beyond country image favorability: how brand positioning via ......2010), havaianas closely...

Beyond country image favorability: How

brand positioning via country personality

stereotypes enhances brand evaluations

Peter Magnusson,Stanford A. Westjohn andNancy J. Sirianni

Culverhouse College of Business, University of

Alabama, Tuscaloosa, AL 35487, USA

Correspondence:P Magnusson, Culverhouse College ofBusiness, University of Alabama, Tuscaloosa,AL 35487, USA.Tel: 205-348-8932;Fax: 205-348-6695;e-mail: [email protected]

AbstractAcross four lab experiments and a field study, we find that brands are evaluated

more favorably when the brand is positioned in a manner that is congruent

with the brand’s home country personality stereotype than when brandpositioning is incongruent. Results demonstrate that cultural authenticity

mediates this effect. We also uncover a moderating effect whereby brands are

viewed more favorably when brand positioning and country personalitystereotypes are incongruent, rather than congruent, under conditions of

consumer animosity toward the firm’s home country. This study is the first to

demonstrate how international firms can leverage country-of-origin personalitystereotypes as a brand-building advantage.

Journal of International Business Studies (2019) 50, 318–338.https://doi.org/10.1057/s41267-018-0175-3

Keywords: country personality; country of origin; international consumer behavior;cultural authenticity; consumer animosity

Havaianas of Brazil has emerged as the world’s largest casual sandalbrand, selling about 200 million pairs of flip flops each year (Baker,2017). Despite Brazil’s generally weak country image (Fetscherin,2010), Havaianas closely associates its brand with its Brazilianorigin and employs common stereotypes about Brazilian peopleand values in its positioning, including the use of vibrant colors,Carnival, joy, and fun. Such a brand positioning strategy contra-dicts common guidance in the international marketing literature,which suggests that only brands from countries with a strongcountry image should emphasize country association; whereasbrands from countries with a weak country image should avoidcountry association (e.g., Herz & Diamantopoulos, 2017; Kumar &Steenkamp, 2013).

Havaianas’ success associating itself closely with Brazil highlightsa gap in the understanding of country-of-origin (COO) effects.Specifically, the extant literature fails to account for how strategicbrand positioning can help firms circumvent a weak COO image orbetter leverage a favorable image. To remedy this issue, we posit thatother COO-related associations beyond country image favorability

Received: 16 May 2017Revised: 20 June 2018Accepted: 30 July 2018Online publication date: 12 September 2018

Journal of International Business Studies (2019) 50, 318–338ª 2018 Academy of International Business All rights reserved 0047-2506/19

www.jibs.net

must be accessed. We suggest that a brand can beeffectively positioned to associate with its homecountry’s personality stereotype. Country personal-ity stereotypes, defined as ‘‘the mental representa-tion of a country on dimensions that typicallycapture an individual’s personality’’ (d’Astous &Boujbel, 2007: 233), differ from general countryimage favorability since they are not typicallyperceived as inherently good or bad. For example,country personality stereotypes include extrover-sion versus introversion (Rojas-Méndez, Papado-poulos, & Alwan, 2015) and warmth versuscompetence (Cuddy et al., 2009).

We posit that a brand’s positioning, defined asthe use of promotional messaging to establish keyassociations in customers’ minds (Heinberg,Ozkaya, & Taube, 2017), will have positive effectson consumers’ brand evaluations when it is con-gruent with its COO personality stereotype (here-after referred to as ‘‘brand positioning-countrypersonality congruity’’). Importantly, we contendthat brand positioning-country personality con-gruity will result in positive effects for brands fromcountries with both favorable and unfavorableimages. Thus, we expect that a brand positioning-country personality congruity strategy can helpinsulate, or circumvent, problems associated withan unfavorable country image.

Our theory of brand positioning-country person-ality congruity is founded in schema congruitytheory (Heider, 1946; Mandler, 1982). Using thisperspective, we offer the following contributions tointernational marketing practice and theory. First,recent studies highlight consumers’ interest inreading product labels and using brand origin as apurchase driver (Gürhan-Canli, Sarıal-Abi, &Hayran, 2018). Consequently, it is important forinternational marketing managers and researchersto understand how and why COO associations—specifically COO personality stereotypes—can bestrategically leveraged to enhance brand perfor-mance. This study demonstrates how brand posi-tioning-country personality congruity isadvantageous—even for brands from countrieswith unfavorable images. This is a notable theoret-ical contribution because existing theory suggeststhat brands associated with countries with unfa-vorable images should de-emphasize their COOassociation. However, our findings suggest analternative strategic path forward through brandpositioning-country personality congruity.

Second, we uncover the underlying mechanismthat drives the relationship between brand posi-tioning-country personality stereotypes andenhanced brand evaluations. We posit that culturalauthenticity, defined as the consumer’s perceptionthat the brand is culturally genuine (Southworth &Ha-Brookshire, 2016), helps explain our proposedeffects. We predict that when a brand is positionedin a manner congruent (incongruent) with con-sumers’ commonly held country personality stereo-types, consumers will perceive the brand as more(less) culturally authentic, and this process willresult in more (less) positive brand evaluations. Theidentification of this mediating process is animportant theoretical contribution because itenhances our understanding of how schema con-gruity operates in the context of country associa-tions and consumer brand evaluations.

Third, we investigate a condition under whichcountry personality stereotype incongruity, ratherthan congruity, is the preferred brand positioningstrategy. Specifically, we examine the moderatinginfluence of consumer animosity, defined as ‘‘anger[toward a specific country] related to previous orongoing political, military, economic, or diplo-matic events’’ (Klein, 2002: 346). We posit thatwhen consumers hold feelings of animosity towardthe target country, brands may actually benefitfrom distancing themselves (via incongruity) fromtheir COO personality stereotype. This illustratesan important difference between a merely unfavor-able country image and overtly negative animositytoward a country. This is a unique and importantcontribution as it forges a theoretical and manage-rially valuable connection between the consumeranimosity and brand positioning literature streams.

Finally, experimental research designs, whichhave been rare in IB research, are now beingencouraged because of their ability to isolate causalvariables, and when combined with field studies inrealistic settings, offer convincing evidence (Zell-mer-Bruhn, Caligiuri, & Thomas, 2016). In Study 1,we test the main thesis (H1) with a preliminary fieldstudy and investigate the mediating effect ofcultural authenticity (H2) across two laboratoryexperiments. In Study 2, we examine the moderat-ing effect of consumer animosity (H3) across twodifferent country contexts in two separate experi-ments. Thus, we build evidence to support ourhypotheses across a series of five experiments thattest our predictions across a variety of countrypersonality stereotypes and across countries withvarying levels of overall favorability. Across our

Beyond country image favorability Peter Magnusson et al

319

Journal of International Business Studies

studies, we examine two related brand evaluationconstructs, which capture different facets of abrand’s operational performance in the consumers’mindset (Katsikeas, Morgan, Leonidou, & Hult,2016). We examine brand evaluations in the formof quality ratings of a product following a taste test(Veale & Quester, 2009) and we examine customer-based brand equity, defined as ‘‘the differentialeffect of brand knowledge on consumer response tothe marketing of the brand’’ Keller (1993: 1). Byexamining two distinct, yet related, outcome mea-sures, we extend the generalizability of the concep-tual framework. Given that both are relatedcustomer mindset performance constructs (Kat-sikeas et al., 2016), we expect the effects of brandpositioning-country personality congruity will holdin the same pattern for each construct.

We proceed by providing a review and concep-tual background on country image and personality,as well as schema congruity theory. Then, wedevelop and present hypotheses, which are sup-ported with evidence from five experiments. Weconclude by discussing the theoretical and man-agerial implications, and identifying limitationsand future research directions.

CONCEPTUAL FOUNDATIONDichter (1962: 116) first suggested that ‘‘the littlephrase ‘Made in [y]’ can have a tremendousinfluence on the acceptance and success of prod-ucts.’’ Since then, it has received significantresearch attention; we summarize influential arti-cles that have advanced the field in Table 1.Following the early realization that country asso-ciations can influence brand evaluations, a vastbody of literature has found corroborating evi-dence of a significant link between consumers’perceptions of a country (i.e., country image) andtheir evaluations of products associated with thatcountry, highlighted by several notable reviewarticles (e.g., Bilkey & Nes, 1982; Verlegh &Steenkamp, 1999). A consistent conclusion acrossthe literature is that COO ‘‘has been observed toinfluence people’s attitudes, even when subjectsare given a chance to see, touch, feel or taste thevery same physical product’’ (Kotler & Gertner,2002: 252). Research suggests that these effects arepowerful and hold when consumers attribute thebrand to the wrong home country (Magnusson,Westjohn, & Zdravkovic, 2011) and when con-sumers deny relying on COO information (Herz &Diamantopoulos, 2017).

Country ImageGiven the role of country associations in influenc-ing consumers’ brand evaluations, it is not surpris-ing that significant research efforts have tried togain a better understanding of country image. Afundamental assumption underlying countryimage is that consumers form stereotypes aboutcountries (e.g., Shimp, Samiee, & Madden, 1993;Verlegh & Steenkamp, 1999). Country stereotypesreflect a person’s perceptions about the features of acountry and are developed through socializationprocesses and exposure to information about coun-tries (Diamantopoulos, Florack, Halkias, & Palcu,2017). Accordingly, country image is a schema, or aknowledge structure, that synthesizes what weknow about a country and it has been defined as‘‘the total of all descriptive, inferential and infor-mational beliefs one has about a particular coun-try’’ (Martin & Eroglu, 1993: 193). Even thoughthere seems to be agreement in viewing countryimage associations as a schema of stored knowl-edge, some researchers have viewed it as a broad,overall construct, based on ‘‘representative prod-ucts, national characteristics, economic and polit-ical background, history, and traditions’’(Nagashima, 1970: 68). Other researchers havefocused on a narrower product country imageemphasizing consumers’ perceptions about thequality of products made in a particular country(Roth & Romeo, 1992).

However, regardless of the breadth of conceptu-alizations of country image, researchers have gen-erally viewed this construct from a positive–negative valence perspective. For example, productcountry innovativeness measures have ranged fromvery innovative to not innovative, workmanshipfrom high to low quality (Roth & Romeo, 1992),and the macro image of a country has beenevaluated as, for example, more or less economi-cally developed and with higher versus lower levelsof technological research capacity (Martin & Ero-glu, 1993). Whereas the country image favorabilityperspective has contributed valuable insights intothe understanding of COO effects, it fails toaccount for country stereotype associations thatare not always viewed as more or less favorable.Specifically, country personality associations can bepositive, negative, or neither, but they are animportant part of consumers’ country schema.

Country PersonalityA growing body of evidence indicates that con-sumers spontaneously associate countries with


320


human personality traits as part of their mentalschema (e.g., Barbarossa, De Pelsmacker, & Moons,2018; Kim, Shim, & Dinnie, 2013; Rojas-Méndez,Murphy, & Papadopoulos, 2013). Using humantraits to stereotype countries is appealing, because‘‘human traits are associated with universal mentalrepresentations that are easily activated becausethey have been used previously to characterizepeople in our environment and predict theirbehavior’’ (d’Astous & Boujbel, 2007: 232).

Furthermore, unlike favorability perceptions,which can often be contained to specific productcategories (Pappu, Quester, & Cooksey, 2007; Roth& Romeo, 1992), country personality stereotypesrepresent broad macro-level associations of acountry.

With the realization that country personalitystereotypes may be a salient country association,researchers have aimed to identify the dominantcountry personality dimensions that people use to

Table 1 Overview of influential articles in country-of-origin and country personality literatures

Author(s) (year) Context Key findings/conceptualizations

Representative studies on how country image favorability influence brand evaluations

Schooler (1965) 200 Guatemalan part-time students First empirical assessment illustrating that products from

Mexico and Guatemala are preferred over products from El

Salvador and Costa Rica

Bilkey and Nes

(1982)

Qualitative review of 11 COO articles, published

1965–1974

Review of findings confirm that COO does indeed influence

product evaluations

Verlegh and

Steenkamp

(1999)

Quantitative meta-analytic review of 41 studies

(278 effect sizes), published 1980–1996

Findings confirm that COO influences product evaluations.

Effects are stronger for, for example, quality judgments over

attitude and purchase intentions, and COO effects are

stronger when comparing developed with less developed

country products

Representative studies defining country associations as a schema

Shimp et al.

(1993: 323)

64 in-depth interviews with American consumers ‘‘Cognitive structure connotes consumers’ encoded

representations of information in memory, which, in present

usage, refers to what consumers know/believe about

products from different countries’’

Martin and

Eroglu (1993:

193)

200 American students Country image is defined as ‘‘the total of all descriptive,

inferential and informational beliefs one has about a

particular country’’

Pappu et al.

(2007: 727)

539 Australian consumers Country image (similar to brand image) is a set of COO

associations organized into groups in a meaningful way

Representative studies focusing on fit between product category and country image

Roth and Romeo

(1992)

368 US, Mexican, and Irish consumers COO influence varies depending on how well the country’s

perceived production and marketing strengths are related to

the product category

Pappu et al.

(2007)

539 Australian consumers COO effect depends on match between country’s macro and

micro country image and product category

Representative studies identifying country personality dimensions

d’Astous and

Boujbel (2007)

170 French-speaking Canadians evaluated 11

different countries.

First country personality effort. Identified (1) agreeableness,

(2) wickedness, (3) snobbism, (4) assiduousness, (5)

conformity, and (6) unobtrusiveness

Cuddy et al.

(2009)

755 European evaluated 15 EU countries Warmth and competence are negatively correlated.

Stereotypes of countries are either high warmth, low

competence, or high competence, low warmth

Kim et al. (2013) 1221 university students from nine countries Identified country personality dimensions of (1) leadership,

(2) excitement, (3) sophistication, (4), tradition, and (5)

peacefulness

Rojas-Méndez

et al. (2015)

561 Saudi Arabian consumers Evaluated the US based on the big five personality

dimensions

This study Five experiments in the US Extends literature by demonstrating how positioning brands

via COO personality stereotypes signals cultural authenticity

and enhances consumer brand evaluations


321


stereotype countries. Several broad frameworkswith multiple dimensions, akin to the Big Fivepersonality framework (John & Srivastava, 1999),have been proposed. For example, d’Astous andBoujbel (2007) developed a framework based on sixdimensions (agreeableness, wickedness, snobbism,assiduousness, conformity, and unobtrusiveness).Kim et al. (2013) found five personality dimensions(leadership, excitement, sophistication, tradition,and peacefulness), and Rojas-Méndez et al. (2013)identified three dimensions (amicableness,resourcefulness, and self-centeredness).

Others have developed frameworks that attemptto identify the most salient stereotype peopleassociate with other countries. For example, JWThompson conducted a large-scale study to under-stand how people stereotype other countries byfocusing on the (perceived) dominant personalitytrait for each country (Ayala, Truss, & Montecuollo,2012). Similarly, the stereotype content model hassuccessfully been used to categorize countries,using two primary dimensions that people use tostereotype outgroups (Fiske, Cuddy, Glick, & Jun,2002). Warmth reflects friendliness, good-natured-ness, and sincerity; whereas competence reflectscapability, confidence, and skillfulness (Cuddyet al., 2009).

Whereas multi-dimensional personality frame-works enable the development of a descriptivecountry personality profile across a number oftraits, a more useful approach with respect to brandpositioning is focusing on a salient, dominantpersonality stereotype. We argue that a single-traitfocus enables consumers to process country per-sonality imagery more clearly, and thus, createstronger associations in their existing schema forthe focal brand and country, and this processshould result in more positive brand evaluations.Similarly, Aaker’s (1997) brand personality frame-work is a rich multi-dimensional framework, yetresearch has shown that brands may only be able toleverage one dominant brand personality dimen-sion to enhance brand equity (Sirianni, Bitner,Brown, & Mandel, 2013).

Although some country personality dimensionsthat have been identified in the literature (e.g., self-centeredness (Rojas-Méndez, Papadopoulos, &Murphy, 2013) or wickedness (d’Astous & Boujbel,2007) may fit the more versus less favorable con-tinuum common in the traditional COO literature,several country personality stereotypes are notinherently positive or negative. For example, bothwarmth and competence are desirable attributes,

yet countries are typically stereotyped as predom-inantly one or the other (Cuddy et al., 2009). Ineffect, the stereotype content model suggests thatambivalent stereotypes are the norm and outgroupsare generally viewed as either competent, but cold,or warm, but incompetent, and it is unclear if oneof these is more favorable than the other. Similarly,countries can be viewed as introverted or extro-verted; however, one trait is not necessarily morefavorable than the other. Given that firms areunlikely to actively position their brands usingobviously negative traits, e.g., wickedness, thefocus of this study is on country personalitystereotypes that are not inherently valenced, suchas warmth (but incompetent) versus competence(but cold), or extroversion versus introversion.

In sum, the current state of knowledge in theCOO and country personality literatures suggeststhat (1) consumers organize their associationsabout countries into mental schemas, which influ-ence brand evaluations, and (2) beyond generalfavorability perceptions, consumers develophuman personality associations about countries.Such personality stereotypes are often not inher-ently valenced. This suggests an important researchgap and we posit that country personality stereo-types offer a strategic opportunity for brand man-agers to capitalize on COO associations in ways thatprior research based on general favorability percep-tions is unable to explain.

HYPOTHESES DEVELOPMENT

Schema CongruityConsumers organize their associations about acountry into a schema, defined as a ‘‘cognitivestructure that represents knowledge about a con-cept or type of stimulus, including its attributes andthe relations among those attributes’’ (Fiske &Taylor, 1991: 98). The schema provides a framethrough which consumers interpret new informa-tion and serves as an anchor in forming brandevaluations (Stumpf & Baum, 2016). A family oftheories in psychology, including balance theory(Heider, 1946), cognitive dissonance theory (Fes-tinger, 1957), the principles of persuasion (Cart-wright, 1949), and schema congruity theory(Mandler, 1982) emphasize the benefits of cogni-tive consistency or schema congruity.

Schema congruity is defined as a match betweenan object and the schema it invokes. Cognitiveconsistency theories suggest that consumers are


322


predisposed to prefer elements cognitively consis-tent with their existing knowledge (e.g., Festinger,1957). New information (i.e., brand positioninginformation) that is consistent with prior knowl-edge (i.e., country schema) reinforces previousstereotypes and is viewed more favorably. Thisimplies that consumers should value congruitybetween the personality associations evoked by abrand through its brand positioning and con-sumers’ country personality stereotypes. In con-trast, when the new information (i.e., brandpositioning information) is inconsistent with aperson’s prevailing cognitive structure (i.e., countryschema), consumers are more likely to reject it(Cartwright, 1949).

Related empirical evidence also supports thebenefits of cognitive congruity. For example, cor-porate social responsibility campaigns are perceivedmore favorably when there is a stronger alignmentbetween the brand and the social cause (Zdravko-vic, Magnusson, & Stanley, 2010). Anthropomor-phized products are evaluated more favorably whena human schema has been activated beforehand,creating congruity (Aggarwal & McGill, 2007), andconsumers are more likely to refer a brand to othersif there is congruity between the referral reward andthe brand (Stumpf & Baum, 2016).

Accordingly, we posit that brand evaluations willbe more favorable when brand positioning iscongruent with consumers’ country personalitystereotypes. When objects are placed in a predictiveconceptual context, this leads to a sense of famil-iarity that is associated with positive evaluations ofthe object (Whittlesea, 1993). In sum, brand posi-tioning-country personality congruity fits con-sumer expectations about brands from thatcountry, ‘‘makes sense,’’ and produces cognitiveharmony and balance that engenders favorablebrand evaluations. In contrast, brand positioning-country personality incongruity contradicts expecta-tions, produces cognitive dissonance, and imbal-ance that engenders unfavorable brand evaluations(Mandler, 1982). Thus, on the basis of this discus-sion, and under the assumption that the country’sdominant personality stereotype is not obviouslynegative, e.g., wickedness or snobbism, we proposethe following hypothesis:

Hypothesis 1: Consumer brand evaluations aremore favorable under conditions of brand posi-tioning-country personality congruity than underbrand positioning-country personality incongruity.

The Mediating Effect of Cultural AuthenticityCultural authenticity is defined as the consumer’sperception of whether a brand is culturally genuineor real (Southworth & Ha-Brookshire, 2016). Argu-ably, some firms have strategically infused countrypersonality stereotypes into their branding effortsto appear more culturally authentic; yet priorresearch does not explain why this may or maynot be a successful branding strategy. Notable ex-amples include Volkswagen’s use of German lan-guage (Das Auto) and strong German accents in itsUS advertisements, the incorporation of the kan-garoo into the brand symbol for Australia’s Qantasairline, and Ricola’s herbal lozenge commercials,which have included characters dressed in leder-hosen, scenes of the Swiss Alps, and the sound ofthe iconic Swiss Alpenhorn. While such brandingefforts are frequently practiced, there is a lack ofempirical evidence explaining any interveningmechanisms between brand positioning-countrypersonality congruity and favorable consumerbrand evaluations.

Recent research evidences that cultural authen-ticity positively affects brand attitudes (Spielmann,2016) and consumers’ willingness to try brands(Southworth & Ha-Brookshire, 2016). We build onthese findings and posit that cultural authenticityserves as a key mediating variable that helpsexplain why brand positioning-country personalitystereotype congruity leads to enhanced brandevaluations. A brand’s COO acts as a source ofcultural meaning (Alden, Steenkamp, & Batra,1999), and we argue that positioning brands toemphasize COO personality stereotypes will serveto reinforce consumers’ previously formed schemafor that country. Moreover, we propose this schemareinforcement process permits consumers to cred-ibly link the brand to the home country’s culture(Kumar & Steenkamp, 2013) and allows the brandto be perceived as more culturally authentic.Finally, because consumers have shown a prefer-ence for more authentic market offerings, wepredict that consumer brand evaluations will bemore favorable as a result of this brand positioningstrategy. In contrast, if a brand’s personality expres-sion is inconsistent with consumers’ culturalstereotype for that brand’s home country, (i.e.,brand positioning-country personality incon-gruity), we argue that the brand will be viewed asless culturally authentic, and consumer brandevaluations will be less favorable as a result.


323


Hypothesis 2: Cultural authenticity mediatesthe positive effect of brand positioning-countrypersonality congruity on consumer brandevaluations.

The Moderating Effect of Consumer AnimosityIn H1, we offered theoretical arguments suggestingthat when a brand’s personality-based positioning iscongruent with its COO personality stereotype,consumers will evaluate brands more favorably as aresult of cognitive congruity. However, a fit betweenan object (e.g., country-associated brand) and itscorresponding schema can also activate affect thattransfers to the object to the degree that it fits theschema, a process referred to as schema-triggeredaffect (Fiske, 1982). Importantly, since affect associ-ated with a schema can be just as often negative aspositive, the transferred affect could also be negativeor positive. The stronger the fit between the brandpositioning and the country schema, the more easilyschema-triggered affect will transfer to the brand(Mandler, 1982). This raises the question whether anincongruent country-personality brand positioningwould be advantageous in instances of schema-triggered negative affect.

One instance where the country schema maytrigger negative affect is in the case of consumeranimosity. Consumer animosity is defined asintense negative affect toward a specific countrydue to a perceived injustice, based on past orongoing military, political, or economic events(Klein, 2002). Consumer animosity differs fromunfavorable country image in that it is a strongnegative emotion (Harmeling, Magnusson, &Singh, 2015). In contrast, an unfavorable countryimage is generally viewed as a cognitive construct,reflecting perceptions about a country’s economicdevelopment or the quality of the products itproduces (Roth & Diamantopoulos, 2009). Signifi-cant examples of consumer animosity include thehistorical conflict between China and Japan (Gao,Wang, & Che, 2017), Egyptian animosity towardIsrael (Papadopoulos, El Banna, & Murphy, 2017),and animosity toward Russia in former Sovietrepublics, such as Lithuania and Ukraine (Ginei-kiene & Diamantopoulos, 2017). In sum, con-sumers can hold temporary or persistentanimosity feelings toward countries that otherwisehave generally favorable country images (Leonget al., 2008).

This raises questions as to how a brand shouldbest position itself if consumers harbor animositytoward the brand’s home country. We posit that

consumer animosity will moderate the effect ofbrand positioning-country personality congruitysuch that when animosity is high, incongruentbrand positioning may lead to more favorableevaluations than congruent brand positioning. Inthis sense, congruity turns into a disadvantage andincongruent brand positioning may be preferred,because negative associations with the schemainterfere with the positive effects of congruity. Ineffect, even though congruity has generally positiveoutcomes, we predict that in the case of consumeranimosity, congruity can activate negative schema-triggered affect embedded in consumers’ mentalassociations (Fiske, 1982). The intense negativeaffect caused by consumer animosity will over-power the positive experience of cognitive con-gruity, and should result in less-favorable brandevaluations (Lee & Labroo, 2004). Thus, in caseswhere consumer animosity is high, incongruentbrand positioning weakens the association betweenthe brand and its COO, and the negative schema-triggered affect will not transfer as easily (Lane &Fastoso, 2016). Conversely, when animosity is low,congruent brand positioning, for the reasons laidout in H1, should lead to more favorable brandevaluations than incongruent brand positioning.

Hypothesis 3: Consumer animosity moderatesthe brand positioning-country personality con-gruity effect. When consumer animosity is high(low), brand positioning that is incongruent withthe brand’s country personality leads to more(less)-favorable brand evaluations than whenbrand positioning is congruent with the brand’scountry personality.

STUDY 1Study 1 is comprised of a preliminary field studyand two laboratory experiments. The field studymeasured the effects of brand positioning-countrypersonality congruity on consumer wine tasteperceptions at a wine bar during regular operatinghours. The field study, which used German andSpanish wines as the study context, providedpreliminary support for H1, and served as a plat-form for the remaining experiments. The fulldetails of the field study are provided in the ‘‘WebAppendix’’.

Study 1-A largely replicated the field study with afew notable differences. First, we included a controlcondition (neutral country personality stereotypepositioning), second, we examined customer-based


324


brand equity as the outcome variable, and third, weincluded a test of the mediating effect of culturalauthenticity (H2). Study 1-B provides additionalevidence in support of H1 and H2, with threeimportant extensions. First, we examined a differ-ent product category and international brandingcontext: airlines. We chose this context in order toextend the generalizability of the framework fromtangible goods to a more experiential service offer-ing. Second, we expanded our investigation beyondGermany and Spain to two new countries: Italy andIndia. We investigated (American) consumers’ per-sonality stereotypes for the two new countries,which were found to be slightly different from thedimensions used for Germany and Spain in Study1-A. Third, because India has a significantly lessfavorable country image (Fetscherin, 2010), Study1-B allowed us to explore whether the benefits ofbrand positioning-country personality congruityalso holds for countries with less-favorable countryimages.

Study 1-A: Participants and ProceduresWe focused on wine in the preliminary field studyand in this study since it is a product closelyassociated with its home country (Veale & Quester,2009). Further, we chose Germany and Spain asfocal countries because they are both top-ten wine-producing countries of the world (UN Food andAgriculture Division, 2013), relatively similar interms of economic wealth, generally elicit favorablecountry images (Fetscherin, 2010), and are in thesame geographic region; yet Germany and Spainhave different salient country personalitystereotypes.

As a first step, we determined the salient countrypersonality stereotypes for Germany and Spain bytriangulating evidence from three sources. First, weconducted interviews with ten American con-sumers (even gender split, ages 23–63) to determineperceptions of the most salient country personalitystereotypes. Respondents were asked to describeGermany’s and Spain’s respective country person-alities. Adjectives used to describe Germany were‘‘organized, disciplined, methodical, and focused,’’whereas adjectives used to describe Spain were‘‘charming, warm, and romantic.’’

Second, we examined the JW Thompson Person-ality Atlas (Ayala et al., 2012), which asked 6075adults from 27 countries how they perceived thepersonalities of other countries. Germany waslabeled as ‘‘Uniformany’’ and described as ‘‘hard-working, successful, orderly, and disciplined.’’ In

contrast, Spain was labeled ‘‘Funlandia’’ anddescribed as ‘‘charismatic and fun.’’ Third, wereviewed Cuddy et al.’s (2009) analysis of Europeancountries, which found that Germany was per-ceived as a notably ‘‘high-competence, low-warmth’’ country, whereas Spain was perceived asa ‘‘high-warmth, low-competence’’ country.Accordingly, based on the triangulation of evi-dence, we deemed competence (Germany) andwarmth (Spain) as the two most salient descriptorsof those countries.

We recruited 174 non-student participants fromthe US (see ‘‘Appendix’’ for demographic character-istics) using Amazon’s MTurk online consumerpanel. We included a control condition to enhanceconfidence in our findings. Thus, the experimentwas a 3 (brand positioning: competent vs. warm vs.neutral-control) 9 2 (country personality stereo-type: competent vs. warm) between-subjectsdesign. Participants were randomly assigned toconditions, which meant that some participantswere exposed to a brand positioning-country per-sonality stereotype congruity condition, whereasothers were exposed to a brand positioning-countrypersonality stereotype incongruity condition, andsome received a neutral-control condition. For thecompetent and warm brand positioning condi-tions, we used the same brand positioning promo-tional brochure stimuli as in the preliminary fieldstudy (see ‘‘Web Appendix 3’’). For the controlcondition, we created a similar-looking descriptionabout each of the wines, but it emphasized brandattributes that were intended to be neither compe-tent nor warm. For example, the wine wasdescribed as ‘‘food-friendly and suitable for pairingwith a wide range of foods.’’

Participants were instructed to imagine that theyhad just entered a wine shop and were looking topurchase a bottle of white wine, and that theywould be presented with information about aspecific bottle of wine. On the next page, partici-pants viewed the promotional material for the focalwine brand for at least 30 s. The next pagemeasured the brand evaluation variables of interest.Customer-based brand equity was measured usingfour items to capture perceptions of quality versuscompetitors, value for the cost, brand uniqueness,and willingness to pay a premium for the brand(Netemeyer et al., 2004). We created a customer-based brand equity index by averaging the fourmeasures (a = 0.87). Cultural authenticity wasassessed with a three-item scale measuring percep-tions of the cultural authenticity of the brand


325


positioning. The scale is adapted from Moulard,Garrity, and Rice (2015) and included items such as‘‘the brand… (1) embodies genuinely [country]characteristics, (2) exhibited real [country] behav-iors, and (3) is authentically [country].’’ We aver-aged the three items to form a cultural authenticityindex (a = 0.92). For brand positioning competenceversus warmth, we used a three-item manipulationcheck based on (Cuddy et al., 2009). We averagedthe three items to form a competence versuswarmth index (a = 0.92).

Study 1-A: ResultsFirst, we tested the effectiveness of the manipula-tion in a 3 9 2 ANOVA with brand positioningcompetence versus warmth as the dependent vari-able. It revealed a main effect of brand positioning(MCompetent positioning = 5.50 vs. MNeutral positioning-= 4.53 vs. MWarm positioning = 3.53, F(1,

168) = 21.34, p\0.001). As expected, the maineffect of country and the interaction betweencountry and brand positioning were non-signifi-cant. Post hoc tests revealed that the neutral-control condition was perceived as significantlyless competent than the competent condition andsignificantly more competent than the warm con-dition, suggesting that the manipulation wassuccessful.

To examine H1, we conducted a 3 9 2 ANOVAwith customer-based brand equity as the depen-dent variable. As expected, the main effects ofcountry and brand positioning were not significant(MCompetent positioning = 4.91 vs. MNeutral positioning-= 4.70 vs. MWarm positioning = 4.84, F(1, 168) = 0.72,p = 0.36), (MGermany = 4.88 vs. MSpain = 4.75; F(1,168) = 0.85, p = 0.36), indicating that neithercountry nor brand positioning alone influencedthe evaluation of the brand. Importantly, in sup-port of H1, the interaction between country andbrand positioning was significant (F(1, 168) = 4.65,p = 0.01). Follow-up planned contrasts indicatedthat participants gave more favorable customer-based brand equity ratings to the German brandwith competent brand positioning (M = 5.26) thanwith neutral brand positioning (M = 4.73;t(63) = 1.91, p = 0.03) or warm positioning(M = 4.66; t(63) = 2.46, p = 0.01)1. In contrast, theSpanish brand was rated more favorably with warmbrand positioning (M = 5.02) than with neutralbrand positioning (M = 4.67; t(55) = 1.68, p = 0.05)or competent positioning (M = 4.56; t(53) = 1.67,p = 0.05). There were no significant differencesbetween the incongruent and neutral positioning

for both the German and Spanish wine brands. Weillustrate this effect in Figure 1a.

To test H2, the mediating effect of culturalauthenticity, we first conducted a 3 9 2 ANOVAwith cultural authenticity as the dependent vari-able. As expected, the main effects of brand posi-tioning and country were not significant, but theinteraction between brand positioning and countrywas significant (F(l, 168) = 13.79, p\0.001). Forthe German brand, the competent positioning(M = 5.26) was viewed as more culturally authenticthan the neutral positioning (M = 4.75;t(55) = 1.69, p = 0.05) or the warm positioning(M = 3.85; t(63) = 5.58, p\0.001). In contrast, forthe Spanish brand, the warm positioning(M = 5.31) was viewed as more culturally authenticthan the neutral positioning (M = 4.71;t(55) = 2.29, p = 0.01) or the competent position-ing (M = 4.48; t(53) = 2.71, p = 0.01), which isshown in Figure 1b.

4.66 4.73

5.265.02

4.67 4.564.50

5.00

5.50

WarmPositioning

NeutralPositioning

CompetentPositioning

WarmPositioning

NeutralPositioning


German Brand Spanish Brand

Cus

tom

er-B

ased

B

rand

Equ

ity

(Study 1-A)

3.85

4.755.25 5.31

4.71 4.48

3.50

4.50

5.50

WarmPositioning

NeutralPositioning


WarmPositioning

NeutralPositioning


German Brand Spanish Brand

Cul

tura

l Aut

hent

icity

(Study 1-A)

4.95

6.086.51

5.75

4.75

5.75

6.75

ExtrovertedPositioning

IntrovertedPositioning



Indian Brand Italian Brand

Cus

tom

er-B

ased

B

rand

Equ

ity

(Study 1-B)

6.567.51

8.24

5.155.00

7.00

9.00





Indian Brand Italian Brand

Cul

tura

l Aut

hent

icity

(Study 1-B)

A

C

D

B

Figure 1 The mediating effect of cultural authenticity.


326


Next, to establish whether cultural authenticitymediated the interactive effect of brand position-ing-country personality congruity on customer-based brand equity (H2), we conducted a mediatedmoderation analysis using 5000 bootstrap samplesand Model 8 of the PROCESS macro (Hayes, 2013).Because they were not significantly different fromone another, we collapsed the neutral and incon-gruent brand positioning conditions into one con-dition. The interaction between brand positioningand country was significantly related to culturalauthenticity (b = 1.72, t(170) = 4.49, p\0.001)and cultural authenticity was significantly relatedto customer-based brand equity (b = 0.38,t(169) = 6.95, p\ 0.001). A 95% bias-correctedbootstrap confidence interval for the indirect effectof brand positioning-country personality congruityon brand equity was wholly above zero (0.35–1.03).Therefore, consistent with H2, cultural authenticitymediated the moderated effect of brand position-ing-country personality congruity on customer-based brand equity evaluations for the Germanand Spanish wine brands (Table 2).

Study 1-B: Participants and ProcedureThe objective for Study 1-B was to extend ourinvestigation to a different branding context anddifferent country personality stereotypes. Weselected airlines, an experiential service offering,as the brand context, and chose to focus on Italyand India. Similar to Study 1-A, we first determinedwhich personality characteristics (American) con-sumers most closely associate with Italy and India,using a triangulating process. Based on qualitativeinterviews, adjectives used to describe Italy were‘‘passionate, loud, energetic, and outgoing.’’ Incontrast, adjectives used to describe India included‘‘cautious, humble, calm, and refined.’’ Combined,these descriptions suggest that consumers perceiveItaly as more extroverted and India as moreintroverted.

Following the qualitative interviews, we quanti-tatively confirmed the dominant personalitydimensions for the two countries by adapting theBig Five Inventory (BFI) (John & Srivastava, 1999)and asking respondents to assess the personality ofa country rather than that of a person. In abetween-subjects design, 150 non-student partici-pants recruited via MTurk were randomly assignedto assess the country personality of either Italy orIndia. Confirming the qualitative interviews, theANOVA revealed a significant difference in extro-version/introversion between the two countries

(MIndia = 3.23 vs. MItaly = 4.06; F(1, 148) = 57.03,p\0.001). Differences in other personality dimen-sions were modest in comparison to the differencesin extroversion/introversion. There were no signif-icant differences on the agreeableness, openness, orneuroticism dimensions, however; India was ratedas slightly more conscientious than Italy. However,the differences in extroversion/introversion wereconsiderably larger than the differences in theconscientiousness dimension, suggesting thatextroversion/introversion is the most salient coun-try personality dimension for these countries. Fur-ther, we assessed the country images of Italy andIndia, with a three-item scale (a = 0.90) adoptedfrom (Rojas-Méndez et al., 2015). Consistent withprior literature, India’s country image (M = 4.15)was significantly less favorable than Italy’s(M = 5.87; F(1, 150) = 82.62, p\ 0.001). Havingestablished extroversion/introversion as salientcountry personality stereotypes for Italy and India,we proceeded to examine whether brand position-ing that is congruent with these country personal-ity stereotypes would influence consumer brandevaluations.

We recruited 144 non-student participants (see‘‘Appendix’’ for demographic characteristics) fromthe US using MTurk and randomly assigned themto conditions. The experiment was a 2 (brandpositioning: extroverted vs. introverted) 9 2 (coun-try personality stereotype: extroverted vs. intro-verted) between-subjects design. Thus, we fullycrossed extroverted and introverted brand position-ing with extroverted and introverted country per-sonality stereotypes. We selected the nationalflagship airline for each country, Alitalia and AirIndia, to test our predictions. A professional graphicdesigner produced print advertisements that posi-tioned the airlines as either extroverted or intro-verted. The extroverted ad emphasized words suchas friendly, outgoing, and cheerful, and the intro-verted ad emphasized words such as quiet, gentle,and soft-spoken (see ‘‘Web Appendix 4’’).

The advertisements were developed through aniterative process to ensure that the stimuli success-fully conveyed the intended personality and weresimilar in likability. We pre-tested the stimuli witha small sample of non-student MTurk participants(n = 33) to examine whether they conveyed theintended position and whether there was a signif-icant difference in the attitude toward the adver-tisements. Participants were asked to identify thepositioning of the advertisement on four semanticdifferential items: (1) introverted/extroverted, (2)


327


quiet/expressive, (3) reserved/outgoing, and (4)thoughtful/talkative. We averaged the four itemsto form a perceived extroversion versus introver-sion advertising positioning index (a = 0.98). Wealso measured attitude toward the advertisementwith three items (a = 0.97) (MacKenzie & Lutz,1989). As intended, there was a significant differ-ence in the perceived extroversion versus introver-sion positioning of the ads (MExtroverted ad = 5.79,MIntroverted ad = 2.18, F(1, 31) = 82.66, p\0.001).However, there was no difference in the attitudetoward the advertisement (MExtroverted ad = 5.41,MIntroverted ad = 5.50, F(1, 31) = 0.03, p = 0.87).Thus, based on the pre-test, we were satisfied thatthe advertisements conveyed the intended posi-tioning, and that neither advertisement was

significantly more appealing than the other onthe basis of design alone.

In the full study, after viewing the advertisement,participants responded to the outcome questions aswell as manipulation check questions. Customer-based brand equity (a = 0.89) and cultural authen-ticity (a = 0.95) were assessed with the same scalesthat we used in Study 1-A. To ensure the manipu-lation was successful, we used the same four-itemscale to measure perceived extroversion versusintroversion ad positioning as in the pre-test(a = 0.97).

Study 1-B: ResultsFirst, we tested the effectiveness of the manipula-tion in a 2 9 2 ANOVA with brand positioningextroversion versus introversion as the dependent

Table 2 Summary of results for all studies

Preliminary Field Study: Germany–Spain (wine bar customers)

ANOVA results: DV—wine taste rating F p

Country (Germany/Spain) (1, 91) = 1.58 0.21

Positioning (competent/warm) (1, 91) = 0.00 0.99

Country 9 positioning (1, 91) = 7.24 0.01

Study 1-A: Germany–Spain (online panel)

ANOVA results: DV—customer-based brand equity F p

Country (Germany/Spain) (1, 168) = 0.85 0.36

Positioning (competent/warm/neutral) (1, 168) = 0.72 0.36

Country 9 positioning (1, 168) = 4.65 0.01

PROCESS results b t p

Country 9 positioning ? cultural authenticity 1.72 (170) = 4.49 \0.001Cultural authenticity ? customer-based brand equity 0.38 (170) = 6.95 \0.001

LLCI–ULCI

Mediation test—95% bootstrap confidence interval 0.35–1.03

Study 1-B: Italy–India (Online panel)

ANOVA Results: DV—customer-based brand equity F p

Country (Italy/India) (1, 140) = 5.42 0.02

Positioning (extroverted/introverted) (1, 140) = 0.48 0.49

Country 9 positioning (1, 140) = 12.93 \0.001PROCESS results b t p

Country 9 positioning ? cultural authenticity 4.04 (140) = 7.36 \0.001Cultural authenticity ? customer-based brand equity 0.34 (139) = 4.52 \0.001

LLCI–ULCI

Mediation test—95% bootstrap confidence interval 0.79–2.16

Study 2-A: Germany Animosity (students—lab)

ANOVA results: DV—chocolate taste rating F p

Animosity (high/low) (1, 156) = 0.08 0.78

Positioning (competent/warm) (1, 156) = 0.00 0.99

Animosity 9 positioning (1, 156) = 6.50 0.01

Study 2-B: Colombia Animosity (online panel)

ANOVA results: DV—customer-based brand equity F p

Animosity (high/low) (1, 139) = 6.23 0.01

Positioning (extroverted/introverted) (1, 139) = 0.02 0.89

Animosity 9 positioning (1, 139) = 5.82 0.02

DV dependent variable, LLCI-ULCI lower and upper limit confidence interval, PROCESS refers to the SPSS macro by Hayes (2013), b unstandardizedregression coefficient.


328


variable. It revealed a main effect of brand posi-tioning (MExtroverted positioning = 5.44 vs. MIntro-verted positioning = 2.24, F(1, 140) = 177.47,p\0.001). As expected, the main effect of countryand the interaction between country and brandpositioning were non-significant. To examine H1,we conducted a 2 9 2 ANOVA with customer-basedbrand equity as the dependent variable. Asexpected, the main effect of brand positioningwas not significant (MExtroverted positioning = 5.73 vs.MIntroverted positioning = 5.92, F(1, 140) = 0.48,p = 0.49). However, the main effect of countrywas significant (MItaly = 6.13 vs. MIndia = 5.52; F(1,140) = 5.42, p = 0.02), which is consistent withtraditional COO literature, as Italy generally has amore favorable country image than India(Fetscherin, 2010).

More importantly, in support of H1, the interac-tion between country and brand positioning wassignificant (F(1, 140) = 12.93, p\0.001). Follow-upplanned contrasts indicate that participants gavemore favorable customer-based brand equity rat-ings to the Italian airline brand with extrovertedbrand positioning (M = 6.51) than with introvertedbrand positioning (M = 5.75; t(74) = 2.15,p = 0.02). In contrast, for the Indian airline brand,they gave more favorable customer-based brandequity ratings with introverted brand positioning(M = 6.08) than with extroverted brand positioning(M = 4.95; t(66) = 2.89, p = 0.002). We illustratethis effect in Figure 1c.

To test H2, the mediating effect of culturalauthenticity, we conducted a 2 9 2 ANOVA withcultural authenticity as the dependent variable. Themain effects of brand positioning and country werenot significant, but as expected, the interactionbetween brand positioning and country was signif-icant (F(l, 140) = 54.24, p\ 0.001). For the Italianairline brand, the extroverted positioning (MExtro-verted positioning = 8.24) was viewed as significantlymore culturally authentic than its incongruentcounterpart (MIntroverted positioning = 5.15;t(74) = 8.78, p\0.001). In contrast, for the Indianairline brand, the introverted positioning wasviewed as more culturally authentic (MIntroverted posi-tioning = 7.51) than its incongruent counterpart(MExtroverted positioning = 6.56; t(66) = 2.23, p = 0.02),which is illustrated in Figure 1d.

Next, to establish whether cultural authenticitymediated the interactive effect of brand position-ing-country personality stereotype congruity oncustomer-based brand equity (H2), we conducted amediated moderation analysis using 5000 bootstrap

samples and Model 8 of the PROCESS macro(Hayes, 2013). Consistent with the ANOVA, theinteraction between brand positioning and countrywas significantly related to the mediating variablecultural authenticity (b = 4.04, t(140) = 7.36,p\0.001) and cultural authenticity was signifi-cantly related to customer-based brand equity(b = 0.34, t(139) = 4.52, p\0.001). A 95% bias-corrected bootstrap confidence interval for theindirect effect of brand positioning-country per-sonality stereotype congruity on customer-basedbrand equity was wholly above zero (0.79–2.16).Therefore, consistent with H2, cultural authenticitymediated the effect of brand positioning-countrypersonality congruity on customer-based brandequity for brands from both Italy and India.

Study 1: DiscussionThe three experiments in Study 1 provide strongsupport for our theory of brand positioning-coun-try personality stereotype congruity. In the prelim-inary field study, we provide initial support for H1by establishing the relationship between brandpositioning-country personality congruity and con-sumer brand evaluations, as customers evaluatedthe wine as tasting better when brand positioningand country personality stereotypes were congru-ent. Because it was conducted in a live retailenvironment, the field study offers external validityto our conceptual framework. While both countriestested (Germany and Spain) have generally a veryfavorable country image and both countries havebuilt strong reputations for producing certain vari-etals of wine, results suggest that favorability for acountry’s brands and products can be improvedthrough alignment of brand positioning and coun-try personality stereotypes. The conclusion is thatcountry personality characteristics reachingbeyond typical COO favorability can significantlyinfluence brand evaluations.

Testing the same two countries and countrypersonality stereotypes, Study 1-A provides addi-tional support for H1 in that consumer brandevaluations are more favorable when the brandpositioning is congruent with consumers’ countrypersonality stereotypes. The evidence is furtherstrengthened by the inclusion of a neutral controlcondition as the congruent condition leads to morefavorable brand evaluations over a neutral non-personality based positioning. Furthermore, Study1-A sheds initial light on why brand positioning-country personality stereotype congruity leads tomore favorable brand evaluations. In support of H2,


329


we demonstrate the underlying role of culturalauthenticity as a mediating factor. That is, whenbrand positioning is congruent with country per-sonality stereotypes, it fits with the consumers’mental schema about the country. Consistent withour theoretical predictions, congruity in brandpositioning satisfies consumers’ desire for consis-tency and harmony. The brand is viewed as moreculturally authentic, and it is evaluated morefavorably due to enhanced credibility.

Finally, the findings from Study 1-B offer specificcontributions beyond what was learned in the fieldstudy and Study 1-A. First, we extend the general-izability of our results by examining two additionalcountries and different salient country personalitystereotypes (Italy and India). Importantly, weinclude a country that has a generally less favorablecountry image. The main effect findings are con-sistent with the traditional COO literature in thatthe Indian brand was rated significantly lower thanthe Italian brand. However, when the Indian brandis positioned in a way that is congruent withconsumers’ personality stereotypes of India, thebrand’s evaluation improves significantly. This is asignificant finding, as it suggests a strategic path forbrands from countries with overall unfavorablecountry images to successfully capitalize on theirCOO. The context of our study, airline brands, alsoadds to the generalizability of our framework, sincethe findings have been consistent for both atangible good (wine) and a more experientialservice offering (airlines).

STUDY 2In this final set of experiments, we turn ourattention to H3, which investigates the moderatingeffect of consumer animosity. We conducted twoseparate experiments to test the moderating effectof consumer animosity from the perspective of USconsumers’ animosity reactions toward a countrywith a generally more favorable country image(Germany: Study 2-A) and toward a country with aless favorable country image (Colombia: Study 2-B).We selected Germany (Study 2-A) because of itspotential to be viewed as a threat to US companiesand workers. The US runs a large trade deficit withGermany and some accuse Germany of adoptingneo-mercantilist policies as evidenced by its largecurrent account surplus (Becker, 2015). Further,German brands compete directly with US alterna-tives (e.g., Adidas/Nike, Volkswagen/GM), and thecountry was the enemy of the US in both World

Wars that continue to be popularized in Hollywoodmovies and TV shows. Thus, there is a potentialbasis for consumer animosity as a result of pastwars, brand rivalries, and current economic policy.

Colombia (Study 2-B) was deemed suitable due toits close proximity and the US-Colombian TradePromotion Agreement, which makes the US animportant market for Colombian companies. Fur-ther, although the relationship between the US andColombia is generally stable, there is a history ofconflict surrounding drug cartels and the flow ofnarcotics to the US. Thus, economic competitionfrom Colombia and the role of Colombian drugcartels in supplying narcotics to the US serves as apotential basis for animosity. Lastly, we selectedchocolate as our study context across both exper-iments to provide an additional test of the gener-alizability of our findings across a new andpreviously untested product category. Both Ger-many and Colombia actively export chocolates tothe US, therefore the product context has strongexternal validity and represents a realistic con-sumer advertising situation.

Study 2-A: Participants and ProcedureWe recruited 160 student participants from a USuniversity student subject pool (see ‘‘Appendix’’ fordemographic characteristics). In a 2 (country ani-mosity: high vs. low) 9 2 (brand positioning: com-petent vs. warm) between-subjects design, we firstmanipulated consumers’ feelings of animosity byadapting the animosity prime developed by Russelland Russell (2006). Participants were randomlyassigned to either a high or low-animosity condi-tion, and read a brief story discussing details aboutUS-German trade relations. The high-animositycondition presented US-German relations as turbu-lent, filled with restrictions against American prod-ucts, and hurting the US economy. In contrast, thelow-animosity condition presented US-Germanrelations as open and friendly, including significantUS exports to Germany, which create well-payingjobs in the US and a higher quality of life forAmericans.

In a controlled lab setting, participants weresubsequently presented a cover story that a Germanchocolate brand was considering entry into the USmarket pending a favorable response by US con-sumers. Participants were randomly assigned toproduct positioning conditions that emphasizedeither the competence or the warmth of the focalproduct. In the competent brand positioning con-dition, the chocolate was presented as being


330


meticulously produced by chocolatiers who dili-gently selected each cocoa bean and who roastedeach bean using a pin-point roasting technique,with each morsel engineered to contain exactly74% cocoa. In contrast, in the warmth brandpositioning condition, the chocolate was presentedas fun and flirty, with a silky texture and decadentflavor notes, produced by chocolatiers who reliedon their playful imaginations (see ‘‘Web Appendix5’’).

After reading the product description, partici-pants were asked to sample an actual piece ofchocolate handed to them by a lab assistant. Thechocolate was identical across all conditions. It wasunwrapped and served in a clear, plastic bag so as toconceal its original branding and to fit with ourcover story. Immediately following the chocolatetasting, participants completed a brief question-naire that measured the dependent variable as wellas manipulation check questions. Similar to thefield study, we operationalized our dependentvariable of consumer brand evaluations as theparticipants’ taste evaluations. The respondentswere asked to indicate how much they liked thechocolate on a scale using 1) like-dislike, 2) veryappealing-very unappealing, and 3) very desirable-very undesirable as end points. We averaged thethree items to form a perceived taste index(a = 0.97). To examine the effectiveness of themanipulations, we included a four-item angertoward Germany scale based on Harmeling et al.(2015). We averaged the four items to form aconsumer animosity index (a = 0.90). For brandpositioning competence, we used the same three-item competence versus warmth scale as in Study1-A (a = 0.85).

Study 2-A: ResultsFirst, we tested the manipulations with a series of2 9 2 ANOVAs. The ANOVA with the brand posi-tioning competence versus warmth index as thedependent variable revealed a main effect of brandpositioning (MCompetent = 5.26 vs. MWarm = 4.88,F(1, 156) = 3.98, p = 0.05). Second, an ANOVA withconsumer animosity toward Germany as the depen-dent variable revealed a significant main effect ofthe animosity prime (MHigh animosity = 2.68 vs.MLow animosity = 2.10, F(1, 156) = 11.31, p\0.001).As expected, all other main and interaction effectsfor these ANOVAs were non-significant.

To examine H3, we conducted a 2 9 2 ANOVAwith taste evaluation as the dependent variable. Itrevealed non-significant main effects of consumer

animosity (MHigh animosity = 5.74 vs. MLow animosity-= 5.80; F(1, 156) = 0.08, p = 0.78) and brand posi-

tioning (MCompetent = 5.76 vs. MWarm = 5.78, F(1,156) = 0.01, p = 0.92). However, the interactionbetween consumer animosity and brand position-ing was significant (F(1, 156) = 6.50, p = 0.01).Consistent with the earlier experiments, follow-upplanned contrasts showed that participants in thelow animosity condition rated the German choco-late more favorably when presented in the congru-ent competent brand positioning (M = 6.08) thanwhen presented in the incongruent warm brandpositioning (M = 5.53; t(82) = 1.94, p = 0.03). Incontrast, participants in the high animosity condi-tion preferred the incongruent warm positioning(M = 6.04) over the congruent competent position-ing (M = 5.44; t(74) = 1.69, p = 0.05). The resultssupport H3 and are illustrated in Figure 2a.

Study 2-B: Participants and ProcedureAs a first step, we established consumers’ personal-ity stereotypes of Colombia. Based on qualitativeinterviews, Colombia was described as ‘‘flirtatious,enthusiastic, outgoing, and loud,’’ which was sim-ilar to the extroverted adjectives used to describeItaly. Subsequently, we asked a small sample(n = 36) of MTurk participants to evaluate Colom-bia on the Big 5 inventory, the same test used forItaly and India in Study 1-B. The findings con-firmed that Colombia was viewed as having a moreextroverted personality stereotype (M = 4.20),which was significantly more extroverted thanIndia (M = 3.23), but the same as Italy (M = 4.06).Further, on country image, Colombia (M = 4.44)was rated significantly less favorable than Italy(M = 5.87), but no better than India (M = 4.15).

Having established that Colombia is most closelyassociated with extroversion, we developed a 2(consumer animosity: high vs. low) 9 2 (brandpositioning: extroverted vs. introverted) between-subjects design. We recruited 143 non-studentparticipants (see ‘‘Appendix’’ for demographic char-acteristics) from the US via MTurk and randomlyassigned them to conditions. First, we manipulatedparticipants’ feelings of consumer animosity, withan adapted version of the animosity prime devel-oped by Russell and Russell (2006). Then, wepresented them with a cover story that a Colom-bian chocolate brand, Santander, was consideringentry into the US market. Santander is a realColombian chocolate brand, but is not well knownto the vast majority of American consumers.Participants were randomly assigned to either an


331


extroverted or introverted brand positioning con-dition. A professional advertising agency producedthe advertisements. Most attributes were held con-stant across the two advertisements, which used thesame image of the chocolate drawn from thebrand’s website, and featured the word ‘‘happi-ness.’’ To manipulate the brand positioning, theextroverted advertisement illustrated a group ofpeople at a party and with supporting text indicat-ing the chocolate would be best enjoyed withfriends. The introverted advertisement illustrated asolitary environment, suggesting the chocolate wasbest enjoyed during a quiet evening at home (see‘‘Web Appendix 5’’).

We pre-tested the ads with a small sample of non-student MTurk participants (n = 38) to examinewhether they conveyed the intended brand posi-tion and whether there was a difference in theattitude toward the ads. Participants were asked torate the ads on the same extroversion versusintroversion ad positioning scale (a = 0.92) andattitude toward the ad (a = 0.94) as we used inStudy 1-B. As intended, there was a significantdifference in the perceived extroversion versusintroversion of the advertisements (MExtroverted ad-= 6.09, MIntroverted ad = 3.32, F(1, 36) = 45.68,p\0.001). However, there was no difference inthe attitude toward the ads (MExtroverted ad = 5.91,

MIntroverted ad = 6.07, F(1, 36) = 0.37, p = 0.55).Thus, based on the pre-test, we were satisfied thatthe stimuli conveyed the intended brand position-ing and that neither advertisement was signifi-cantly more appealing than the other based on itsdesign alone.

In the full study, after viewing the advertisement,participants immediately responded to the out-come measures as well as manipulation checkquestions. Given that this was an online study, ataste test was not possible. Thus, we used the samecustomer-based brand equity measure as in Studies1-A and 1-B (a = 0.89). To examine the effectivenessof the consumer animosity manipulations, we usedthe same four-item consumer animosity scale as inStudy 2-A (a = 0.97) and the same four-item scalefor perceived brand positioning extroversion versusintroversion as in the pre-test (a = 0.96).

Study 2-B: ResultsWe again tested the effectiveness of the manipula-tions with a series of 2 9 2 ANOVAs. First, anANOVA with perceived brand positioning extro-version versus introversion as the dependent vari-able revealed a main effect of brand positioning(MExtroverted ad = 5.87 vs. MIntroverted ad = 3.32, F(1,139) = 145.40, p\0.001). Second, an ANOVA withconsumer animosity toward Colombia as thedependent variable revealed a main effect of ani-mosity prime (MHigh animosity = 4.05 vs.MLow animosity =1.49, F(1, 139) = 108.11, p\0.001). As expected, allother main and interaction effects for these ANOVAswere non-significant.

To examine H3, we conducted a 2 9 2 ANOVAwith customer-based brand equity as thedependent variable. It revealed that the maineffect of brand positioning was not significant(MExtroverted positioning = 3.85 vs. MIntroverted positioning =3.87, F(1, 139) = 0.02, p = 0.89). However, asexpected, the main effect of consumer animositywas significant (MHigh animosity = 3.61 vs. MLow animosity =4.11, F(1, 139) = 6.23, p = 0.01). More central to thepurpose of this study, the interaction betweenconsumer animosity and brand positioning wassignificant (F(1, 139) = 5.82, p = 0.02). Consistentwith the German animosity study (and Study 1),the follow-up planned contrasts revealed that inthe low animosity condition, the congruent extro-verted brand positioning (M = 4.34) led to morefavorable customer-based brand equity evaluationsas compared with the incongruent introvertedbrand positioning (M = 3.88; t(70) = 1.68, p =0.05). However, in the high animosity condition,

6.04

5.44 5.53

6.08

5.25

5.75

6.25

WarmPositioning


WarmPositioning


High Animosity Low Animosity

Cho

cola

te T

aste

(Germany)

3.35

3.86

4.34

3.88

3.00

3.50

4.00

4.50





High Animosity Low Animosity

Cus

tom

er-B

ased

B

rand

Equ

ity

(Colombia)

A

B

Figure 2 The moderating effect of consumer animosity.


332


the incongruent introverted brand positioning(M = 3.86) led to more favorable customer-basedbrand equity evaluations compared with the con-gruent extroverted brand positioning (M = 3.35;t(69) = 1.73, p = 0.04), in support of H3 and illus-trated in Figure 2b.

Study 2: DiscussionStudy 1 offers robust support for the effect of brandpositioning-country personality stereotype con-gruity on brand evaluations. Results from Study 2reveal that in a context where consumers harboranimosity toward the COO, brand positioning-country personality congruity results in a negativeeffect, instead of a positive effect on brand evalu-ation as previously demonstrated. This highlights apotential risk of country personality stereotypepositioning. The positive effect as evidenced bythe early experiments is not merely weakened, butreverses valence. Under conditions of animosity,brands are viewed more favorably when brandpositioning and country personality are incongruent.The theorized predictions were supported in twoseparate contexts, which lend further generalizabil-ity to the findings.

Further, in the Colombian context, we found amain effect of consumer animosity, which is con-sistent with the general animosity literature. Some-what surprisingly, this effect did not materialize inthe preceding German context. We can only spec-ulate as to why this occurred, but two potentialexplanations may be that the degree of animositygenerated toward Germany was somewhat modest,which may have led to the non-significant maineffect for animosity. Alternatively, the outcomevariable in that study was taste-related, which maybe somewhat less malleable than attitudinal out-comes. Nonetheless, the expected interaction withbrand positioning emerged in both the Germanand Colombian contexts, which provided supportfor our theorized prediction.

GENERAL DISCUSSIONAcross five experiments, we built conclusive evi-dence to support our theory of brand positioning-country personality congruity. We supported ourpredictions across a variety of country personalitystereotypes, countries with varying levels of overallfavorability, and different product categories. Thefield study, conducted in a retail store duringoperating hours, revealed that customers evaluatedbrands more favorably when positioned in a manner

that was congruent, rather than incongruent, withthe brand’s home country personality stereotype.Therefore, this study supported H1 and developedexternal validity for our research premise. Internalvalidity was established across the more controlledremaining studies. Studies 1-A and 1-B providedadditional tests of H1 and also revealed support forH2, the mediating effect of cultural authenticity.Study 2 demonstrated support for H3 by examiningthe moderating effect of consumer animosity. Insum, our findings shed light on an understudiedaspect of COO brand positioning, which explainhow and why brand managers can use COOpersonality stereotypes to their advantage. Thus,our research has significant implications for inter-national marketing theory and practice.

Theoretical ImplicationsMore than a half century of COO research hasprovided insight into how country associationsaffect brand evaluations. However, much of theCOO literature conceptualizes country image as ageneral favorable/unfavorable continuum. Thisstudy adds to our knowledge by examining theunder-explored dimension of country personality.As an extension of the common usage of overallfavorability attitudes in country image research,our investigation recognizes new relevant stereo-type associations. Evidence gleaned from our fiveexperiments indicates that country personalityinfluences consumer evaluations of brands associ-ated with a country. Thus, country image effects aremore complex than previously thought.

The overarching theoretical explanation for ourfindings is schema congruity, which is a novelperspective from which to examine the influence ofCOO on consumer brand evaluations. Cognitiveconsistency theories suggest that consumers arepredisposed to prefer elements that are cognitivelyconsistent with their existing knowledge (e.g.,Festinger, 1957). So when presented with newinformation (i.e., brand positioning information)that is consistent with prior knowledge (i.e., coun-try schema), it reinforces the consumer’s previouslyheld stereotypes and is viewed more favorably. Theuse of schema congruity is novel, because iteschews traditional reliance on country imagefavorability and attractiveness of specific personal-ity attributes, i.e., generalized attractiveness ofextroversion. Instead, it relies on the congruity ofbrand positioning and country personality, suchthat when congruent, it facilitates more favorablebrand evaluations.


333


The mediating and moderating variables improveour understanding of how and when brand posi-tioning-country personality congruity matters.First, the effect of congruity enhances perceptionsof cultural authenticity of the brand. Trends ofidentity-related and authentic experiential con-sumption are becoming more relevant to con-sumers as they navigate a landscape of marketglobalization and subsequent standardized strate-gies of global brands. This effect offers someunderstanding of how schema congruity affectsconsumer brand evaluations through culturalauthenticity. Second, the moderating effect ofanimosity on the relationship between brand posi-tioning-country personality stereotype congruityand consumer brand evaluations underscores thecapacity of schema congruity to favorably affectbrand evaluations. Whereas Study 1 shows thatcongruity enhances brand evaluation, Study 2evidences that congruity under conditions of highanimosity toward the country of association resultsnot in a weakened positive effect, but rather areversed effect. This reversal offers evidence that ageneral favorability spectrum lacks the dexterity tocapture other more complex effects of countryimage.

A common criticism of the broad COO literatureis that research findings have been artificiallyinflated due to ecologically unrealistic researchdesigns and it has even been suggested that ‘‘underexternally valid, real-life conditions most consumerdo not consider the COO of the products theypurchase’’ (Samiee & Leonidou, 2011: 75). Thisstudy adds to the body of literature (e.g., Herz &Diamantopoulos, 2013, 2017), which has illus-trated the significant effect of country associationsin realistic settings. For example, in the field study,we worked with the wine shop owner to ensure thatpresentation of the wines mirrored how wines aretypically presented during normal wine samplings.The chocolate advertisements in Study 2 employedimages of Santander chocolate from the brand’swebsite. Further, participants were never informedof the country’s salient personality being investi-gated. Thus, we conclude that focal participantsheld the same country personality stereotypes asthe participants in the qualitative and quantitativepre-tests. In effect, this study combined externallyvalid, real-life conditions with more controlled labenvironments to find evidence of significant brandpositioning-country personality congruity.

Furthermore, across multiple experiments, weutilize different samples (wine shop customers,

Mturk participants, and student subject pool par-ticipants), and we use two related, but distinct,outcome variables to operationalize consumerbrand evaluations (taste quality and customer-based brand equity). We also find consistent resultswith brands from both countries with and withoutfavorable country images. Finally, we manipulatethe brands’ positioning with tangible goods (wineand chocolate), and with an intangible serviceoffering (airlines). Combined, this suggests thatthe conceptual framework is relatively generaliz-able and not confined to a specific context ormeasurement.

Managerial ImplicationsThe traditional COO literature has mostly beendescriptive and advice for managers has beenlimited to ‘‘emphasize’’ or ‘‘hide’’ country associa-tions depending on the country’s degree of favor-ability. With this study, we offer a clearermanagerial lever for managers as our findingssuggest that how the brand emphasizes the countryassociation is of great importance. Particularly, thefindings offer a strategic path for brands fromcountries with unfavorable country images to notonly circumvent the challenge of an unfavorablehome country image, but capitalize on congruity forstrategic gain. Consistent with the traditional COOliterature, in Study 1-B, the Italian brand is signif-icantly preferred over the Indian brand. However,by positioning the brand in a schema-congruentmanner consistent with the country personalitystereotype, brands from countries with a weakcountry image can ‘‘close the gap’’ compared withbrands from advanced economies. Thus, it becomescritically important for managers to be aware of andtrack consumers’ perceptions about its home coun-try’s personality.

Further, merely relying on a favorable countryimage may be sub-optimal when positioning is notcongruent with the dominant country personality.For example, in our follow-up conversations withthe wine shop owner and his wine distributors,they shared an anecdote that German wine makersregularly emphasize the ‘‘passion and enthusiasm’’involved in the production process, perhaps in aneffort to match the passion and love typicallyexpressed by Spanish and French wine makers.Despite the seemingly beneficial assumption that‘‘passion and enthusiasm’’ connote positive asso-ciations for wine (i.e., engagement and love ofcraft), such a positioning is incongruent withGermany’s country personality and, somewhat


334


counterintuitively, German brands would appearto benefit more by emphasizing competence. Theopposite perception may be prevalent in otherindustries. For some industries, it may be easy toassume that competence is a favorable brandattribute that all brands should tout. However,based on our results, there is no evidence of asignificant main effect for either competence orwarmth, it appears that it is only schema con-gruity with country personality stereotypes thatdrives enhanced brand evaluations.

We caution managers that it is highly likely thatconsumers from different countries will perceivecertain target countries differently. Research oncultural values (e.g., Hofstede, Hofstede, & Minkov,2010) has found that our perceptions of othercountries are influenced by our own cultural values,and we suspect this will also influence countrypersonality stereotypes. Thus, firms which attemptto align brand positioning with consumers’ countrypersonality stereotypes must verify the most salientcountry personality stereotypes in each given mar-ket and it is likely that cultural adaptations arenecessary.

Limitations and Future Research DirectionsThe literature provides significant evidence thatcountry associations are a powerful brand associa-tion for many brands competing in internationalmarkets (e.g., Diamantopoulos et al., 2017; Herz &Diamantopoulos, 2017). Accordingly, it should beimperative for brand managers to understand andtry to leverage such country associations, as shownin this study. However, we can, of course, not claimthat brand positioning based on country associa-tions is the only way to be successful. There may beother brand positioning strategies that also havepositive effects.

To combat COO research criticism of overlyartificial research designs, we employed a mix ofstimuli, including ‘‘real’’ wine tasting notes in thefield study, simple product descriptions, and moreelaborate professionally-designed advertisements.Combined, the consistent results across all thestudies with the ‘‘clean, but less realistic’’ simplestimuli to the ‘‘more realistic, but less clean’’elaborate stimuli gives us confidence in the

robustness of the results, but additional researchcan examine consequential outcome variables.Potential extensions also include investigatingwhether the effects hold for other country contextswith different country personality stereotypes. Fur-ther, we adopt the common perspective that brandorigin refers to the country which a consumerassociates with a brand as being its source, regard-less of where the product is made or parts aresourced from (Diamantopoulos et al., 2017). Apotential confounding effect of a diversified globalvalue chain is probably also less of a concern giventhat brand origin and country of manufacture tendto be one and the same for the product categories(e.g., wine and chocolate) examined in this study.However, a potential issue would arise if consumersview a brand as bicultural and particularly if thetwo home countries have different country person-ality stereotypes. For example, Royal Dutch Shell isa British-Dutch company with bicultural identities.The number of bicultural companies, however, isrelatively limited and it was thus beyond the scopeof this study, but it may be a source for futureresearch.

Finally, this research has identified one signifi-cant mediator, cultural authenticity, and one sig-nificant moderator, consumer animosity, that helpus understand the main thesis of this study.However, we cannot rule out that there may beother significant mediators or moderators thatinfluence the brand positioning-country personal-ity congruity relationship, which can be uncoveredin future research. Limitations notwithstanding, byexamining brand positioning-country personalitystereotype congruity, this research advances ourunderstanding of the role of country image associ-ations in a more complex and precise way thatmoves beyond a general favorability continuum tooffer a deeper understanding of the role of COO onconsumer brand evaluations.

ACKNOWLEDGEMENTSThe authors gratefully acknowledge Nicolas Papado-poulos, Colleen Harmeling, and participants at theAMA Global Marketing Conference in Havana, Cubafor feedback on early versions of this manuscript and


335


the valuable guidance provided by Editor CostasKatsikeas and the three anonymous JIBS reviewers.

NOTES1One-tailed tests were performed on all planned

contrasts.

REFERENCESAaker, J. L. 1997. Dimensions of brand personality. Journal ofMarketing Research, 34(3): 347–356.

Aggarwal, P., & McGill, A. L. 2007. Is that car smiling at me?Schema congruity as a basis for evaluating anthropomor-phized products. Journal of Consumer Research, 34(4):468–479.

Alden, D. L., Steenkamp, J.-B. E. M., & Batra, R. 1999. Brandpositioning through advertising in Asia, N America, andEurope: The role of global consumer culture. Journal ofMarketing, 63(1): 75–87.

Ayala, N., Truss, M., & Montecuollo, K. 2012. JWT personalityatlas. http://personalityatlas.com/. Accessed 27 March 2018.

Baker, V. 2017. Havaianas: How a Brazilian flip-flop took over theworld. http://www.bbc.com/news/world-latin-america-40610739. Accessed 27 March 2018.

Barbarossa, C., De Pelsmacker, P., & Moons, I. 2018. Effects ofcountry-of-origin stereotypes on consumer responses to pro-duct-harm crises. International Marketing Review, 35(3):362–389.

Becker, J. 2015. German neo-mercantilism: Contradictions of a(non-)model. In B. Unger (Ed.), The German model: Seen by itsneighbours. London: SE Publishing.

Bilkey, W. J., & Nes, E. 1982. Country-of-origin effects onproduct evaluations. Journal of International Business Studies,13(1): 89–99.

Cartwright, D. 1949. Some principles of mass persuasion.Human Relations, 2(3): 253–267.

Cuddy, A. J. C., Fiske, S. T., Kwan, V. S. Y., Glick, P., Demoulin,S., Leyens, J.-P., et al. 2009. Stereotype content model acrosscultures: Towards universal similarities and some differences.British Journal of Social Psychology, 48(1): 1–33.

d’Astous, A., & Boujbel, L. 2007. Positioning countries onpersonality dimensions: Scale development and implicationsfor country marketing. Journal of Business Research, 60(3):231–239.

Diamantopoulos, A., Florack, A., Halkias, G., & Palcu, J. 2017.Explicit versus implicit country stereotypes as predictors ofproduct preferences: Insights from the stereotype contentmodel. Journal of International Business Studies, 48(8):1023–1036.

Dichter, E. 1962. The world customer. Harvard Business Review,40(4): 113–122.

Festinger, L. 1957. A theory of cognitive dissonance. Stanford,CA.: Stanford University Press.

Fetscherin, M. 2010. The determinants and measurement of acountry brand: The country brand strength index. Interna-tional Marketing Review, 27(4): 466–479.

Fiske, S. T. 1982. Schema-triggered affect: Applications to socialperception. In M. S. Clark & S. T. Fiske (Eds.), Affect andcognition, 17th annual Carnegie Symposium on Cognition (pp.55–78). Hillsdale, NJ: Erlbaum.

Fiske, S. T., Cuddy, A. J. C., Glick, P., & Jun, X. 2002. A model of(often mixed) stereotype content: Competence and warmth

respectively follow from perceived status and competition.Journal of Personality and Social Psychology, 82(6): 878–902.

Fiske, S. T., & Taylor, S. E. 1991. Social cognition. New York:McGraw-Hill.

Gao, G. Y., Wang, D. T., & Che, Y. 2017. Impact of historicalconflict on FDI location and performance: Japanese invest-ment in China. Journal of International Business Studies,advance online publication.. https://doi.org/10.1057/s41267-016-0048-6.

Gineikiene, J., & Diamantopoulos, A. 2017. I hate where itcomes from but I still buy it: Countervailing influences ofanimosity and nostalgia. Journal of International BusinessStudies, 48(8): 992–1008.

Gürhan-Canli, Z., Sarıal-Abi, G., & Hayr

beyond country image favorability: how brand positioning via ......2010), havaianas closely...

Documents