1 witha variety ofestimation strategies.

'3Evenifonesubscribes tothisstrategy thereis evidencethatModel4 performs quite

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

causecultural valueshavebeenwidelyperceived as diffuse andpermanent

'9Inthepost-World WarII period,thelifecyclemodelarguesthatsinceJapanand

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

norms"and"cultural"factors (Cole,Malaith,andPostlewaite 1992;Elster

Manuscriptsubmitted3 March 1995.

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

We outlinethe F-testforparameterencompassingbelow. Proceduresshould be

A2) Estimatea "restricted"regressionthatsets F = 0. Save the "restricted"

A null findingindicatesthatModel 1 4 Model 2.

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM

This content downloaded from 198.11.25.98 on Sun, 17 Nov 2013 18:13:07 PM