fashion with a foreign flair: professional experiences

FASHION WITH A FOREIGN FLAIR: PROFESSIONALEXPERIENCES ABROAD FACILITATE THE CREATIVE

INNOVATIONS OF ORGANIZATIONSFRÉDÉRIC C. GODART

INSEAD

WILLIAM W. MADDUXINSEAD

ANDREW V. SHIPILOVINSEAD

ADAM D. GALINSKYColumbia University

This research explores whether the foreign professional experiences of influential execu-tives predict firm-level creative output. We introduce a new theoretical model, the foreignexperience model of creative innovations, to explain how three dimensions of executives’foreign work experiences—breadth, depth, and cultural distance—predict an organiza-tion’s “creative innovations,” which we define as the extent to which final, implementedproducts or services are novel and useful from the standpoint of external audiences. Weexamined 11 years (21 seasons) of fashion collections of the world’s top fashion houses andfound that the foreign professional experiences of creative directors predicted the creativ-ity ratings of their collections. The results revealed individual curvilinear effects for allthree dimensions: moderate levels of breadth and cultural distance were associated withthe highest levels of creative innovations, whereas depth showed a decreasing positiveeffect that never turned negative. A significant three-way interaction shows that depth isthe most critical dimension for achieving creative innovations, with breadth and culturaldistance important at low but not high levels of depth. Our results show how and whyleaders’ foreign professional experiences can be a critical catalyst for creativity andinnovation in their organizations.

Karl Lagerfeld is an icon of the fashion industry.With his trademark sunglasses, tight black suits,and shock of white hair, he is nothing if not con-

spicuous. By virtue of his decades-long career asthe creative director of two of the world’s top fash-ion houses, Chanel and Fendi, he has establishedhimself as a major creative force. Lagerfeld is cred-ited with saving the Chanel brand from decay in theearly 1980s when he was appointed creative direc-tor of both its haute-couture and ready-to-wearlines (Vergani, 2010: 262). Largely due to his influ-ence, both Chanel and Fendi now belong to the top10 of global luxury brands (BrandZ, 2013), withChanel’s value reaching $7 billion and Fendi’svalue exceeding $3.5 billion. These are exceptionalachievements. Lagerfeld is clearly a talented de-signer and business leader.

One of the keys to Lagerfeld’s success may be hismulticultural background: Born in Hamburg to aSwedish father and German mother, Lagerfeldworks in France and Italy, often commuting be-tween the two countries during the course of the

Previous versions of this paper were presented at the2012 American Sociological Association Meeting in Den-ver, Colorado; at the 2012 Academy of Management con-ference in Boston, Massachusetts; and at the 2013 NewDirections in Leadership Research Conference at INSEAD,Fontainebleau, France. We would like to thank AMJ asso-ciate editor Adam Grant and three anonymous reviewersfor their suggestions, as well as Cameron Ford, Shira Mor,Michael Pratt, and Florence Rambaud who have providedcomments and suggestions. We are grateful to Liselott Pet-tersson and her team at the INSEAD Social Science Re-search Centre in Paris, and to INSEAD research associateDilla Wong, for their assistance in data collection and cod-ing. This research was supported by a grant from theINSEAD Research and Development Committee and wethank the INSEAD Alumni Fund for their support.

195

! Academy of Management Journal2015, Vol. 58, No. 1, 195–220.http://dx.doi.org/10.5465/amj.2012.0575

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

same day. Further, he once proclaimed that hewould like to be a “one-man multinational fashionphenomenon” (Shapiro, 1984), someone who usesa diversity of cultural influences in his collectionsto make a lasting imprint on the global fashionindustry.

Although the fashion industry is a unique con-text, creativity and innovation are critical to suc-cess across a variety of organizational domains(Amabile, 1996). Indeed, as organizations becomemore globally oriented, it is increasingly importantto understand how culturally diverse experiences,such as working in a foreign country, affect thecreativity of professionals and their organizations,both in terms of generating new ideas and imple-menting them as products or services (Hammond,Neff, Farr, Schwall, & Zhao, 2011). Currently, thepotential benefits of such experiences remainlargely anecdotal. In addition, even though the gen-eral relationship between certain kinds of experi-ences abroad and their subsequent creative benefitsis beginning to find initial empirical support in thepsychological literature (e.g., Leung & Chiu, 2010;Leung, Maddux, Galinsky, & Chiu, 2008; Maddux &Galinsky, 2009; Tadmor, Galinsky, & Maddux,2012), it is still unknown what types of profes-sional experiences abroad will be powerful enoughto impact the creativity of organizational outputs(i.e., products and services).

The current analysis explores whether the for-eign professional experiences of influential execu-tives can predict firm-level creative output. To doso, we introduce a new theoretical framework—theforeign experience model of creative innova-tions—to understand when and why individual ex-ecutives’ foreign professional experiences can im-pact the creativity of their organizations’ output.We used a unique dataset to examine the life his-tories of fashion houses’ creative directors to deter-mine how their individual experiences predictedthe creativity ratings of their collections shown be-tween 2000 and 2010.

Our study design and theoretical model allow usto directly address numerous gaps in the extantliterature. First, the current study is the first toexamine whether and how the foreign professionalexperiences of certain influential individuals canimpact organizational-level output. We investigatewhether the foreign professional experiences offashion house creative directors predict the creativ-ity of their firms’ innovations. Second, althoughprevious work has examined the impact of multi-cultural experiences on creativity in general (e.g.,

Leung et al., 2008), the current study design al-lowed us to explore the particular impact of foreignwork experiences on an organization’s creative in-novations. Third, we offer the first examination ofhow creativity is affected not only by the depth offoreign experiences (e.g., Maddux & Galinsky,2009), but also by the breadth of foreign experienceand by the cultural distance between the countriesin which individuals were socialized and the coun-tries in which they later worked. Fourth, we intro-duce a new theoretical model—the foreign experi-ence model of creative innovations—to explainhow and why breadth, depth, and cultural distancecan affect organizational-level creative output. Fi-nally, our model builds off three heretofore sepa-rate theoretical frameworks: (a) Campbell’s (1960)blind variation and selective retention (BVSR)model of creativity, (b) the theory of internationaladjustment (Black, Mendenhall, & Oddou, 1991),and (c) the social embeddedness perspective oncreativity (Burt, 2004; Godart, Shipilov, & Claes,2014; Perry-Smith & Shalley, 2003).

Overall, we bring together previously separatetheoretical perspectives to develop a model for un-derstanding specific mechanisms by which profes-sional foreign experiences of individuals translateinto the creative innovations of organizations. Thismultilevel perspective allows us to generate newinsights for the psychological (Hammond et al.,2011) and sociological (e.g., Baum, Shipilov, &Rowley, 2003; Cattani & Ferriani, 2008; Godart etal., 2014; Uzzi & Spiro, 2005) theories of creativityand innovation, as well as for the literatures ofwork experience (Tesluk & Jacobs, 1998), diversity(Joshi & Roh, 2009), and cross-cultural management(Molinsky, 2007).

FOREIGN EXPERIENCES ANDCREATIVE INNOVATIONS

Creativity is defined as something novel and use-ful (Amabile, 1996; Zhou & Shalley, 2003). Noveltyrefers to the extent to which a concept, an idea, ora product differs from conventional practices in aparticular domain; usefulness is the degree towhich a given output is recognized to have func-tional utility for a given audience. As novelty andusefulness are determined within “the bounds ofsocial, cultural, and historical precedents of thefield” (Perry-Smith & Shalley, 2003: 91), a givenoutput is viewed as creative “to the extent thatappropriate observers independently agree it is cre-ative” (Amabile, 1996: 33).

196 FebruaryAcademy of Management Journal

It is important to note that, in the organization-al context, some scholars distinguish the conceptof creativity from that of innovation. Whereascreativity involves the generation of novel anduseful ideas by individuals or teams, innovationencompasses both the generation of ideas and theselection of some subset of these ideas for imple-mentation by internal audiences, such as seniorexecutives, in an organization (Clegg, Unsworth,Epitropaki, & Parker, 2002; Hammond et al.,2011). Although all innovations are, by defini-tion, creative to some extent, there is great vari-ance in the exact level of creativity a given inno-vation might exhibit. Indeed, innovation scholarshave made a distinction between incrementaland radical innovations, with the latter beingmore creative than the former (Damanpour, 1991;Dewar & Dutton, 1986; Henderson & Clark, 1990).

Creative innovations are critical to success inmany industries. In most creative industries, suchas fashion, art, video game making, technology,publishing, or film, success depends not on thecreativity of each idea generated during the entiredevelopment and production process, but, rather,on external audiences’ evaluation of the final prod-uct brought to market. In such industries, then, itis difficult, and often impossible, to separate cre-ativity from innovation (Caves, 2000). For exam-ple, in the film industry, audience members andcritics do not evaluate the novelty and usefulnessof tens of thousands of scripts that were neverturned into movies, nor do they evaluate ideasthat were suggested by scriptwriters but later cutfrom the film. Rather, they make their evalua-tions based on films’ final released versions (Cat-tani & Ferriani, 2008). Similarly, buyers and jour-nalists who evaluate the novelty and usefulnessof fashion collections do not look at the design-er’s initial drawings, nor do creative teams com-pile and save all ideas proposed. Rather, buyersand journalists evaluate only the finished cloth-ing items. Thus, the success of a fashion housedepends on the creativity of implemented ideas(Godart, 2012a). Given the difficulty of disentan-gling creativity from innovation in such contexts,we label such organizational output in creativeindustries as “creative innovations,” defined asthe extent to which final, implemented productsare novel and useful from the standpoint of rele-vant external audiences.

Although creativity remains a somewhat mys-terious phenomenon, much is now known aboutthe underlying psychology of creative personali-

ties and of the creative process (for reviews, seeFeist, 1998, 1999; MacKinnon, 1978; Simonton,2000; Zhou & Shalley, 2003). One theme that thisresearch has highlighted is the importance of in-dividual and contextual diversity in facilitatingcreativity. For example, at the individual level,first- or second-generation immigrants are morecreative compared to individuals raised in a sin-gle country (Lambert, Tucker, & d’Anglejan,1973; Simonton, 1994, 1997, 1999). Similar ef-fects have been shown for bilinguals (Nemeth &Kwan, 1987; Simonton, 1999), who exhibit en-hanced creativity compared to monolingual indi-viduals. At the group and organization level,moderate levels of team diversity are importantbecause they not only produce the right amountof novel creative inputs, but also enough inter-personal tension to spark creativity, though notso much tension as to impede group performance(Baer, Leenders, Oldham, & Vadera, 2010;Guimerà, Uzzi, Spiro, & Nunes Amaral, 2005;Shin, Kim, Lee, & Bian, 2012). Furthermore, in-dividuals with ties to diverse informational do-mains inside their organizations are likely to ex-hibit higher creativity (Burt, 2004; Perry-Smith &Shalley, 2003). And research has found that, thelonger individuals have lived abroad and the morethey adapted to their host counties, the better theyperform on standard psychological tests of creativity(Maddux & Galinsky, 2009).

Reviewing this and other research, Leung et al.(2008) suggested that certain types of multicul-tural exposure or diverse cultural experiencescan enhance general creative ability. However, anopen question is whether the foreign professionalexperiences of individuals influence the creativ-ity of implemented organizational products orservices (Anderson, De Dreu, & Nijstad, 2004;Clegg et al., 2002).

Recent meta-analyses suggest that individualscan indeed implement their ideas more easily, orhave a particularly strong influence on organiza-tional output, if they find themselves in particularorganizations or positions (Clegg et al., 2002; Ham-mond et al., 2011). For example, executives in po-sitions of formal power, such as C-level executives,will be especially likely to influence firm-level out-comes (Staw, 1980). High levels of formal influenceover organizational outcomes can also be seen inentrepreneurial start-ups, which reveal the behav-iors and biases of their owners due to their highcentralization and small size (Staw, 1991; Uzzi,1996). In addition, if individuals have well-devel-

2015 197Godart, Maddux, Shipilov, and Galinsky

oped social networks, they will also have influenceover the implementation of their ideas inside theorganization (Baer, 2012; Dutton & Ashford, 1993).Thus, when individuals who have importantprofessional experiences abroad are also organiza-tional leaders in charge of innovative activities(e.g., head of R&D in a technology-based firm, cre-ative director in the fashion industry, or producerin the movie industry), their foreign professionalexperiences may be profound enough to influencethe creativity of their organizations’ innovations.

The Foreign Experience Model ofCreative Innovations

To understand the link between individual for-eign experiences and organizational creativity, wepropose a new comprehensive theoretical modelcalled the foreign experience model of creative in-novations. The first component of this model in-volves insights from the BVSR model of creativity(Campbell, 1960). From this perspective, creativitymechanisms are similar to those of natural selec-tion in biological evolution, with ideas’ develop-ment initially proceeding via a relatively randomvariation process of either completely new con-ceptualizations or novel combinations of existingideas. Such a process is random or “blind” in thesense that there is no particular logic or a priorirationale for the ideas’ generation. Instead, the ini-tial search proceeds whereby many new conceptu-alizations or combinations are created relativelyhaphazardly based on whatever different inputs areavailable, in the hope that something of value willeventually be produced. In the second step, that ofselective retention, a subset of the most promisingvariations are then selected for further explorationand refinement, eventually leading to an end prod-uct that is considered both novel and useful—inother words, creative (Campbell, 1960; Simonton,1999, 2011). When individuals work abroad, suchexperience exposes them to a larger number andmore diverse array of new inputs, concepts, andideas than they could have access to within theirown country.1

In addition, exposure to different environ-ments will make individuals more able to takerisks. This will occur because new inputs, con-cepts, and ideas will make individuals comfort-able with challenging the status quo, as they willhave a greater ability to recognize non-traditionalopportunities in changing processes and sche-mas. As implementing creative ideas tends to berisky (Baer, 2012), professional foreign experi-ence will help in these ideas’ implementation.

However, we also suggest that exposure to vari-ation by itself is not enough to stimulate creativeinnovations. Importantly, the second component ofour model—one of psychological adaptation—isneeded to transform foreign experiences into lastingand tangible psychological benefits. Indeed, previousresearch has shown that not all foreign experienceslead to enhanced creativity; instead, people mustadapt themselves to the new culture (Maddux & Ga-linsky, 2009), undergo deep learning experiences(Maddux, Adam, & Galinsky, 2010), or integrate thenew culture into their own identity (Tadmor et al.,2012) for foreign experiences to produce creative ben-efits. These findings suggest that the process of ad-justment that individuals go through when they liveor work in a new country is a key factor (Bhaskar-Shrinivas, Harrison, Shaffer, & Luk, 2005; Black et al.,1991). For example, Black et al. (1991) suggested thatemployees need to go through a period of difficultand intense socialization and sensemaking with re-gard to both the cultural and work environment inorder to make the necessary personal and profes-sional transition that international assignments de-mand. To the extent that situational, interpersonal, ororganizational factors facilitate such adjustment, thenexpatriates can better adapt to new cultures. Thisadaptation, in turn, helps individuals to find creativeways of approaching problems in the future (Mad-dux, 2011).

Third, professional foreign experiences willalso affect creative innovations by facilitatingindividuals’ embeddedness in professional net-works (Godart et al., 2014) and their general net-working ability (e.g., Baer, 2010). Social em-beddedness provides valuable professional infor-mation and tacit knowledge about how to gener-ate and implement ideas, as these tend to becodified and transmitted through informal rela-tionships (Uzzi, 1996). Moving across various ge-ographies helps build wide-reaching bridgingties (Reagans & McEvily, 2003) across the pools ofgeographically localized knowledge. These tiescan thus provide exposure to a greater blind vari-

1 Similar predictions have been made in the study ofintrapersonal diversity, which show that executives’prior work in different functional areas help them togenerate and implement ideas for achieving profitabilitytargets (Bunderson & Sutcliffe, 2002).


ation of ideas, which then enhance the ability tocombine insights from different domains (e.g.,Fleming, 2001; Galunic & Rodan, 1998; Mednick,1962) and generate novel insights through selec-tive retention. Foreign experiences also assist theimplementation of these ideas because intercul-tural collaboration offers exposure to a widerange of political knowledge and issue-sellingtactics from around the world (e.g., Tesluk &Jacobs, 1998). Such exposure increases individu-als’ ability to communicate with a variety ofstakeholders inside the firm (Bunderson & Sut-cliffe, 2002), help build better intra-firm coali-tions (Dutton & Ashford, 1993), mobilize spon-sorship and advocacy (Obstfeld, 2005), and driveorganizational change more effectively (Ferris etal., 2005).

In sum, in order to produce creative innova-tions, professional foreign experiences must pro-vide individuals with exposure to novelty, butalso with sufficient opportunities to psychologi-cally adapt to these foreign environments and tobecome embedded in different professionalnetworks.

From these dimensions of our theoretical model,we are able to derive three main predictions aboutwhen and how foreign professional experienceswill affect creative innovations. First, we hypothe-size that moderate levels of breadth of organizationalleaders’ professional foreign experience will be as-sociated with more creative innovations. In es-sence, breadth involves exposure to a variety ofinputs. Such variety can help individuals betterrealize multiple approaches to the same problem,or conceive of new, unique ways of solving a spe-cific issue, both by observing how things are donein different countries as well as from receiving in-formation from intercultural collaborations withinprofessional networks across these countries(Laursen, Masciarelli, & Prencipe, 2012; Sorenson& Stuart, 2001). In addition, breadth increases thenumber of country-spanning bridging ties (Oettl &Agrawal, 2008). The diversity of information ex-changed through these bridging ties may help thegeneration of novel ideas (Burt, 2004), increase com-fort with risk taking (Baer, 2010), and offer exposureto a variety of political skills and influence tactics(e.g., Tesluk & Jacobs, 1998), all of which will have apositive effect on the creative innovations.

However, based on the adaptation component ofour model, we argue that very high levels ofbreadth may begin to preclude one’s ability toadapt to each of their many new experiences,

which could end up having a detrimental effect oncreative innovations. For example, an executivemay find that working in two different countriesmakes it possible to integrate and embed oneselfinto the new cultural contexts and networks, yetalso be enriching enough to stimulate the genera-tion of novel ideas and the capacity to get themimplemented. That same executive may find, how-ever, that working in six different countries is toooverwhelming to be able to adapt to each, and toodifficult to become effectively embedded in themyriad different networks encountered. In otherwords, high levels of breadth might lead an indi-vidual to experience information overload, whichhas been shown to impede creativity (Chua & Iyen-gar, 2008; Edwards, 2001; Gurteen, 1998). The pos-itive effects derived from an influx of novel infor-mation and access to new social connections isexpected to satiate at a high level of breadth,whereas the adverse influences from informationoverload and the inability to adapt may eventuallyescalate, the satiation and escalation processes bothcontributing to the overall theorized effect (Grant &Schwartz, 2011). Thus, we expected that breadth offoreign professional experiences would be optimalwhen experienced at moderate levels, after whichthe marginal benefits of greater breadth will declineand eventually may turn negative.

Hypothesis 1. The effect of breadth of an or-ganizational leader’s foreign professional ex-periences on the firm’s creative innovationswill have an inverted U-shaped relationship,such that relatively moderate levels of breadthwill be associated with the highest level of cre-ative innovations.

Second, we hypothesize that moderate levels ofdepth of organizational leaders’ professional for-eign experience will be associated with more cre-ative innovations. Similar to breadth, depth of pro-fessional foreign experience can also provideindividuals with requisite variety, because, the lon-ger a person works in a foreign country, the morethis person is exposed to diverse inputs, ideas, andconcepts. In addition, the adaptation component ofour model suggests that deep experiences will pro-duce greater opportunities and incentives to psy-chologically adapt and to truly internalize theforeign culture. Less deep experiences may not pro-vide enough opportunities or incentives for truepsychological transformations, such as adaptation,learning, and identity change (Maddux & Galinsky,2009), for the discovery of informal influence strat-


egies (Baer, 2012; Dutton & Ashford, 1993), or forbenefiting from intercultural collaborations (Blacket al., 1991). Depth will also provide opportunitiesto integrate in a variety of different audiences,which can facilitate translation and communica-tion of their ideas and the building of support co-alitions inside their own organization (Bunderson& Sutcliffe, 2002).

Furthermore, deeper professional foreign experi-ences can allow opportunities to more successfullyembed oneself into foreign professional networks.Most of the tacit knowledge exchanged within com-munities happens through strong ties (Uzzi, 1996)and dense networks (Reagans & McEvily, 2003),both of which require effort to build. Such tieswill not only provide an individual with fine-grained information about how things are done inspecific cultural and professional settings, but alsooffer access to resources, buy-in, and support,which will help the individual to increase the use-fulness of his or her products or services within thefocal foreign environment and offer help in imple-menting them (Cattani & Ferriani, 2008).

However, a foreign environment may act as acatalyst to creative innovations only as long as theenvironment continues to be perceived as stimulat-ing and novel. Indeed, research has demonstratedthat individuals who have completely assimilatedto a new culture and have lost their original cul-tural identity lose the creative benefits of livingabroad (Tadmor et al., 2012). Additionally, once anindividual has achieved very deep professional ex-periences in a foreign country, this individual canbecome “overembedded” (Uzzi, 1996) within thatcountry’s professional networks, focusing on infor-mation received from ties in this country to thedetriment of ties to other countries. As this personincreasingly focuses on the information circulatingin a single geographical network, their ability togenerate and implement ideas will cease to be dif-ferent from people who never left that country.2

Furthermore, after spending too much time in aforeign country, individuals might start experienc-ing cognitive entrenchment, a decrease in mentalflexibility when one is exposed too deeply to adomain of knowledge. Cognitive entrenchmentleads to heightened functional fixedness (Dane,2010: 584) and, consequently, to lower creativity.With an increase in depth, the negative effects ofcognitive entrenchment and overembeddednesswill escalate, whereas the positive effects of oppor-tunities for psychological transformation and net-work embeddedness will gradually satiate. Thus,we expected that the depth of foreign professionalexperiences would be optimal when experienced atmoderate levels, after which the benefits of greaterdepth may level off or decline.

Hypothesis 2. The effect of depth of an organ-izational leader’s foreign professional experi-ences on the firm’s creative innovations willhave an inverted U-shaped relationship, suchthat relatively moderate levels of depth will beassociated with the highest levels of creativeinnovations.

Finally, the cultural distance between one’shome country (i.e., the country where one was so-cialized) and the foreign countries in which one isworking may be an important determinant ofwhether individuals experience optimal levels ofvariation but also have the psychological resourcesto adapt to the new environment. A host of researchhas noted that there are a number of cultural di-mensions and values along which countries vary(Hofstede, 1980; House, Hanges, Javidan, Dorfman,& Gupta, 2004; Schwartz, 1994). This means thatsubjective experiences abroad will vary dependingon the closeness of the cultural characteristics be-tween home and host countries. Indeed, the blindvariation and network embeddedness componentsof our model suggest that the cultural distance be-tween the home country and the various host coun-tries should be an important dimension of foreignexperience for creative innovations. Because expo-sure to novel variation is a critical means by whichforeign experiences promote creativity (Simonton,1999; Weick, 1979), cultural distance is likely tohave a positive effect on the ability to generate andimplement novel and useful ideas. For example,working in a host country that is close in culturaldistance to one’s home country (say, Canada andthe United States, respectively) may not providethe requisite novelty to either impact general cre-ative processes or the motivation and ability to

2 This is not to say that this individual will be cut offfrom information circulating through the Internet or otherpublic media as a result of living in a foreign country for avery long time. Rather, this person will be cut off from thediverse tacit information exchanged through bridging tiesin professional social networks across geographies—for ex-ample, what are the contemporary sources of inspiration,how are new ideas implemented, who are the best partnersto work with—information that is not available in the pub-lic domain (Cross & Parker, 2004), especially in the creativeindustries (Currid, 2007).


implement creative ideas (Baer, 2012; Hammond etal., 2011). In contrast, professional stints in coun-tries that show more variance across cultural valuesand norms, such as an American having an inter-national assignment in Korea (Oh, Chung, & Labi-anca, 2004) or Ukraine (Danis & Shipilov, 2002)would provide more novel inputs. Such experi-ences would also expose individuals to more het-erogeneous professional networks and practices,which might, in turn, become useful in gainingaccess to ideas from different cultures, communi-cating and translating these ideas to people withdifferent backgrounds (Kostova & Roth, 2003), andbuilding intra-organizational coalitions in supportof these ideas.

Nevertheless, the adaptation component of ourmodel also implies that high cultural distance may,at some point, begin to preclude one’s ability toadapt (Black et al., 1991). In line with this argu-ment, a meta-analysis has shown that the quality ofan individual’s adaptation to a new country de-creases as the cultural novelty to which this personis exposed increases (Hechanova, Beehr, & Chris-tiansen, 2003). If the cultural distance is so high asto preclude adaptation, then we should expect lim-ited benefits both for the ability to generate novelideas and for the ability to implement them. Highcultural distance might prove so overwhelmingthat it impedes the ability to absorb the requisitevariety (Weick, 1979) of creative inputs as well ashinders the learning of socialization, interculturalcollaboration, and coalition building (Morris,Podolny, & Sullivan, 2008). A person working in acountry with a high cultural distance to their home-land might also be so stressed by the experiencethat they could become less likely to generate novelideas or to take risks in implementing them due tothe inability to foresee consequences in vastly dif-ferent cultural environments. In other words, as thecultural distance increases, the benefits from ac-cessing the requisite variety will satiate, whereasthe stress due to the inability of individuals tosocialize, engage in intercultural collaborations,and build coalitions will escalate. Thus, levels ofcultural distance may be most optimal when expe-rienced at relatively moderate levels.

Hypothesis 3. The effect of cultural distance ofan organizational leader’s foreign professionalexperiences on the firm’s creative innovationswill have an inverted U-shaped relationship,such that relatively moderate levels of cultural

distance should be associated with the highestlevels of creative innovations.

DATA AND METHODS

Study Context

The high-end fashion industry, which is the set-ting for this study, is a prototypical creative indus-try that can be used as a good illustration for howcreative innovations emerge (Caves, 2000; Crane,1999; Crane & Bovone, 2006; Godart & Mears,2009). Sales and profit in fashion are largely de-rived from, and thus are highly dependent on, cre-ative innovations. This is perhaps most vividly il-lustrated by the fact that the most significant publicfigures and most influential organizational leadersof the high-end fashion houses are their “creativedirectors”—the individuals in charge of definingthe houses’ bi-annual collections—rather than theirCEOs. These creative directors, who can sometimeshave a different title such as “artistic director,” caneither be the founders of their own house (e.g.,Marc Jacobs is the founder and creative director ofMarc Jacobs) or work for a house founded by some-one else (e.g., Alber Elbaz is the creative director ofLanvin, which was founded in 1889 in Paris byJeanne Lanvin). Industry stalwarts such as MarcJacobs, Karl Lagerfeld, Giorgio Armani, Tom Ford,Miuccia Prada, or Alber Elbaz exert enormous con-trol over their houses’ creative vision and collec-tions, as well as set the tone for the entire fashionindustry—they “are the primary creators of fashionwithin the fashion industry” (Sproles & Burns,1994: 45). Although creative directors of somewhatless well-known fashion houses (e.g., Alice Roi,Antonio Berardi) have less industry influence, theystill wield almost complete control over theirhouses’ collections, generating and implementingideas concerning looks, colors, fabrics, or patterns,for example (Kawamura, 2005).

It should be noted that, even though creativedirectors do not work in isolation (they oftentimeshave stylists, photographers, PR professionals, as-sistant designers, and others who help them intheir professional endeavors), they are, withoutquestion, in charge of defining the vision of a col-lection. This has led Kawamura (2005: 57) to writethat “although it is important to remember that[creative directors] are not the only players, [they]are and must be portrayed as ‘stars’ in the produc-tion of fashion.” Creative directors are personallyevaluated by fashion buyers and journalists based


on what they are able to produce for their fashionshows in a context of high interpersonal competi-tion among creative directors of different houses(Blumer, 1969). Thus, the process of generating andimplementing creative ideas in fashion is very cen-tralized, and is attached to the person of the cre-ative director.3

Other industries also have positions that yield astrong influence on the creativity of the organiza-tional output. In the film industry (Cattani & Ferri-ani, 2008), for example, the position equivalent tothe fashion industry’s “creative director” is the“film director” (e.g., Steven Spielberg, Jean-Luc Go-dard, or Alfred Hitchcock), and most of the successor failure of a movie is attributed to this person. Inthe Broadway musical show industry (Uzzi &Spiro, 2005), it would be the “stage director” or“impresario” who influences the major creative el-ements; for example, Andrew Lloyd Webber, thecreator of The Phantom of the Opera, or CatherineJohnson, the creator of Mamma Mia!. Following theprocess highlighted by Staw (1991), the high cen-tralization of fashion houses’ creative operationswill help their creative directors to drive the organ-izational outcomes. Because creative directors fullydefine their collections, their professional experi-ences abroad should also have a strong impact onthe collections’ creative innovations.

Data Collection and Variables

We collected industry-wide data on the globalhigh-end fashion industry over 21 fashion seasons(covering both fall/winter and spring/summer, thetwo main fashion seasons) between 2000 and 2010.The total number of fashion houses studied was270. Most of the data came from publicly availablesources, such as industry publications and com-pany websites, as detailed below. The first step of

the data collection was to identify the firms (i.e.,fashion houses) competing in the market. We did soby collecting the names of all the houses that orga-nized a major fashion show in one of the four“fashion capitals”—Paris, New York, Milan, andLondon—which constitute the core of the “fashionsystem” for the time period we considered(Breward, 2003; Kawamura, 2005, 2011). This strat-egy excluded houses that do not have the means toorganize a fashion show, and bigger mass-marketclothing companies (e.g., H&M, Forever 21, Uniqlo)that do not usually organize shows.

Based on this definition of our population ofhigh-end fashion houses, we also collected life andcareer histories of creative directors who workedfor these houses from industry encyclopedias(Price Alford & Stegemeyer, 2009; Vergani, 2010),as well as from leading industry publications (suchas Women’s Wear Daily, Journal du Textile, orVogue). Websites such as www.fashionmodeldirec-tory.com (FMD), www.nymag.com (New York mag-azine), or www.style.com (fashion website), as wellas www.factiva.com (Dow Jones & Company), com-plemented the aforementioned sources. Data on de-signers span a period starting in the 1930s andending in 2010. To most accurately use our archivaldata to operationalize our key constructs, we con-ducted more than 30 interviews with industry in-siders. These interviews took place between 2007and 2011. We complemented them with an exten-sive review of industry reports.

Independent variables. For all of our indepen-dent variables, we defined one’s “home country” asthe country in which one was socialized; thatis, the country where one spent the most time be-fore the age of 18. For breadth of professional ex-periences abroad, we calculated the number of for-eign countries in which individuals had worked.For depth of professional experiences abroad, wecalculated the number of years each director hadworked abroad in their professional career (Mad-dux & Galinsky, 2009). Both variables included thebreadth and depth accumulated while this personwas not yet a creative director, as well as thebreadth and depth accumulated following the per-son’s promotion to the creative director’s position.We also constructed the index of cultural distancebetween the countries in which the person hasbeen working. We used Hofstede’s (1980; Hofstede,Hofstede, & Minkov, 2010) cultural distance scores.The distances between countries based on the Hof-stede’s dimensions were aggregated using the Kan-dogan (2012) approach, which is a modified and

3 A good example of the role played by the creativedirector as the defining force of a high-fashion collectioncomes from the 1995 documentary, Unzipped, about thelife of American designer Isaac Mizrahi (Godart, 2012b).In this documentary, Mizrahi is shown preparing a col-lection that is inspired by the 1922 silent documentaryfilm Nanook of the North and the 1935 adventure filmThe Call of the Wild. The designer’s vision about thiscollection becomes associated with him individually.His team supports this vision—for example, by scoutingthe press to see what other designers are doing, or byhandling relations with suppliers and buyers—but he isthe one translating his visionary idea into actual designs.


improved version of the Kogut and Singh (1988)method. The idea behind this approach is that onecomputes an aggregated score of cultural distancebetween two countries based on the distances oneach of the Hofstede’s dimensions (Kogut & Singh,1988), while taking into account the possible biasresulting from positive or negative correlations be-tween the pairs of dimensions (Kandogan, 2012).4

To compute our measure of cultural distance, wecompared the home country to the foreign countryin which the creative director has been working. Incases where there were several countries—say,work experiences in Canada and Japan for a personraised in the United States—we added up the ab-solute values of cultural distances between theUnited States and Canada as well as those of theUnited States and Japan. We chose the sum becauseit reflects the entire requisite variety to which anindividual is currently exposed.

Dependent variable. To evaluate the creativeinnovations produced by fashion houses, we usedthe only industry-validated measure available—theratings in the renowned French trade magazineJournal du Textile (JdT) (Barkey & Godart, 2013;Crane, 1997). The JdT scores are widely used byinternational fashion industry professionals to fol-low major market trends, identify up-and-comingdesigners, and assess the financial value of thefashion houses. Previous research has used thisdata to capture creativity in the fashion industry,albeit for different research questions (Barkey &Godart, 2013; Crane, 1997). JdT constructs its rank-ing by asking industrial buyers to evaluate the cre-ativity of fashion collections in both fall/winter and

spring/summer fashion shows (the actual numberof buyers varies from 65 to 70, averaging at 67.5across 21 seasons). Buyers comprise an appropriatejury because they represent the vital constituentsthat decide whether or not fashion collections areactually sold in stores. These individuals are accus-tomed to judging collections, and their career de-pends on the ability to evaluate creativity in fash-ion: All of them either own or buy for fashionretailers and distributors around the world. Buyersevaluate the creativity of collections by attendingthe actual shows, reading magazines such as Voguethat report on the shows, or watching the collec-tions online. Using buyers’ ratings as a dependentvariable is also consistent with the “consensualassessment” definition of creativity where some-thing is considered to be creative if knowledgeablethird parties consider it to be creative (Amabile,1982, 1996). In addition, the evaluation made bybuyers looks at both novelty and usefulness. Nov-elty comes from the fact that fashion is character-ized by “recurrent change” (Aspers & Godart,2013), and critics typically praise collections thatcontain new designs, as compared to those in thepast seasons; usefulness comes from the fact thatbuyers are concerned with collections’ potentialcommercial success. In this ranking, buyers areasked to give 20 points to what they perceive as themost creative collection for the given season, and 0points are given to collections that are considerednot creative. The editor of the JdT told us that thebuyers are asked the following question: “Pleaseevaluate the creativity of the collections from all ofthe fashion houses that presented in the [e.g., Fall2006] season.” Fashion collections can be consid-ered creative innovations because they consist offinished products based on the implemented ideas,and these finished products are evaluated for theircreativity by third parties (i.e., buyers). Thus, welabeled the resulting dependent variable creativeinnovations.

Each buyer can give points to a maximum of 20houses. The points awarded by the buyers aresummed and yield the final score. The theoreticalmaximum depends on the exact number of buyers:With 70 buyers, for example, it is 1,400 points (i.e., ifall 70 buyers give 20 points to the same collection).We collected both the total number of points receivedby each fashion house and the ratings individual buy-ers gave to each house. This allowed us to calculateinter-rater reliability scores for this measure. Thisscore (Cronbach’s alpha) varies from year to year, butthe average alpha is very high, 0.97.

4 This method of measuring cultural distance has beenused extensively in management research, most notablyto examine the modes of foreign entry (Kogut & Singh,1988), cross-border acquisition performance (Morosini,Shane, & Singh, 1998), choices between licensing andforeign direct investment (Shane, 1994), cross-culturalvariations in the R&D investment intensity (Varsakelis,2001), longevity in international joint ventures (Barkema,Shenkar, Vermeulen, & Bell, 1997), and the formation oftechnological alliances (Steensma, Marino, Weaver, &Dickson, 2000). It assumes that individuals are exposedto the same amount of requisite variety when they workin a country the culture of which is characterized by highdifference on one Hofstede’s dimension to the home cul-ture with little differences on the other dimensions, ascompared to a country the culture of which has moderatedifferences with the home culture on all dimensions. Inother words, theoretically speaking, all dimensions a pri-ori play the same role for creative processes.


To further assess the robustness of the buyers’evaluation, we looked at another creativity rankingdeveloped by the JdT. Although the buyers’ evalu-ation yields the JdT’s flagship ranking, the journalalso collects evaluations from journalists, who areasked the same question as the buyers and evaluatethe same houses. There are 15 expert raters, includ-ing well-known journalists such as the fashioncritic Suzy Menkes. The correlation between thetwo rankings was 0.8, indicating the convergence ofviews between buyers and journalists. We kept thebuyers’ ranking in our main analysis because of thegreater importance of buyers in the fashion indus-try, and because of the much bigger sample of raters(around 70 buyers vs. around 15 journalists), whichwould produce a more reliable result.

Control variables. We controlled for a number oforganizational-level variables in our analyses. First,because the JdT rankings are published twice a yearfor each fashion season (fall/winter and spring/summer), we included season-based fixed effects inour models. In this context, many firms are privatelyowned or affiliated with a conglomerate, and, hence,they either do not release, or release only a partialamount of, their financial information such as reve-nue or number of employees (Caves, 2000). Thus, wehad to rely on a proxy for organizational size. Weconstructed this proxy by counting the number ofdesigns displayed for a given season by a fashionhouse (variable, number of designs). The intuition isthat bigger houses create more designs than dosmaller houses (Crane, 1997).

We also controlled for the amount of media cov-erage of the focal fashion house by tallying thenumber of articles published about each house inmedia outlets for the time period preceding a fash-ion season. We used all 25 languages available inFactiva to avoid geographic bias. We focused onmedia outlets that are centered on fashion (such asVogue) and the fashion-dedicated sections of gen-eralist outlets (for example, the style section of theNew York Times). Media coverage can be consid-ered a proxy of the attention from the stakeholdersto the fashion house (Godart & Mears, 2009). Be-cause media coverage was a highly skewed vari-able, we used a log transformation.

Although our dependent variable, creative inno-vations, is measured at the level of the organization(i.e., the creativity of the collections produced bythe fashion houses), the main driving force behindthe collections are the houses’ creative directors.Thus, we controlled for a number of individual-level variables associated with creative directors:

age (age of creative director), tenure at the direc-tor’s current house (creative director’s tenure(logged)), and the number of different fashionhouses worked at, including the current position(creative director’s number of houses), and whetherthe director’s position was solo or as part of a team(team of creative directors). When a house was runby more than one creative director (slightly morethan 20% of the observations), we used the averageof the relevant variable across individuals for bothcontrol variables and our main independent vari-ables (e.g., breadth, depth, and cultural distance)because our interviews confirmed that creative di-rector teams have to work collaboratively and theconsensus of all individuals is the norm.

We coded whether creative directors lived ab-road prior to starting their careers by computing adummy variable, lived abroad, set to “1” if a personhad such experience and “0” otherwise. We alsocoded whether the creative director studied design,since educational experiences in the domain of de-sign can endow individuals with domain- and cre-ativity-specific skills (higher education in designvariable). As a separate variable, we coded the ed-ucation level (both design-related and non-designrelated degrees) of the creative directors with “1”(! no higher education), “2” (! bachelor’s degree),and “3” (! master’s degree and above) (educationlevel). In order to account for a potential role playedby junior designers, we counted the number foreach fashion house (variable, number of designers(non-CD)) and computed their average age (vari-able, age of designers (non-CD)). The former vari-able accounts for the size of a director’s team—abigger team might potentially improve the creativ-ity of a director by supplying him or her with moresupport and ideas—and the latter is a proxy for theexperience of junior designers. It is important toemphasize, however, that designers other than thecreative directors play a subordinate role.

Analyses Overview

We analyzed our data using two-stage leastsquares regressions with instrumental variables.The choice of this estimation strategy was predi-cated by concerns of reverse causality (endogene-ity) and omitted variables bias (Hamilton & Nicker-son, 2003). That is, we first needed to rule out theexplanation that it is innate creative ability thatleads people to get professional experiencesabroad, and not the other way around. Further-more, there were also psychological variables that


our archival analysis made it impossible for us tocollect, such as individuals’ “openness to experi-ence” (e.g., Huang, Chi, & Lawler, 2005), that mightimpact their willingness to seek professional expe-riences abroad and/or their creativity. These issuescan be interpreted as bias associated with the errorterm of the regression equation examining the driv-ers of creative innovations (Bascle, 2008).

A standard approach to simultaneously deal withboth reverse causality and omitted variables (and,incidentally, measurement errors) is to conduct re-gressions with instrumental variables (Shaver,1998). Instrumental variables have a strong fit withthe endogenous variable (such as going abroad), butdo not correlate with the error term in the equationexamining the dependent variable of interest—inthis case, creative innovations (Murray, 2006).

To perform this regression, a predicted probabil-ity of the endogenous event (i.e., having a creativedirector who went abroad at a helm of a fashionhouse) as a function of instrumental variables plusall other theoretical and control variables in themodel is computed. Then, this probability is en-tered as a control variable in stage two of the re-gression analysis with the ultimate dependent vari-able of interest (i.e., creative innovations of afashion house) without including the instrumentalvariables. Greene (2011: 259–296) shows that theinclusion of this probability absorbs the biases as-sociated with reverse causality and omitted vari-ables, effectively yielding conditions that are as goodas “random assignment” for examining the relation-ships between all independent variables and the de-pendent variable in the second stage of the regression(Wooldridge, 2002). This is why instrumental vari-able regressions are referred to as “quasi-experimen-tal research designs” (Angrist & Krueger, 2001) thatcan make accurate causal inferences from archivaldata and lessen biases due to omitted variables andreverse causality (Bollen, 2012). Recent studies inmanagement that use the instrumental variables in-clude the investigation of how firms conform to thedemands from minority resource suppliers (Durand &Jourdan, 2012), the examination of the effect thatinterorganizational ties across different geographieshave on new firm formation (Bae, Wezel, & Koo,2011), or the impact of social structure on creativity(Fleming, Mingo, & Chen, 2007).5

Because regressions with instrumental vari-ables require an endogenous variable, we created adummy, foreign experience, where a value of “1”indicates that a fashion house has a creative direc-tor with foreign experience, while a value of “0”indicates no foreign experience. This variable hadto include more information than that contained inthe breadth, depth, and cultural distance variables.That is, coding this variable as “0” when someonedid not have any professional foreign experience(i.e., foreign experience ! 0 if breadth ! depth !cultural distance ! 0) will yield regression modelswith prohibitively high correlations between pre-dicted value of foreign experience and the threedimensions of actual professional foreign experi-ence. We incorporated in foreign experience infor-mation on other foreign experiences that an indi-vidual could have had before starting work.Consequently, foreign experience was set to “1” if adesigner had experiences that exposed him or herto foreign cultures, and “0” otherwise.

The second issue is to find instruments that cor-relate with one’s foreign experience and do notcorrelate with the creativity of innovations of one’s

5 Specifically, Fleming et al. (2007) hypothesized thatbrokerage opportunities in the inventors’ collaborativenetworks are positively related to the novelty of their

ideas. The authors used records of inventors’ authorshipson patents as a source of archival data. An inventor is abroker to the extent that he or she collaborates withinventors who do not collaborate with each other. Aninventor is in a cohesive social structure (reverse of bro-kerage) when he or she collaborates with inventors whoalso work with each other. The authors faced an endoge-neity problem because they were concerned that the ex-istence of a creative project might drive the formation ofsocial networks. Thus, social structure (brokerage vs. co-hesion) is the endogenous variable. They chose the num-ber of unique patent lawyers for each inventor’s patentsas an instrument for the brokerage/cohesion instrumentalvariable. Brokers will have many different lawyers be-cause they work with many collaborators from differentcompanies, and different companies use different law-yers. Inventors in cohesive networks tend to work withcollaborators from the same company, and thus use thesame lawyers all the time. Because lawyers are assignedto patents without the inventor’s preference, the numberof lawyers has no influence on the creativity of the pat-ent. Thus, the number of lawyers is a good instrumentthat correlates with the social structure (the number oflawyers increases brokerage), but does not correlate withthe dependent variable—creativity of the patents. Theinclusion of a predicted value for this social structureinstrument (i.e., the number of unique patent lawyers)suppresses the endogeneity concerns in the regressionmodel examining the antecedents of ideas’ novelty.


fashion house. This would refer to factors that leadpeople to get professional or non-professional for-eign experience in the first place. Socioeconomicconditions in their countries of birth might causethem to do so, with either better conditions increas-ing the likelihood of going abroad in some cases(e.g., because of greater ease to obtain visas and buytickets), or worse conditions increasing the likeli-hood in other cases (e.g., out of economic neces-sity). To that end, for each designer in our database,we identified their country and year of birth andthen computed per capita gross domestic product(GDP)—in constant thousands of 2,000 U.S. dol-lars—at the time of birth. Furthermore, our endog-enous variable will be affected by whether a personwas born to bicultural parents. We coded a variableintercultural parents as “1” if a person was born toparents who themselves were born in differentcountries. We also constructed an interaction termbetween intercultural parents and per capita GDPat birth to allow for the synergistic effects of thesetwo variables (e.g., someone who was born in amulticultural family in a rich country might bemore willing and able to get experience abroad).6

Our dataset is a panel based on firm–season ob-servations, and our main dependent variable is acount of points. We transformed our dependentvariable into a variable that can be used by anordinary least squares regression analysis; in thiscase, the xtivreg2 command in STATA. Specifi-cally, creative innovations contained the averagenumber of points given to the fashion house by theraters in a given season. Our data comprised infor-mation over 21 fashion seasons (fall/winter and

spring/summer) between 2000 and 2010. STATA’sxtivreg2 used a fixed effects specification at thelevel of the fashion house. This is the equivalent ofincluding a dummy for each fashion house in theanalysis. Statistically, house fixed effects is an ad-ditional check for omitted variables that might bedriving the results, such as, for example, changes inorganizational culture, and whether these are condu-cive to creativity or not. Our final sample comprised2,427 house–season observations. Table 1 gives anoverview of the descriptive statistics and Pearson cor-relation coefficients for our variables.7 We mean-cen-tered the main effects prior to the construction ofinteractions to avoid collinearity, but provide in Ta-ble 1 the means of the un-centered variables.

We inspected the correlations between creativeinnovations, foreign experience, and the instru-ments (per capita GDP at birth, interculturalparents, and their interaction) to see which in-struments were strongly correlated with the en-dogenous variable (i.e., foreign experience). Percapita GDP at birth was correlated neither withforeign experience (r ! 0.02, ns) nor with creativeinnovations (r ! 0.01, ns). Since both correla-tions were not significant, it was not appropriateto use per capita GDP at birth as an instrument.Intercultural parents was strongly correlatedwith foreign experience (r ! 0.40, p " 0.01),suggesting it as a possibly appropriate instru-ment. The interaction of per capita GDP at birth #Intercultural parents was strongly correlatedwith foreign experience (r ! 0.32, p " 0.01),again suggesting its potential appropriateness.We used the latter two in the first-stage analysis,whereas per capita GDP at birth was included inthe second stage.

We examined whether our interactions and theindividual curvilinear effects were robust against

6 Ultimately, the impact of these variables on goingabroad and on creative innovations is determined by twostatistical tests: the Kleibergen–Paap rk Wald F test andthe Sargan test (Baum, Schaffer, & Stillman, 2003). Thefirst test must be significant because it tests whetherinstrumental variables (per capita GDP at birth, intercul-tural parents, and their interaction) are jointly correlatedwith the endogenous variable (foreign experience). Anon-significant Sargan test shows that a null hypothesisabout the lack of correlation between the instrumentsand the dependent variable (creative performance)should not be rejected. However, the significance of thelatter test does not automatically mean that the model ismis-specified (Bascle, 2008). If the Sargan test is signifi-cant in the intermediate models, but it is not significantin the fully specified model, and the theoretical vari-ables’ results in the fully specified model are the same asin the intermediate models, then the researcher can stillbe confident about the intermediate results.

7 Although these correlations are generally low, oneshould expect high correlations among the squared termsof breadth, depth, and cultural distance. High correla-tions between variables—multicollinearity—is the con-sequence of having redundant information in the regres-sion model, which primarily inflates standard errors thatmay lead the researcher to over-reject a relationship thatexists in the data. When the maximum variance inflationfactor (VIF) is high—that is to say substantially above10—one can retain the explanatory power of the regres-sion model by removing highly collinear variables andseeing whether this affects the results (Belsley, Kuh, &Welsch, 2004; Kennedy, 2008). This is what we did inour analyses, as reported below.


TA

BLE

1D

escr

ipti

veSt

atis

tics

and

Pear

son

Cor

rela

tion

Coe

ffic

ient

s

Var

iabl

esM

ean

SD

12

34

56

78

910

1112

1314

1516

1718

1920

1.C

reat

ive

inno

vati

onsa

1.14

2.31

1.00

2.Fo

reig

nex

peri

ence

0.68

0.47

$0.

081.

003.

Inte

rcul

tura

lpa

rent

s0.

260.

44$

0.05

0.40

1.00

4.Pe

rca

pita

GD

Pat

birt

h#

Inte

rcul

tura

lpa

rent

s15

.34

33.0

60.

010.

320.

791.

00

5.Pe

rca

pita

GD

Pat

birt

h(/

1000

)7.

4445

.08

0.01

0.02

$0.

200.

121.

006.

Num

ber

ofde

sign

s43

.00

14.4

00.

25$

0.04

0.03

$0.

01$

0.22

1.00

7.M

edia

cove

rage

(log

ged)

4.22

1.45

0.42

0.06

0.10

0.05

$0.

130.

351.

008.

Age

ofcr

eati

vedi

rect

or42

.65

10.7

90.

13$

0.09

0.10

$0.

04$

0.42

0.35

0.35

1.00

9.C

reat

ive

dire

ctor

’ste

nure

(log

ged)

2.12

0.84

0.21

$0.

130.

10$

0.03

$0.

270.

300.

240.

591.

0010

.Cre

ativ

edi

rect

or’s

num

ber

ofho

uses

2.93

2.11

0.13

0.24

0.07

0.10

$0.

030.

160.

330.

290.

031.

0011

.Tea

mof

crea

tive

dire

ctor

s0.

200.

40$

0.10

0.01

0.07

0.10

0.09

$0.

07$

0.05

$0.

12$

0.09

$0.

161.

0012

.Hig

her

educ

atio

nin

desi

gn0.

650.

460.

060.

08$

0.03

0.06

0.14

$0.

09$

0.07

$0.

35$

0.19

0.05

$0.

041.

0013

.Edu

cati

onle

vel

1.90

0.40

0.07

$0.

02$

0.09

$0.

050.

17$

0.09

0.02

$0.

23$

0.12

$0.

17$

0.07

0.49

1.00

14.A

geof

desi

gner

s(n

on-C

D)

5.66

12.3

70.

230.

070.

080.

06$

0.11

0.25

0.37

0.14

0.09

0.14

$0.

05$

0.01

0.07

1.00

15.N

umbe

rof

desi

gner

s(n

on-C

D)

0.29

0.67

0.23

0.12

0.08

0.08

$0.

090.

280.

370.

150.

090.

16$

0.05

0.04

0.11

0.78

1.00

16.N

umbe

rof

crea

tive

dire

ctor

s1.

230.

48$

0.11

0.02

0.07

0.10

0.09

$0.

06$

0.05

$0.

10$

0.12

$0.

140.

79$

0.07

$0.

10$

0.02

$0.

041.

0017

.Liv

edab

road

0.12

0.32

$0.

050.

250.

040.

01$

0.19

$0.

07$

0.11

$0.

04$

0.12

$0.

040.

07$

0.04

$0.

06$

0.04

$0.

070.

061.

0018

.Bre

adth

0.83

0.87

0.04

0.64

0.06

0.09

0.12

0.00

0.07

$0.

04$

0.17

0.44

$0.

070.

10$

0.01

0.07

0.11

$0.

060.

201.

0019

.Dep

th6.

299.

400.

060.

460.

250.

19$

0.06

0.08

0.19

0.33

0.16

0.51

$0.

05$

0.10

$0.

180.

090.

11$

0.04

0.19

0.55

1.00

20.C

ultu

ral

dist

ance

$0.

921.

450.

050.

430.

190.

190.

00$

0.04

0.04

0.03

$0.

090.

230.

010.

06$

0.03

0.06

0.08

0.00

0.27

0.44

0.65

1.00

Not

e:C

orre

lati

ons

grea

ter

than

|.03

9|ar

esi

gnif

ican

tat

p"

0.05

.a

Dep

ende

ntva

riab

le.

outliers. To do so, we calculated Cook’s distancestatistics for observations in each of the regressionsinvolving curvilinear effects and the regressionmodel with all the main effects and interactions.We re-ran these regressions excluding observationsthat had higher Cook’s distance statistics than thethreshold (determined as 4/N where N is the num-ber of observations). In all four regressions, theresults were the same as those reported below.

RESULTS

Table 2 provides results of our second-stage re-gression analysis with creative innovations as adependent variable. Model 1 is the baseline. Weentered the linear effects of the three independentvariables of theoretical interest in Model 2. Thenwe entered the squared terms of breadth, depth,and cultural distance separately in Models 3, 4, and5, respectively. Model 6 reports linear and squaredeffects for all three variables.

We start with our basic predictions for the threedimensions of interest: breadth, depth, and culturaldistance. Overall, we predicted that relatively mod-erate levels of each dimension would be associatedwith the highest level of creative innovations. Hy-pothesis 1 predicted an inverted U-shaped relation-ship between breadth and creative innovations.Consistent with this hypothesis, there was a posi-tive linear effect in Model 3 for breadth (5.38, p ".001) and a negative quadratic effect ($1.64, p ".001). Hypothesis 2 predicted an inverted U-shapedrelationship between depth and the creative inno-vations. Consistent with this hypothesis, there wasa significant positive linear effect of depth (0.21,p " .001) and a negative quadratic effect ($0.003, p" .001) in Model 4. Hypothesis 3 predicted aninverted U-shaped relationship between culturaldistance and creative innovations. Consistent withthis hypothesis, there was a significant linear effectfor cultural distance (0.25, p " .05), and a negativequadratic effect ($0.06, p " .001) in Model 5.

All quadratic effects remained when we enteredthe three linear terms and the three quadratic terms inModel 6.8 The only exception was a non-significantlinear coefficient for cultural distance (0.07, ns).

However, Model 6 had a maximum VIF higher than10 (VIF % 12), suggesting that the presence of multi-collinearity might inflate standard errors and reducethe significance of coefficients. In Model 7, we re-moved the quadratic term for breadth, which broughtthe maximum VIF to 9 and both the linear (0.18, p "0.1) and quadratic terms ($0.04, p " .05) for culturaldistance became significant. Even without this checkin Model 7, Aiken and West (1991) suggest that thesignificance of the quadratic term is enough to indi-cate the presence of a curvilinear effect. Thus, all ourhypotheses were supported.

Figure 1 plots the effects from Models 3–5 forbreadth, depth, and cultural distance between onestandard deviation below the mean and four stan-dard deviations above. Breadth and cultural dis-tance showed the predicted inverted-U-shaped cur-vilinear effect on creative innovations, where thepositive effect of each variable eventually de-creased, and even turned negative at very high lev-els for breadth and depth. Thus, consistent withour theorizing, the highest levels of creative inno-vations were seen at relatively moderate levels ofbreadth or cultural distance (when the other twodimensions of professional foreign experience wereat their means). In addition, very high levels ofbreadth and cultural distance began producing det-rimental effects, with the highest levels of eachapproaching levels of creative innovations seen bythose with little or no foreign work experience. Incontrast, depth had a decreasing positive effect oncreative innovations, but did not turn negative.9

8 We performed a major robustness check to accountfor an alternative explanation based on Schneider’s(1987) theory of attraction–selection–attrition (ASA).From this perspective, it is possible that creative direc-tors with foreign professional experiences select or at-tract more talented team members with foreign experi-

ences. To rule out this possibility, we looked at thesubsample of fashion houses run by teams, but found thatdirectors with foreign experience do not systematicallyselect or attract other directors who have foreign experi-ence, casting doubt on the ASA alternative explanation.We were not able, however, to rule out the possibilitythat directors attract, select, and retain junior designerswith foreign experiences, due to the lack of data.

9 We did not examine curvilinear moderations (i.e., theinteractions among quadratic terms) because we do notknow of a method for combining a three-way interactionwith the curvilinear moderation of all three variables. Thiswould require adding at least three interactions of quadraticeffects with the linear terms to Model 8, which already hasthe maximum VIF & 38. Effects in such a model would notbe interpretable. Second, we did not have a theoreticalreason to believe that our inverted U-shaped relationshipswould change shape to a regular U at other combinations ofpoints in our data. As Aiken and West (1991: 69–70) ex-plain, this is the assumption behind testing the model withcurvilinear moderation.


TA

BLE

2R

esul

tsof

Seco

nd-S

tage

Pane

lD

ata

Reg

ress

ion

wit

hIn

stru

men

tal

Var

iabl

es

Mod

el1

Mod

el2

Mod

el3

Mod

el4

Mod

el5

Mod

el6

Mod

el7

Var

iabl

esC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

E

Fore

ign

expe

rien

ce$

10.5

6***

(2.5

0)$

17.9

6***

(4.2

5)$

20.2

4***

(4.6

0)$

19.0

6***

(4.3

9)$

21.0

5***

(5.2

6)$

16.3

9***

(3.4

9)$

20.5

5***

(4.9

0)Pe

rca

pita

GD

Pat

birt

h(/

1000

)$

0.00

***

(0.0

0)$

0.00

***

(0.0

0)$

0.00

***

(0.0

0)$

0.00

***

(0.0

0)$

0.00

***

(0.0

0)$

0.00

***

(0.0

0)$

0.00

***

(0.0

0)N

umbe

rof

desi

gns

$0.

00(0

.00)

$0.

00(0

.00)

$0.

00(0

.00)

$0.

00(0

.00)

$0.

00(0

.00)

$0.

00(0

.00)

$0.

00(0

.00)

Med

iaco

vera

ge(l

ogge

d)0.

18*

(0.0

8)0.

12(0

.11)

0.20

*(0

.10)

0.12

(0.1

1)0.

05(0

.13)

0.23

**(0

.08)

0.08

(0.1

2)A

geof

crea

tive

dire

ctor

$0.

05**

*(0

.01)

$0.

12**

*(0

.03)

$0.

10**

*(0

.02)

$0.

11**

*(0

.03)

$0.

13**

*(0

.03)

$0.

07**

*(0

.02)

$0.

11**

*(0

.03)

Cre

ativ

edi

rect

or’s

tenu

re(l

ogge

d)0.

90**

*(0

.19)

1.28

***

(0.2

8)0.

79**

*(0

.18)

1.07

***

(0.2

5)1.

38**

*(0

.32)

0.49

***

(0.1

3)1.

11**

*(0

.26)

Cre

ativ

edi

rect

or’s

num

ber

ofho

uses

0.72

***

(0.1

6)0.

46**

*(0

.13)

0.56

***

(0.1

5)0.

44**

*(0

.13)

0.58

***

(0.1

7)0.

46**

*(0

.12)

0.52

***

(0.1

5)T

eam

ofcr

eati

vedi

rect

ors

0.06

(0.1

5)0.

00(0

.19)

$0.

33(0

.22)

$0.

14(0

.20)

0.14

(0.2

2)$

0.26

(0.1

8)$

0.04

(0.2

1)H

ighe

red

ucat

ion

inde

sign

$0.

58*

(0.2

9)0.

44(0

.37)

$0.

98*

(0.4

7)0.

41(0

.38)

0.40

(0.4

2)$

0.82

*(0

.40)

0.37

(0.4

0)Ed

ucat

ion

leve

l$

1.45

***

(0.4

2)$

2.42

***

(0.6

3)$

1.61

**(0

.56)

$2.

33**

*(0

.63)

$2.

34**

*(0

.68)

$1.

22*

(0.4

8)$

2.21

***

(0.6

5)A

geof

desi

gner

s(n

on-C

D)

0.02

***

(0.0

1)0.

02**

(0.0

1)0.

01(0

.01)

0.01

'(0

.01)

0.02

*(0

.01)

0.01

'(0

.01)

0.01

'(0

.01)

Num

ber

ofde

sign

ers

(non

-CD

)$

0.07

(0.1

0)$

0.30

**(0

.12)

$0.

14(0

.12)

$0.

28*

(0.1

2)$

0.33

*(0

.13)

$0.

16(0

.10)

$0.

30*

(0.1

3)N

umbe

rof

crea

tive

dire

ctor

s0.

92**

(0.2

9)1.

41**

*(0

.42)

1.37

***

(0.3

9)1.

42**

*(0

.41)

1.56

**(0

.48)

0.94

***

(0.2

8)1.

46**

*(0

.44)

Live

dab

road

2.00

***

(0.4

4)3.

47**

*(0

.75)

5.26

***

(1.1

2)3.

73**

*(0

.80)

4.19

***

(0.9

6)4.

62**

*(0

.92)

4.15

***

(0.9

2)B

read

th(H

ypot

hesi

s1)

2.77

***

(0.6

4)5.

38**

*(1

.26)

2.56

***

(0.5

9)3.

29**

*(0

.81)

4.11

***

(0.9

3)2.

84**

*(0

.68)

Dep

th(H

ypot

hesi

s2)

0.08

**(0

.03)

0.04

*(0

.02)

0.21

***

(0.0

5)0.

10**

(0.0

3)0.

13**

*(0

.03)

0.22

***

(0.0

6)C

ultu

ral

dist

ance

(Hyp

othe

sis

3)0.

20*

(0.0

9)0.

12(0

.08)

0.15

'(0

.09)

0.25

*(0

.11)

0.07

(0.0

7)0.

18'

(0.0

9)B

read

th(s

quar

ed)

(Hyp

othe

sis

1)$

1.64

***

(0.4

3)$

1.25

***

(0.3

2)D

epth

(squ

ared

)(H

ypot

hesi

s2)

$0.

00**

*(0

.00)

$0.

00**

*(0

.00)

$0.

00**

*(0

.00)

Cul

tura

ldi

stan

ce(s

quar

ed)

(Hyp

othe

sis

3)$

0.06

**(0

.02)

$0.

03**

(0.0

1)$

0.04

*(0

.02)

Bre

adth

#D

epth

Bre

adth

#C

ultu

ral

dist

ance

Dep

th#

Cul

tura

ldi

stan

ceB

read

th#

Dep

th#

Cul

tura

ldi

stan

ceO

bser

vati

ons

2427

2427

2427

2427

2427

2427

2427

Kle

iber

gen–

Paap

rkLM

stat

isti

c32

.793

***

23.2

41**

*25

.109

23.8

64**

*19

.437

***

32.7

07**

*21

.547

***

Sarg

an16

.969

***

3.86

9*0.

778

1.19

81.

127

4.97

2*0.

282

Max

VIF

2.98

5.26

10.6

49.

295.

3411

.89.

46N

umbe

rof

hous

es27

027

027

027

027

027

027

0

Not

e:St

anda

rder

rors

inpa

rent

hese

s.**

*p

"0.

001

**p

"0.

01*

p"

0.05

'p

"0.

10

TA

BLE

2(c

onti

nued

)

Mod

el8

Mod

el9

Mod

el10

Mod

el11

Var

iabl

esC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

EC

oeff

icie

ntS

E

Fore

ign

expe

rien

ce$

18.0

0***

(4.9

5)$

24.1

3***

(6.8

4)$

23.8

4***

(6.7

4)$

23.6

7***

(6.6

4)Pe

rca

pita

GD

Pat

birt

h(/

1000

)$

0.00

**(0

.00)

$0.

00**

(0.0

0)$

0.00

***

(0.0

0)$

0.00

**(0

.00)

Num

ber

ofde

sign

s$

0.00

(0.0

0)$

0.00

(0.0

0)$

0.00

(0.0

0)$

0.00

(0.0

0)M

edia

cove

rage

(log

ged)

0.21

*(0

.09)

0.20

'(0

.11)

0.13

(0.1

2)0.

07(0

.14)

Age

ofcr

eati

vedi

rect

or$

0.06

**(0

.02)

$0.

08**

(0.0

3)$

0.09

**(0

.03)

$0.

09**

(0.0

3)C

reat

ive

dire

ctor

’ste

nure

(log

ged)

0.40

**(0

.14)

0.61

**(0

.19)

0.70

***

(0.2

1)0.

82**

(0.2

5)C

reat

ive

dire

ctor

’snu

mbe

rof

hous

es0.

48**

(0.1

6)0.

63**

(0.2

1)0.

59**

(0.2

0)0.

53**

(0.1

7)T

eam

ofcr

eati

vedi

rect

ors

$0.

36'

(0.2

0)$

0.42

'(0

.25)

$0.

41(0

.25)

$0.

43'

(0.2

5)H

ighe

red

ucat

ion

inde

sign

$0.

65'

(0.3

9)$

0.03

(0.4

4)0.

30(0

.43)

0.36

(0.4

4)Ed

ucat

ion

leve

l$

0.98

'(0

.53)

$1.

74*

(0.6

9)$

2.18

**(0

.75)

$1.

36*

(0.6

4)A

geof

desi

gner

s(n

on-C

D)

0.01

(0.0

1)0.

01(0

.01)

0.00

(0.0

1)0.

00(0

.01)

Num

ber

ofde

sign

ers

(non

-CD

)$

0.13

(0.1

2)$

0.08

(0.1

5)$

0.15

(0.1

4)$

0.17

(0.1

4)N

umbe

rof

crea

tive

dire

ctor

s1.

11**

(0.4

0)1.

80**

(0.5

9)1.

77**

(0.5

8)1.

87**

(0.6

2)Li

ving

abro

ad4.

76**

*(1

.26)

5.73

***

(1.6

2)5.

16**

*(1

.45)

4.47

***

(1.1

2)B

read

th(H

ypot

hesi

s1)

3.77

***

(0.9

2)2.

55**

(0.8

0)3.

12**

(0.9

7)2.

61**

(0.8

3)D

epth

(Hyp

othe

sis

2)0.

18**

(0.0

6)0.

36**

*(0

.10)

0.33

***

(0.0

9)0.

25**

*(0

.06)

Cul

tura

ldi

stan

ce(H

ypot

hesi

s3)

0.56

*(0

.25)

1.25

**(0

.38)

1.10

**(0

.34)

1.78

**(0

.59)

Bre

adth

(squ

ared

)(H

ypot

hesi

s1)

$1.

11**

*(0

.22)

Dep

th(s

quar

ed)

(Hyp

othe

sis

2)$

0.00

(0.0

0)C

ultu

ral

dist

ance

(squ

ared

)(H

ypot

hesi

s3)

$0.

03*

(0.0

2)B

read

th#

Dep

th$

0.01

(0.0

6)$

0.23

*(0

.10)

$0.

15'

(0.0

8)B

read

th#

Cul

tura

ldi

stan

ce$

0.30

(0.2

3)$

0.98

**(0

.36)

$0.

91**

(0.3

4)$

1.36

**(0

.51)

Dep

th#

Cul

tura

ldi

stan

ce$

0.07

***

(0.0

2)$

0.12

***

(0.0

3)$

0.03

*(0

.01)

$0.

12**

*(0

.03)

Bre

adth

#D

epth

#C

ultu

ral

dist

ance

0.02

(0.0

2)0.

09**

(0.0

3)0.

04**

(0.0

2)O

bser

vati

ons

2427

2427

2427

2427

Kle

iber

gen–

Paap

rkLM

stat

isti

c18

.984

***

15.5

38**

*15

.54*

**15

.243

***

Sarg

an4.

287*

0.01

20.

018

0.02

5M

axV

IF31

.21

13.4

59.

6910

.49

Num

ber

ofho

uses

270

270

270

270

Not

e:St

anda

rder

rors

inpa

rent

hese

s.**

*p

"0.

001

**p

"0.

01*

p"

0.05

'p

"0.

10

Three-Way Interaction between Breadth, Depth,and Cultural Distance

We also ran post-hoc analyses to test whetherthere was a three-way interaction among breadth,depth, and cultural distance indicating their jointeffects on creative innovations. We tested for thepresence of a three-way interaction between linearterms of each dimension in Model 8. We also in-cluded lower-level terms (i.e., all two-way interac-tions among the three variables). The maximumVIF in this model was very high, due to the corre-lations between the interactions and the quadratic

effects, and many coefficients were not significant;thus, results in Model 8 cannot be interpreted. InModel 9, we eliminated the quadratic effects, be-cause they were redundant with the interactions.Even though maximum VIF is 13 in Model 9, thethree-way interaction was significant. Since the ef-fects of collinearity appear in inflated standard er-rors (Kennedy, 1998), we can still interpret signif-icant coefficients in Model 9. This model shows apositive three-way interaction between depth,breadth, and cultural distance (0.09, p " .01). Thus,the interactions of our foreign professional experi-ence variables appear to explain the effects attrib-utable to their individual quadratic terms.

We also checked whether collinearity affectedour results in Model 9. To that end, in Model 10, weremoved the three-way interaction to inspect thesignificance of the two-way interactions among the-oretical variables, whereas, in Model 11, we re-moved one two-way interaction (breadth # depth)to inspect the three-way effect. None of thesechanges had any impact on the coefficients of theremaining variables, even though collinearity wasreduced to 10. Thus, we can conclude that col-linearity does not affect our results in Model 9.

In order to better understand effects in Model 9,we plotted the relationships among three variablesat plus or minus one standard deviation around themean of breadth. We also tested for the significanceof slopes (Dawson & Richter, 2006). Figure 2 con-tains this plot, constructed using the coefficientestimates from Model 9 as well as the slope signif-icance tests. The plot and corresponding tests indi-cate that depth of foreign work experience seems tobe the most critical of the three factors for obtainingcreative innovations. When depth was high, cul-tural distance and breadth had essentially no effecton creative innovations. However, when depth waslow, breadth and cultural distance had a morenoticeable impact, but they also seemed to act assubstitutes for providing variety: having one orthe other seemed to be helpful, but the combina-tion of both did not provide additional benefits.This substitutability occurs presumably becauseeither breadth or distance may be enough to pro-vide the requisite variety needed for creative in-novations. High depth, on the other hand, mayprovide enough exposure to variety such thatbreadth and cultural distance are no longer crit-ical. In addition, because depth also provides theopportunity to adapt and integrate different cul-tural elements that breadth and cultural distance

FIGURE 1Main Effects of Breadth, Depth,

and Cultural Distance

Note: Since STATA’s xtivreg2 does not report the constantterm in the analysis with the fixed effects, we can interpret onlythe relative and not absolute values of innovations’ creativity onthe y-axis.


do not, depth emerged as the most critical factorin our analyses.

In summary, the significant linear and quadraticterms of our three dimensions show that moderatelevels of breadth, depth, and cultural distance areassociated with the highest levels of creative inno-vations, though the diminishing effect at high lev-els of depth was not pronounced. It is important tonote that these effects show the individual relation-ship between one theoretical variable and creativeinnovations when the other two are at their mean.For example, when a person has average breadthand depth, this person will benefit from averagecultural distance. Since the effects of depth neverturn negative, a person with average breadth anddistance will still benefit from high depth. Thelinear three-way interaction shows a joint effect ofthe combination of three variables and points totwo conclusions: (1) depth seems to be the mostimportant dimension for creative innovations, and(2) breadth and cultural distance are also impor-tant, but primarily at low levels of depth, wherethey act as substitutes. Thus, it seems that the high-est level of creative innovations was achievedwhen high depth was coupled with moderatebreadth and moderate cultural distance.

DISCUSSION AND CONCLUSIONS

The present study is the first to demonstrate thatcompanies benefit when senior leaders in charge ofcreative operations have foreign professional experi-ence. We presented a new theoretical model—theforeign experience model of creative innovations—toshow how the breadth, depth, and cultural distanceof such experiences affect organizational output. Ourresults revealed that these three dimensions of for-eign professional experiences had both independentand joint effects.

When taken individually, depth, breadth, andcultural distance of foreign professional experi-ences had curvilinear relationships with creativeinnovations. The positive effect of breadth and cul-tural distance increased but eventually turned neg-ative at high levels, whereas depth initially in-creased but then showed a decreasing positiveeffect at high levels, though its effect never turnednegative. We demonstrate, for the first time, curvi-linear effects on any of these three dimensions,extending previous research that has typicallyfound linear effects—in particular, as far as depth isconcerned (Maddux & Galinsky, 2009). This devel-opment was made possible because we looked at

FIGURE 2Three-Way Interaction among Breadth, Depth, and Cultural Distance (!1 Standard Deviation)


large-scale, historical field data that gave us abroader, more complex, and more nuanced per-spective than have previous laboratory experi-ments. In addition, our theoretical model helps ex-plicate exactly when and why the positive effects ofbreadth, depth, and cultural distance level offand/or turn negative at high levels.

Our supplementary analyses established how theeffects of these three variables interacted. Depthemerged as the most important dimension for cre-ative innovations, whereas breadth and culturaldistance mattered only when depth was low, actingas substitutes for each other. High depth presum-ably compensates for the negative effects of havingtoo much requisite variety when breadth and dis-tance are too high, because high depth facilitatesadaptation. Yet, once high levels of depth arereached, there is little or no added benefit ofbreadth or cultural distance. In other words, depthmay also provide enough requisite variety to renderbreadth and cultural distance less important. Thefinding that depth was the most important dimen-sion for determining creative innovations is consis-tent with Maddux and Galinsky (2009), who foundthat, the longer undergraduate and MBA studentshad lived abroad, the more creative they were onstandard psychological tests of creativity.

Our theoretical model also suggests why depth ofexperiences may be so critical: Deep foreign expe-riences not only afford the opportunity for greateradaptation to one’s foreign experience, but, becausethey are, by definition, occurring in a foreign coun-try, they will also offer exposure to a variety ofinputs to the creative process. Breadth and culturaldistance, on the other hand, may help provide ex-posure to variety, but not the opportunity to adapt,which is why they seem to be important at low butnot high levels of depth. This means that, when onelacks depth of foreign experiences, one can stillimprove the probability of creative innovations byseeking either greater breadth or cultural distance.

Finally, our results make clear that by far thelowest level of creative innovations were seen atlow levels of all three variables. Thus, having noforeign experience puts one at a distinct disadvan-tage relative to others with different types of foreignwork experiences.

Theoretical Contributions

We believe our findings contribute to a numberof literatures, including those on creativity, workexperience, diversity, and cross-cultural manage-

ment. Past research on the psychology of creativityhas produced a valuable body of knowledge aboutfactors stimulating the generation of novel and use-ful ideas (Amabile, 1996; Baas, De Dreu, & Nijstad,2008; Baer, Leenders, Oldham, & Vadera, 2010;George, 2007; Shalley & Gilson, 2004; Shin et al.,2012; Zhou & Shalley, 2003). One key insight fromthis research is that creativity requires a variety ofinputs. Our research shows how several dimen-sions of foreign experience can help to provide thisvariety. However, our results also go beyond suchwork to show that deeper experiences provide thecritical opportunity for psychological transforma-tion to make sense of these diverse inputs, as wellas the ability to embed oneself in professional net-works to produce creative innovations.

Indeed, a key contribution of our theory and find-ings to the literature on the social psychology ofcreativity (Amabile, 1996)—and especially the lit-erature on the relationship between creativity andculture (Hennessey & Amabile, 2010)—is that cer-tain types of individual experiences to differentcultures can actually affect the creative output oforganizations. Not all types of foreign work experi-ences (and, presumably, other types of novel envi-ronments) are equally effective at leading to cre-ative innovations: Our study shows that it is theinteraction between exposure to novelty, the abilityto adapt to a foreign culture, and the ability tobecome embedded in diverse professional net-works that are critical for translating individuals’foreign work experiences into creative innovationsin organizations. Thus, it is not just a matter ofworking or having any type of experience abroadthat affects subsequent creativity: it is the combi-nation of specific aspects that helps derive a cre-ative benefit.

Research on the sociology of creativity (Fleminget al., 2007) can also benefit from these findings. Arange of work has advanced purely structural ex-planations for creativity and innovations (e.g.,J. A. C. Baum et al., 2003; Cattani & Ferriani, 2008;Godart et al., 2014; Uzzi & Spiro, 2005). Yet, cre-ativity occurs at the intersection of psychologicalprocesses and the social and organizational con-text. Foreign professional experiences are very im-portant in this regard, because they affect not onlythe individuals’ cognition and motivations but alsoshape their professional networks. These experi-ences can channel diverse information to mindsthat are ready to absorb this information, to take therisks with new ideas, and to build intra-organiza-tional coalitions for their implementation. Thus,


the creativity of scientific teams (Fleming, 2001), ofBroadway musical producers (Uzzi & Spiro, 2005),or of movie makers (Cattani & Ferriani, 2008) mayall be shaped by their members’ foreign profes-sional experiences.

Another contribution is to the literature on workexperience (Quiñones, Ford, & Teachout, 1995),which examines how professional lives are shapedby contextual and individual factors, as well ashow such experiences are translated into work-based knowledge, skills, attitudes, motivation, andperformance (Tesluk & Jacobs, 1998). One key con-tribution that we make is to show that foreign pro-fessional experience is an important factor provid-ing an individual with unique skills, attitudes, andmotivation both for the generation and the imple-mentation of creative ideas. We also show that thelack of this experience is detrimental to one’s abil-ity to produce creative innovations. Importantly,while the work experience literature focusing oninternational assignments and cross-cultural ad-justments highlights the rewards and costs associ-ated with foreign experiences (Takeuchi, Tesluk,Yun, & Lepak, 2005), we explicitly decompose suchexperiences into the three dimensions of depth,breadth, and cultural distance, and integrate re-wards and costs in the underlying mechanisms ex-plaining the three curvilinear effects.

The literature on diversity, which examines theconditions under which diversity is beneficial(Joshi & Roh, 2009), can also benefit from our find-ings that suggest that professional foreign experi-ences can be a critical source of diversity of inputsinto the creative process. Organizations hiring in-dividuals with such experiences to lead their cre-ative operations may be more capable at bringingabout creative innovations. Also, professional for-eign experiences expose individuals to a variety ofdifferent approaches to solving problems, whichmay make them more proficient at other types oforganizational tasks that demand creative thinking,such as working well in diverse or geographicallydistributed teams, or negotiating or resolving inter-personal conflicts. Such experiences can providean extra dimension of diversity not yet emphasizedin the literature. For example, a team might becomprised of only White males born in the samecountry (Ibarra, 1993), but it can still exhibit aconsiderable diversity if its members have broadprofessional foreign experiences.

Finally, cross-cultural management scholars areinterested in what makes some individuals betterthan others at communicating across cultures

(Meyer, 2014; Molinsky, 2007). Our results suggestthat broad, deep, and culturally distant profes-sional foreign experiences may enable individualsto engage in “cross-cultural code-switching” (Mo-linsky, 2007), which is a critical component ofmanaging across cultures. These individuals canfacilitate intercultural collaboration by acting asbridges between colleagues, business units, or evenalliance partners from disparate cultural or na-tional contexts.

Limitations and Future Research

Although we ruled out alternative explanationsby adding theoretically relevant controls to our re-gression models, such as individuals’ ages, gender,and previous experiences, as well as organization-level characteristics that might impact the creativeinnovations, and although our use of instrumentalvariables allows us to make causal claims whenanalyzing archival data (Angrist, Imbens, & Rubin,1996; Winship & Morgan, 1999), our study alsocontains a number of limitations. Regarding thegeneralizability of our findings, we would expectsimilar results in knowledge-intensive and creativeindustries, where success is determined by the abil-ity of individuals to generate and implement noveland useful ideas, such as music, publishing, cin-ema, or art (Caves, 2000), or even technology sec-tors and pharmaceutical R&D. Indeed, there aremany companies and industries where the creativ-ity of the final, implemented products is what isevaluated, rather than the creativity of each ideagenerated during the development process. Novels,Hollywood movies, video games, mobile phones,and computer software would all fall into this cat-egory. Of course, our results are less relevant forindustries that are not as dependent on creativeinnovations for survival.

We are also limited to contexts in which oneindividual—or a small team—has a significant im-pact on the output of firms. This will happen whena firm is small, or when the decision making of alarge firm is centralized and confined to a smallgroup of senior executives (Staw, 1980). Many ad-mired companies have individuals with outsizedinfluence over the final output; this is especiallytrue for movie directors, but can also be the case forinfluential CEOs of other types of companies, suchas those in technology (i.e., Steve Jobs, Bill Gates,or Mark Zuckerberg). It may also happen in firmsrun by larger teams characterized by four mecha-nisms of “collective creativity”: members’ willing-


ness to seek help, give help, to reframe collectiveexperiences, and to provide support to one another(Hargadon & Bechky, 2006). Archival researchdoes not allow access to creative teams’ thoughtprocesses, conversations, and the emergence of col-lective cognitions (Weick & Roberts, 1993). We wel-come future studies using in-depth field research orexperiments. Such approaches could also discoverexactly when all mechanisms in our model—expo-sure, adaptation, and embeddedness—are neces-sary for transforming experiences of organizationalleaders into creative innovations.

Our use of instrumental variables allowed us tocontrol for biases due to omitted variables; forexample, directors’ creative self-efficacy, open-ness of experience, or networking ability. Unfor-tunately, we were not able to examine directlywhich part of creative innovations can be attrib-uted to individuals’ networking or self-efficacy,and which is due to the openness of experienceor net of the professional foreign experience. Forexample, given recent evidence that shows thatindividuals high in openness to experience aremore likely to sojourn for extended periods oftime (Zimmermann & Neyer, 2013), our findingssuggest that openness may typically be associatedwith individual creativity because it can increasethe likelihood of having the types of novel expe-riences that broaden and deepen individuals’perspectives. Thus, our work adds to traditionalintra-psychic explanations of creativity by con-sidering the cognitive effect of facilitating diver-gent thinking, recognizing patterns across dispa-rate ideas, and integrating various experiences asa result of individuals’ working abroad.

We assumed that foreign professional experi-ences affect both the generation and implementa-tion of novel ideas, and, as a result, produce cre-ative innovations. While existing research suggeststhat this is a reasonable assumption, and the exter-nal audience members in our study were asked toevaluate creativity of implemented ideas, we can-not cleanly test this assumption with the currentdataset. A limitation of archival research is that itcannot distinguish between ideas that were gener-ated but not implemented. In our case, this wouldhave required collecting all designers’ drawingsover 21 seasons, including all of the intermediatedesigns that they came up with as well as theexact sources for each. Given these logistical im-possibilities, the assumption about the impact offoreign professional experiences on the genera-tion and implementation of creative ideas will

benefit from testing using experimental work,surveys, participant observation, or interviews(Clegg et al., 2002). However, the consistencybetween our findings and those in the extantpsychological literature give us confidence in thegeneral validity of our results.

Finally, our measure of depth is the duration ofexperiences abroad, which is admittedly more aboutquantity than quality. In future studies, the depth ofimmersion abroad could be evaluated, for example,via an assessment of the size and intensity of net-works and ties formed in the country, and whetherthe focal individuals learned local languages and cus-toms, among many other possibilities.

Practical Implications for Organizationsand Individuals

Our results suggest a number of practical impli-cations. First, companies produce more creativeinnovations if their leaders have professional expe-riences abroad. Although hiring executives withsuch career profiles is relatively straightforward,developing talent internally may require institutinginternational rotational programs into human re-sources policies, such as mandatory internationalassignments for those in management and leader-ship positions (Kopp, 1994).

Second, individuals who want to enhance theircreativity might proactively look for work abroad forsubstantial periods of time. Doing this would not onlyincrease their creativity, but also their appeal to or-ganizations as hires (Brimm, 2010).

CONCLUSIONS

Although aspiring designers and prospectiveleaders inside and outside of the fashion industrymight never be able to exactly replicate the creativ-ity of the world’s best creative directors, they canincrease their odds of creative successes by capital-izing on the multicultural aspects of their careerpaths. Foreign experiences, especially ones withsignificant duration and spread across a number ofculturally diverse countries, may provide careerboosts to individuals and help enhance the creativ-ity of the organizations they join. The first steptowards being the next Karl Lagerfeld might startwith something as simple as finding an opportunityto work abroad.


REFERENCES

Aiken, L. S., & West, D. S. G. 1991. Multiple regression:Testing and interpreting interactions. London,U.K.: Sage Publications, Inc.

Amabile, T. M. 1982. Social psychology of creativity: Aconsensual assessment technique. Journal of Per-sonality and Social Psychology, 43: 997–1013.

Amabile, T. M. 1996. Creativity in context. Boulder, CO:Westview Press.

Anderson, N., De Dreu, C. K. W., & Nijstad, B. A. 2004.The routinization of innovation research: A con-structively critical review of the state-of-the-science.Journal of Organizational Behavior, 25: 147–173.

Angrist, J. D., Imbens, G. W., & Rubin, D. B. 1996. Iden-tification of causal effects using instrumental vari-ables. Journal of the American Statistical Associa-tion, 91: 444–455.

Angrist, J. D., & Krueger, A. B. 2001. Instrumental vari-ables and the search for identification: From supplyand demand to natural experiments. Journal of Eco-nomic Perspectives, 15: 69–85.

Aspers, P., & Godart, F. 2013. Sociology of fashion: Orderand change. Annual Review of Sociology, 39: 171–192.

Baas, M., De Dreu, C. K., & Nijstad, B. A. 2008. A meta-analysis of 25 years of mood-creativity research: He-donic tone, activation, or regulatory focus? Psycho-logical Bulletin, 134: 779–806.

Bae, J., Wezel, F. C., & Koo, J. 2011. Cross-cutting ties,organizational density, and new firm formation inthe U.S. biotech industry, 1994–98. Academy ofManagement Journal, 54: 295–311.

Baer, M. 2010. The strength-of-weak-ties perspective oncreativity: A comprehensive examination and exten-sion. Journal of Applied Psychology, 95: 592–601.

Baer, M. 2012. Putting creativity to work: The implemen-tation of creative ideas in organizations. Academy ofManagement Journal, 55: 1102–1119.

Baer, M., Leenders, R. T. A. J., Oldham, G. R., & Vadera,A. K. 2010. Win or lose the battle for creativity: Thepower and perils of intergroup competition. Acad-emy of Management Journal, 53: 827–845.

Barkema, H. G., Shenkar, O., Vermeulen, F., & Bell, J. H. J.1997. Working abroad, working with others: Howfirms learn to operate international joint ventures.Academy of Management Journal, 40: 426–442.

Barkey, K., & Godart, F. C. 2013. Empires, federated ar-rangements, and kingdoms: Using political modelsof governance to understand firms’ creative perfor-mance. Organization Studies, 34: 79–104.

Bascle, G. 2008. Controlling for endogeneity with instru-

mental variables in strategic management research.Strategic Organization, 6: 285–327.

Baum, C. F., Schaffer, M. E., & Stillman, S. 2003. Instru-mental variables and GMM: Estimation and testing.STATA Journal, 3: 1–31.

Baum, J. A. C., Shipilov, A. V., & Rowley, T. J. 2003.Where do small worlds come from? Industrial andCorporate Change, 12: 697–725.

Belsley, D. A., Kuh, E., & Welsch, R. E. 2004. Regressiondiagnostics: Identifying influential data andsources of collinearity. New York, NY: Wiley.

Bhaskar-Shrinivas, P., Harrison, D. A., Shaffer, M. A., &Luk, D. M. 2005. Input-based and time-based modelsof international adjustment: Meta-analytic evidenceand theoretical extensions. Academy of Manage-ment Journal, 48: 257–281.

Black, J. S., Mendenhall, M., & Oddou, G. 1991. Towarda comprehensive model of international adjustment:An integration of multiple theoretical perspectives.Academy of Management Review, 16: 291–317.

Blumer, H. 1969. Fashion: From class differentiation tocollective selection. Sociological Quarterly, 10:275–291.

Bollen, K. A. 2012. Instrumental variables in sociologyand the social sciences. Annual Review of Sociol-ogy, 38: 37–72.

BrandZ. 2013. Top 100 most valuable global brands2013. New York, NY: MillwardBrown Optimor.

Breward, C. 2003. Fashion. Oxford; New York, NY:Oxford University Press.

Brimm, L. 2010. Global cosmopolitans: The creativeedge of difference. Basingstoke, U.K.; New York,NY: Palgrave Macmillan.

Bunderson, J. S., & Sutcliffe, K. M. 2002. Comparingalternative conceptualizations of functional diver-sity in management teams: Process and performanceeffects. Academy of Management Journal, 45: 875–893.

Burt, R. S. 2004. Structural holes and good ideas. Amer-ican Journal of Sociology, 110: 349–399.

Campbell, D. T. 1960. Blind variation and selective re-tention in creative thought as in other knowledgeprocesses. Psychological Review, 67: 380–400.

Cattani, G., & Ferriani, S. 2008. A core/periphery per-spective on individual creative performance: Socialnetworks and cinematic achievements in the Holly-wood film industry. Organization Science, 19: 824–844.

Caves, R. E. 2000. Creative industries: Contracts be-tween art and commerce. Cambridge, MA; London,U.K.: Harvard University Press.

Chua, R. Y., & Iyengar, S. S. 2008. Creativity as a matter


of choice: Prior experience and task instruction asboundary conditions for the positive effect of choiceon creativity. Journal of Creative Behavior, 42:164–180.

Clegg, C., Unsworth, K., Epitropaki, O., & Parker, G. 2002.Implicating trust in the innovation process. Journalof Occupational and Organizational Psychology,75: 409–422.

Crane, D. 1997. Globalization, organizational size, andinnovation in the French luxury fashion industry:Production of culture theory revisited. Poetics, 24:393–414.

Crane, D. 1999. Diffusion models and fashion: A reassess-ment. Annals of the American Academy of Politi-cal and Social Science, 566: 13–24.

Crane, D., & Bovone, L. 2006. Approaches to materialculture: The sociology of fashion and clothing. Po-etics, 34: 319–333.

Cross, R. L., & Parker, A. 2004. The hidden power ofsocial networks: Understanding how work reallygets done in organizations. Boston, MA: HarvardBusiness Press.

Currid, E. 2007. The Warhol economy: How fashion,art, and music drive New York City. Princeton, NJ:Princeton University Press.

Damanpour, F. 1991. Organizational innovation: A meta-analysis of effects of determinants and moderators.Academy of Management Journal, 34: 555–590.

Dane, E. 2010. Reconsidering the trade-off between ex-pertise and flexibility: A cognitive entrenchmentperspective. Academy of Management Review, 35:579–603.

Danis, W. M., & Shipilov, A. V. 2002. A comparison ofentrepreneurship development in two post-communistcountries: The cases of Hungary and Ukraine. Journalof Developmental Entrepreneurship, 7: 67–94.

Dawson, J. F., & Richter, A. W. 2006. Probing three-wayinteractions in moderated multiple regression: De-velopment and application of a slope difference test.Journal of Applied Psychology, 91: 917–926.

Dewar, R. D., & Dutton, J. E. 1986. The adoption of radicaland incremental innovations: An empirical analysis.Management Science, 32: 1422–1433.

Durand, R., & Jourdan, J. 2012. Jules or Jim: Alternativeconformity to minority logics. Academy of Manage-ment Journal, 55: 1295–1315.

Dutton, J. E., & Ashford, S. J. 1993. Selling issues to topmanagement. Academy of Management Review,18: 397–428.

Edwards, S. M. 2001. The technology paradox: Efficiencyversus creativity. Creativity Research Journal, 13:221–228.

Feist, G. J. 1998. A meta-analysis of personality in scien-tific and artistic creativity. Personality and SocialPsychology Review, 2: 290–309.

Feist, G. J. 1999. Personality in scientific and artisticcreativity. In R. J. Sternberg (Ed.), Handbook of cre-ativity: 273–296. New York, NY: Cambridge Univer-sity Press.

Ferris, G. R., Treadway, D. C., Kolodinsky, R. W.,Hochwarter, W. A., Kacmar, C. J., Douglas, C., &Frink, D. D. 2005. Development and validation of thepolitical skill inventory. Journal of Management,31: 126–152.

Fleming, L. 2001. Recombinant uncertainty in technolog-ical search. Management Science, 47: 117–132.

Fleming, L., Mingo, S., & Chen, D. 2007. Collaborativebrokerage, generative creativity, and creative suc-cess. Administrative Science Quarterly, 52: 443–475.

Galunic, D. C., & Rodan, S. 1998. Resource recombina-tions in the firm: Knowledge structures and the po-tential for Schumpeterian innovation. StrategicManagement Journal, 19: 1193–1201.

George, J. M. 2007. Creativity in organizations. The Acad-emy of Management Annals, 1: 439–477.

Godart, F. 2012a. Unveiling fashion: Business, culture,and identity in the most glamorous industry. Bas-ingstoke, U.K.; New York, NY: Palgrave Macmillan.

Godart, F. C. 2012b. Trend networks: Multidimensionalproximity and the formation of aesthetic choices inthe creative economy. Regional Studies: 1–12.

Godart, F. C., & Mears, A. 2009. How do cultural produc-ers make creative decisions? Lessons from the cat-walk. Social Forces, 88: 671–692.

Godart, F. C., Shipilov, A. V., & Claes, K. 2014. Makingthe most of the revolving door: The impact of out-ward personnel mobility networks on organizationalcreativity. Organization Science, 25: 377–400.

Grant, A. M., & Schwartz, B. 2011. Too much of a goodthing: The challenge and opportunity of the invertedU. Perspectives on Psychological Science, 6: 61–76.

Greene, W. H. 2011. Econometric analysis (7th ed.). NewYork, NY: Prentice Hall.

Guimerà, R., Uzzi, B., Spiro, J., & Nunes Amaral, L. A.2005. Team assembly mechanisms determine collab-oration network structure and team performance.Science, 308: 697–702.

Gurteen, D. 1998. Knowledge, creativity and innovation.Journal of Knowledge Management, 2: 5–13.

Hamilton, B. H., & Nickerson, J. A. 2003. Correcting forendogeneity in strategic management research. Stra-tegic Organization, 1: 51–78.

Hammond, M. M., Neff, N. L., Farr, J. L., Schwall, A. R.,


& Zhao, X. 2011. Predictors of individual-level inno-vation at work: A meta-analysis. Psychology of Aes-thetics, Creativity and the Arts, 5: 90–105.

Hargadon, A. B., & Bechky, B. A. 2006. When collectionsof creatives become creative collectives: A fieldstudy of problem solving at work. Organization Sci-ence, 17: 484–500.

Hechanova, R., Beehr, T. A., & Christiansen, N. D. 2003.Antecedents and consequences of employees’ ad-justment to overseas assignment: A meta-analyticreview. Applied Psychology, 52: 213–236.

Henderson, R. M., & Clark, K. B. 1990. Architecturalinnovation: The reconfiguration of existing producttechnologies and the failure of established firms.Administrative Science Quarterly, 35: 9–30.

Hennessey, B. A., & Amabile, T. M. 2010. Creativity.Annual Review of Psychology, 61: 569–598.

Hofstede, G. 1980. Culture’s consequences: Interna-tional differences in work-related values. New-bury Park, CA: Sage Publications.

Hofstede, G., Hofstede, G. J., & Minkov, M. 2010. Cul-tures and organizations: Software of the mind.New York, NY: McGraw Hill Professional.

House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., &Gupta, V. (Eds.). 2004. Culture, leadership and or-ganizations: The GLOBE study of 62 societies.Thousand Oaks, CA: Sage.

Huang, T., Chi, S., & Lawler, J. J. 2005. The relationshipbetween expatriates’ personality traits and their ad-justment to international assignments. The Interna-tional Journal of Human Resource Management,16: 1656–1670.

Ibarra, H. 1993. Personal networks of women and minor-ities in management: A conceptual framework.Academy of Management Review, 18: 56–87.

Joshi, A., & Roh, H. 2009. The role of context in workteam diversity research: A meta-analytic review.Academy of Management Journal, 52: 599–627.

Kandogan, Y. 2012. An improvement to Kogut and Singhmeasure of cultural distance considering the rela-tionship among different dimensions of culture. Re-search in International Business and Finance, 26:196–203.

Kawamura, Y. 2005. Fashion-ology: An introduction tofashion studies. New York, NY: Berg.

Kawamura, Y. 2011. Doing research in fashion anddress: An introduction to qualitative methods. NewYork, NY: Berg.

Kennedy, P. 1998. A guide to econometrics (4th ed.).Cambridge, MA: MIT Press.

Kennedy, P. 2008. A guide to econometrics (6th ed.).Malden, MA: Blackwell.

Kogut, B., & Singh, H. 1988. The effect of national cultureon the choice of entry mode. Journal of Interna-tional Business Studies, 19: 411–432.

Kopp, R. 1994. International human resource policiesand practices in Japanese, European, and UnitedStates multinationals. Human Resource Manage-ment, 33: 581–599.

Kostova, T., & Roth, K. 2003. Social capital in multina-tional corporations and a micro-macro model of itsformation. Academy of Management Review, 28:297–317.

Lambert, W. E., Tucker, G. R., & d’Anglejan, A. 1973.Cognitive and attitudinal consequences of bilingualschooling. Journal of Educational Psychology, 65:141–159.

Laursen, K., Masciarelli, F., & Prencipe, A. 2012. Regionsmatter: How localized social capital affects innova-tion and external knowledge acquisition. Organiza-tion Science, 23: 177–193.

Leung, A. K.-Y., & Chiu, C.-Y. 2010. Multicultural expe-rience, idea receptiveness, and creativity. Journal ofCross Cultural Psychology, 41: 723–741.

Leung, A. K., Maddux, W. W., Galinsky, A. D., & Chiu,C. Y. 2008. Multicultural experience enhances cre-ativity—The when and how. American Psycholo-gist, 63: 169–181.

MacKinnon, D. W. 1978. In search of human effective-ness: Identifying and developing creativity. Buf-falo, NY: Bearly Limited.

Maddux, W. W. 2011. A moveable feast: How transfor-mational cross-cultural experiences facilitate cre-ativity. In R. Kramer, G. J. Leonardelli & R. Livings-ton (Eds.), Festschrift volume in honor of MarilynnB. Brewer: 339–360. New York, NY: PsychologyPress.

Maddux, W. W., Adam, H., & Galinsky, A. D. 2010. Whenin Rome. . . learn why the Romans do what they do:How multicultural learning experiences enhancecreativity. Personality and Social Psychology Bul-letin, 36: 731–741.

Maddux, W. W., & Galinsky, A. D. 2009. Cultural bordersand mental barriers: The relationship between livingabroad and creativity. Journal of Personality andSocial Psychology, 96: 1047–1061.

Mednick, S. A. 1962. The associative basis of the creativeprocess. Psychological Review, 69: 220–232.

Meyer, E. 2014. The culture map: Breaking through theinvisible boundaries of global business. Philadel-phia, PA: Public Affairs.

Molinsky, A. 2007. Cross-cultural code-switching: Thepsychological challenges of adapting behavior in for-eign cultural interactions. Academy of Manage-ment Review, 32: 622–640.


Morosini, P., Shane, S., & Singh, H. 1998. National cul-tural distance and cross-border acquisition perfor-mance. Journal of International Business Studies,29: 137–158.

Morris, M. W., Podolny, J., & Sullivan, B. N. 2008. Cul-ture and coworker relations: Interpersonal patternsin American, Chinese, German, and Spanish divi-sions of a global retail bank. Organization Science,19: 517–532.

Murray, M. P. 2006. Avoiding invalid instruments andcoping with weak instruments. Journal of EconomicPerspectives, 20: 111–132.

Nemeth, C. J., & Kwan, J. L. 1987. Minority influence,divergent thinking and detection of correct solu-tions. Journal of Applied Social Psychology, 17:788–799.

Obstfeld, D. 2005. Social networks, the tertius iungensorientation, and involvement in innovation. Admin-istrative Science Quarterly, 50: 100–130.

Oettl, A., & Agrawal, A. 2008. International labor mobil-ity and knowledge flow externalities. Journal of In-ternational Business Studies, 39: 1242–1260.

Oh, H., Chung, M.-H., & Labianca, G. 2004. Group socialcapital and group effectiveness: The role of informalsocializing ties. Academy of Management Journal,47: 860–875.

Perry-Smith, J. E., & Shalley, C. E. 2003. The social side ofcreativity: A static and dynamic social network per-spective. Academy of Management Review, 28:89–106.

Price Alford, H., & Stegemeyer, A. 2009. Who’s who infashion (5th ed.). New York, NY: Fairchild Publica-tions.

Quiñones, M. A., Ford, J. K., & Teachout, M. S. 1995. Therelationship between work experience and job per-formance: A conceptual and meta-analytic review.Personnel Psychology, 48: 887–910.

Reagans, R., & McEvily, B. 2003. Network structure andknowledge transfer: The effects of cohesion andrange. Administrative Science Quarterly, 48: 240–267.

Schneider, B. 1987. The People Make the Place. Person-nel Psychology, 40: 437–453.

Schwartz, S. H. 1994. Are there universal aspects in thestructure and contents of human values? Journal ofSocial Issues, 50: 19–45.

Shalley, C. E., & Gilson, L. L. 2004. What leaders need toknow: A review of social and contextual factors thatcan foster or hinder creativity. Leadership Quar-terly, 15: 33–53.

Shane, S. 1994. The effect of national culture on the

choice between licensing and direct foreign invest-ment. Strategic Management Journal, 15: 627–642.

Shapiro, H. 1984. Tout Paris applauds the fashionablevision of Karl Lagerfeld. People. http://www.people.com/people/archive/article/0,,20088043,00.html(accessed January 10, 2012).

Shaver, J. M. 1998. Accounting for endogeneity whenassessing strategy performance: Does entry modechoice affect FDI survival? Management Science,44: 571–585.

Shin, S. J., Kim, T.-Y., Lee, J.-Y., & Bian, L. 2012. Cogni-tive team diversity and individual team member cre-ativity: A cross-level interaction. Academy of Man-agement Journal, 55: 197–212.

Simonton, D. K. 1994. Greatness: Who makes historyand why. New York, NY: The Guilford Press.

Simonton, D. K. 1997. Foreign influence and nationalachievement: The impact of open milieux on Japa-nese civilization. Journal of Personality and SocialPsychology, 72: 86–94.

Simonton, D. K. 1999. Origins of genius: Darwinianperspectives on creativity. New York, NY: OxfordUniversity Press.

Simonton, D. K. 2000. Creativity: Cognitive, personal,developmental, and social aspects. American Psy-chologist, 55: 151–158.

Simonton, D. K. 2011. Creativity and discovery as blindvariation: Campbell’s (1960) BVSR model after thehalf-century mark. Review of General Psychology,15: 158–174.

Sorenson, O., & Stuart, T. E. 2001. Syndication networksand the spatial distribution of venture capital invest-ments. American Journal of Sociology, 106: 1546–1588.

Sproles, G. B., & Burns, L. D. 1994. Changing appear-ances: Understanding dress in contemporary soci-ety. New York, NY: Fairchild Publications.

Staw, B. M. 1980. The consequences of turnover. Journalof Occupational Behavior, 1: 253–273.

Staw, B. M. 1991. Dressing up like an organization: Whenpsychological theories can explain organizationalaction. Journal of Management, 17: 805–819.

Steensma, H. K., Marino, L., Weaver, K. M., & Dickson,P. H. 2000. The influence of national culture on theformation of technology alliances by entrepreneurialfirms. Academy of Management Journal, 43: 951–973.

Tadmor, C. T., Galinsky, A. D., & Maddux, W. W. 2012.Getting the most out of living abroad: Biculturalismand integrative complexity as key drivers of creativeand professional success. Journal of Personalityand Social Psychology, 103: 520–542.


Takeuchi, R., Tesluk, P. E., Yun, S., & Lepak, D. P. 2005.An integrative view of international experience.Academy of Management Journal, 48: 85–100.

Tesluk, P. E., & Jacobs, R. R. 1998. Toward an integratedmodel of work experience. Personnel Psychology,51: 321–355.

Uzzi, B. 1996. The sources and consequences of embed-dedness for the economic performance of organiza-tions: The network effect. American SociologicalReview, 61: 674–698.

Uzzi, B., & Spiro, J. 2005. Collaboration and creativity:The small world problem. American Journal of So-ciology, 111: 447–504.

Varsakelis, N. C. 2001. The impact of patent protection,economy openness and national culture on R&D in-vestment: A cross-country empirical investigation.Research Policy, 30: 1059–1068.

Vergani, G. 2010. The fashion dictionary (2nd ed.). NewYork, NY: Baldini Castoldi Dalai Editore.

Weick, K. E. 1979. The social psychology of organizing(2nd ed.). New York, NY: Random House.

Weick, K. E., & Roberts, K. H. 1993. Collective mind inorganizations: Heedful interrelating on flight decks.Administrative Science Quarterly, 38: 357–381.

Winship, C., & Morgan, S. L. 1999. The estimation ofcausal effects from observational data. Annual Re-view of Sociology, 25: 659–706.

Wooldridge, J. 2002. Econometric analysis of cross sec-tion and panel data. Cambridge, MA: The MITPress.

Zhou, J., & Shalley, C. E. 2003. Research on employeecreativity: A critical review and directions for futureresearch. In J. Martocchio & G. R. Ferris (Eds.), Re-

search in personnel and human resource manage-ment, vol. 22: 165–217. Greenwich, CT: JAI PressInc.

Zimmermann, J., & Neyer, F. J. 2013. Do we become adifferent person when hitting the road? Personalitydevelopment of sojourners. Journal of Personalityand Social Psychology, 105: 515–530.

Frédéric C. Godart ([email protected]) is an assis-tant professor of organisational behaviour at INSEAD inFontainebleau, France. He holds a PhD in sociology fromColumbia University. His research focuses on social net-works, creativity, and the luxury and fashion industries.

William W. Maddux ([email protected]) isan associate professor of organisational behaviour atINSEAD in Fontainebleau, France. His research focuseson culture, creativity, negotiations, decision making, andleadership.

Andrew V. Shipilov ([email protected]) is as-sociate professor of strategy and Akzo Nobel Fellow ofStrategic Management at INSEAD. He received his PhDfrom the Joseph L. Rotman School of Management, Uni-versity of Toronto. His research examines how socialnetworks, strategic alliances, and partnerships evolveand affect competitive advantage.

Adam D. Galinsky ([email protected]) is theVikram S. Pandit Professor of Business at the ColumbiaBusiness School at Columbia University. His researchand teaching focus on leadership, hierarchy, diversity,negotiations, and ethics.


fashion with a foreign flair: professional experiences

Documents