The direct and indirect impact of culture on innovation

Kalanit Efrat n

Ruppin Academic Center, Ruppin 31905, Israel

a r t i c l e i n f o

Available online 2 October 2013

Keywords:InnovationCultural dimensionsCountry-levelLongitudinal

a b s t r a c t

Innovations centrality and importance in international operations has been a subject of enduring debatein the age of globalization. Cross-border knowledge spillovers and the race between nations for increasedinnovativeness only underscore the importance of innovation. One aspect of this discussion concerns theeffect of national culture on the ability to be innovative. Following Shane, 1992. Journal of BusinessVenturing. 7, 2946; Shane, 1993. Journal of Business Venturing 8, 5973 research, the present studyaims to examine the impact, both direct and indirect, of culture on the motivation to innovate at thenational level over a period of time. The ndings support the claim that although some change hadoccurred in this impact, most cultural aspects still demonstrate strong and lasting impact on thetendency to innovate at the national level.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

In times of economic crisis nations constantly seek new waysand ideas to facilitate their economic rejuvenation. In recent yearsthis search has been shaped in large part by the concepts of theknowledge economy and the information age, focusing oninnovation as the core aspect of new economic models. Both theseconcepts reect the shift in recent decades in views of whatconstitutes an economic asset. While traditional economies haveemphasized assets such as labor and machinery, modern economiesview knowledge as their major strength (Jaffe and Trajtenberg,2005). Knowledge, however, is an overly broad, holistic category; itsmanifestation in the form of innovation is regarded as moremeasurable and hence as more amenable to study. For this reason,innovation has been the subject of several studies linking it toeconomic growth (Jaffe and Trajtenberg, 2005; Fagerberg andSrholec, 2008; Freeman, 2002; Thoenig and Verdier, 2003).

No discussion of the knowledge economy or of innovation can becomplete without a consideration of their links with globalization.While innovation has a clear and direct impact on economic growth,its relationship with globalization is to some extent circular. Forcesof globalization largely rely on innovative products and services;one need only look at the tremendous impact of the internet onglobal trade to realize how innovation is embedded in and insepar-able from globalization (Freund and Weinhold, 2004). At the sametime, globalization has led to technological changes which in recentdecades have exerted a major inuence on the worlds economy,both on the macroeconomic and on the market level (Thoenig and

Verdier, 2003). For instance, globalization has given rise to multi-national enterprises (MNEs), large corporations that conduct differ-ent parts of their activities across national borders. Such rms havepromoted homogeneity in customers needs by introducing stan-dardized products and services, thereby impacting the adoption rateof new technology-based products (Nachum et al., 2008) in differentnations (Freeman, 1995). They have also tapped into variouscountries either to exploit their high knowledge resources or tobenet from their proximity to worlds knowledge centers (Nachumet al., 2008). But while globalization possesses many advantages forMNEs, some aspects of it may also be viewed as threats. One majoraspect that inhibits innovation in MNEs subsidiaries in foreigncountries is national culture. Hofstede (1994) claims that the largerthe cultural gap between the home and the host countries, the moreMNEs managers need to reconsider the way they operate theircross-border operations. In their meta-analysis, Tihanyi et al. (2005)link cultural aspects to innovation by concluding that culturaldifferences between countries have a negative impact on interna-tional diversication in high-technology industries. While MNEsoperations are clearly impacted by the national culture of the hostcountries (Dwyer et al., 2005; Steensma et al., 2000), there isevidence for a reciprocal relationship, manifest in the way MNEsprocesses and procedures inuence their subsidiaries operationsand diffuse into local industries. Though this impact occurs at themicro-level, it is bound to resonate at the national level in the longrun (Brock et al., 2000; Hofstede, 1994).

In light of this relationship, and in light of the increasing role ofMNEs in creating and maintaining innovation at the global level,we must ask Does national culture still matter? The mainpurpose of the present study is thus to investigate the impact ofdifferent cultural aspects on countries motivation both to inno-vate and to invest in innovation.

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/technovation

Technovation

0166-4972/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.technovation.2013.08.003

n Tel.: 972 9 8983835; fax: 972 9 8987604.E-mail address: kalanite@ruppin.ac.il

Technovation 34 (2014) 1220

2. Literature review

Schumpeter (1942) has addressed three stages in the process ofnew technologies entering the market: invention, innovation, andadoption. Inventions represent the core idea or concept facilitatingthe new technology. Inventions may but need not be based onpatents. Inventions do not always progress to the next stage: somewill remain a mere idea. Innovation, by contrast, is the stage inwhich invention is commercialized. Firms may base their innova-tions on the inventions of others. In the third stage, innovation isadopted by rms or individuals. The rst two stages occur on therm/industry level and are based mostly on R&D processes,whereas the last stage occurs on the market/national level andconcerns the new technologys diffusion rate. Nevertheless, forinventions and innovations to occur, a supporting infrastructureon the national level is strongly needed. This type of infrastructurehas been termed the national innovation system (NIS) (Balzat andHanusch, 2004).

2.1. Innovation and the national innovation system (NIS)

Though the term national innovation system (NIS) was coinedlong ago, with origins in the nineteenth century (Freeman, 1995),we still lack a clear denition of it (Lundvall, 2007). Earlier researchdescribed the NIS as a network of public and private institutionswithin an economy that fund and perform R&D, translate theresults of R&D into commercial innovations, and effect the diffusionof new technologies (Mowery and Oxley, 1995 p. 80). Etzkowitzand Leydesdorff (2000) describe NIS as a triple helix of state-academia-industry. Several types of organizations participate tocreate the network-like system: in addition to the national rmswhich represent the private sector, the network consists of repre-sentatives of the public sector (agencies and ofces) as well asuniversities. Collaboration among these organizations is crucial forthe system to achieve its goals (Freeman, 1995; Lundvall, 2007;Mowery and Oxley, 1995).

As noted, innovation has been found to have a strong andpositive inuence on economic growth (Freeman, 1995, 2002;Thoenig and Verdier, 2003; Balzat and Hanusch, 2004; Chell,2001), As economists ideas shifted from dependence on tangibleresources to accumulation of intangible resources such as knowl-edge as a core condition for creating value (Freeman and Soete,1997). The view that economic growth depends on innovation andon the presence of an NIS received further support from Fagerbergand Srholecs (2008) nding that the presence of an NIS signi-cantly affects the ability to innovate on the national level and thatcountries with a stable NIS enjoy signicantly higher economicgrowth than others.

While all studies advocate the establishment of NISs due to theirpositive impact on the economy, it is nevertheless clear that countriesbenet differently from investing in the creation and maintenance ofsuch systems. One conclusion that is becoming clearer as researchadvances is that innovation differs in scale and in scope in differentcountries (Fagerberg and Srholec, 2008). Since the concept of an NISincorporates that of a nation, this conclusion has drawn attention toother conditions on the national level which may divert the impact ofinvestment on innovation. Since the extent to which a nations rmstend to innovate has been found to crucially affect the success of anations innovation efforts, we should turn our focus to rmsmotivation to innovate. Such motivation is partly embedded in arms organizational culture (Hurley and Hult, 1998; Ahmed, 1998;Martins and Terblanche, 2003). Since national and organizationalcultures seem to be linked (Hofstede, 2001), certain aspects ofnational culture may well inuence the tendency to innovate at therm level. Whereas a nations NIS represents the formal, overt factorsinvolved in motivating innovation on the national level and is largely

under the nations control, informal cultural aspects may intervene toinuence this motivation. Though these latter aspects are not underthe nations control, it is very important to learn about their potentialimpact.

2.2. Innovation in a cultural perspective

Other factors besides NISs may inuence innovation. Two viewsconcerning such factors are worth noting here. The rst, men-tioned briey earlier, argues for the diminishing relevance ofnational characteristics in the age of globalization (Freeman,1995). On this view, globalization has led to the rise of MNEsand to knowledge spillover, both of which have contributed to theshift from a nations-based to a corporations-based economy (Hu,2004; Audretsch and Lehmann, 2005)hence the non-relevance ofnational innovation.

The second view suggests that several national aspects mayinuence the motivation to innovate on the national level(Lundvall, 2007) and on the rm level (Elenkov and Manev,2005; Mueller et al., 2013). Fagerberg and Srholec (2008) havefurnished strong support for this position with their nding thatseveral factors such as the quality of governance, the politicalsystem, and openness, all interlinked with the concept of NIS,interact with the ability to innovate. More specically, their studyrevealed that certain indicators such as trust, corruption, civicrights, form of governance, and education inuence innovation atthe national level. All these indicators interact with the culturaldimensions identied by Hofstede (2001). In their discussion ofthe impact of national culture on MNEs decisions about thelocation of their R&D units, Jones and Davis (2000) argued thatwhile globalization impacts such rms in various ways, nationalculture still exerts signicant inuence over the performance ofR&D units; MNEs are therefore advised to identify the relevantnational aspects and locate their R&D units in innovation-favorable countries. The upshot of this argument is that regardlessof how much money is spent on creating innovation, nationalculture and its ability either to bolster or to undermine innovationmust be taken into consideration. Similarly, Hayton et al. (2002)suggested a complex model incorporating national as well as rm-level factors in explaining entrepreneurship.

Shane (1992) has presented strong evidence for the impact ofHofstedes (1980) cultural dimensions on nations tendency toinnovate. His positive ndings led him to conclude that nationsthat are inventive would remain so. Two decades after hispioneering research, I set out to learn whether or not globalizationhas somewhat erased the impact of cultural dimensions byneutralizing the cultural distinctions among nations.

Shane (1992) based his research on Hofstedes (1980) culturaldimensions, namely power distance, individualism, masculinity,and uncertainty avoidance, overall, conceptualizing for the abilityof each dimension for predicting the level of a nationsinnovativeness.

Power distance (PDI) measures the distribution of powerwithin a society in terms of the degree to which its membersexpect and accept inequality. High PDI scores translate intohierarchical national or organizational structure. In high-PDIcountries, organizational structure tends to be more centralizedand rigid: decision-making information is the preserve of those inauthority. The key concepts in such organizations are supervisionand rules (Hofstede, 2001). Shane (1992) describes several char-acteristics of low-PDI countries that interact with the tendency toinnovate. In low-PDI countries, he claims, new organizations tendto be smaller and more organic, with high information-processingcapabilities and informal communication between superiors andsubordinates. Such organizations are further characterized aspower-decentralized, with control systems based mainly on trust.

K. Efrat / Technovation 34 (2014) 1220 13

Indeed, Shane (1992) found that high-PDI countries displayedlower per-capita returns on inventions. Further support for thesendings has been found in connection to employees rewardsystems. Hofstede (2001) links low PDI both to the use of manage-ment by objectives and to fair wages. Chandler et al. (2000) foundthat employees who perceived the organization reward system asrewarding innovation tended to be more strongly committed toinnovation. According to Ahmed (1998), certain cultural norms,such as trust and openness, awards and rewards, and autonomyand exibility, facilitate an innovative climate in organizations.Hofstede (2001) has shown all these norms to be closely asso-ciated with low PDI. In their discussion of the determinants oforganizational culture which support creativity and innovation,Martins and Terblanche (2003) identify ve such categories, threeof which structure, support mechanisms, and communication incorporate elements related to low PDI (Hofstede, 2001). Shaneet al. (1995) describe how PDI can impact ones perceptions andhence ones innovativeness.

H1a. Low-PDI countries will invest more in facilitating innovation.

H1b. Low-PDI countries will evince higher rates of innovation.

Individualism (IDV) refers to the degree to which, for eachindividual in a given group, his or her interests prevail over thegroups; in other words, IDV measures the balance between anIcentered and a Wecentered consciousness in a givensociety. In high-IDV countries each individual is expected to takecare of himself or herself and his or her immediate family. Suchsocieties emphasize individual initiative and achievements. High-IDV countries have a strong entrepreneurial orientation whichenables and motivates invention and innovation, both within andwithout formal organizational borders or existing networks(Hofstede, 2001). IDV and PDI share many similar characteristicsin terms of facilitating innovation. Martins and Terblanche (2003)found that elements such as structural exibility and employeesfreedom, which translate into autonomy, empowerment, anddecision making, are all determinants of innovation. Similarndings are presented in Ahmed (1998), where cultural elementsresponsible for a climate of innovation include freedom, trust, andawards and rewards. Hofstede (2001) demonstrates a strongcorrelation between IDV and PDI, with low-PDI countries usuallyreceiving high IDV scores, and vice versa.

In conclusion, Shanes (1992) ndings are consistent with thehypothesis that high-IDV countries will have higher nationalinnovation rates.

H2a. High-IDV countries will invest more in facilitating innovation.

H2b. High-IDV countries will evince higher rates of innovation.

Uncertainty avoidance (UAI) refers to the levels of anxiety, theneed for protection, and the concomitant desire for rules and lawsfelt in a society. Whereas low-UAI societies tend to be more open tochange and new ideas, members of high-UAI societies tend toperceive novelty as dangerous and hence to resist it. While technol-ogy is perceived as a means for creating order and therefore hasappeal for members of high-UAI societies, innovation is regarded asthe carrier of change and is viewed with suspicion. In high-UAIcountries, organizational culture favors a highly formalized concep-tion of management and a hierarchical organizational structure, bothof which contribute to a sense of order and control (Hofstede, 2001).But whereas ndings on the impact of PDI and IDV on innovationhave been conclusive, studies on UAI have yielded conicting results(Shane, 1993; Kedia et al., 1992; Nakata and Sivakumar, 1996). Thiscan be put down to an ambivalent perception of technology. On theone hand, technology is seen as one of the three pillars supportinghigh-UAI societies in their attempt to maintain order and a sense of

predictability: by increasing standardization and thus outcomecertainty, technology helps rms carry out their missions andperform their tasks (Hofstede, 2001). On the other hand, technologyrests on innovation, which, as noted earlier, depends on suchconditions as exibility, informal communication, employeesempowerment, trust, and openness (Ahmed, 1998; Martins andTerblanche, 2003), which are inconsistent with the need for stabilityand predictability at the core of UAI. On the basis of these character-istics of UAI and the interaction between UAI and PDI, I hypothesizethat low UAI levels will contribute to innovation. UAI correlatesstrongly with PDI, suggesting the presence of similar elements inboth (Hofstede, 2001). Previous ndings on the climate for innova-tion (Martins and Terblanche, 2003; Shane, 1995) support a possiblelinkage between low UAI and innovation.

H3a. Low-UAI countries will invest more in facilitating innovation.

H3b. Low-UAI countries will evince higher rates of innovation.

While PDI, IDV, and UAI have received some attention asdeterminants of varying levels of innovation in different countries(Shane, 1992, 1993; Kedia et al., 1992), a fourth cultural aspect,masculinity, has largely been neglected.

Masculinity (MAS) refers to the dominant gender-role patterns.In more masculine societies the emphasis is on ego, money,performance, and achievements, while in more feminine societiesthere is a greater balance between mens and womens roles(Hofstede, 2001). According to previous studies, high-MAS societiesdisplay a greater tendency towards invention and innovation basedon achievement triggers (Shane, 1993). One of the core aspects ofMAS, success (as manifested in nancial rewards or other means ofacknowledgement), is highly appreciated in high-MAS societies andis also a strong motivator of innovation (Ahmed, 1998; Chandleret al., 2000; Morris et al., 1994). There are thus solid grounds forproposing a link between MAS and innovation Fig. 1.

H4a. High-MAS countries will invest more in facilitating innovation.

H4b. High-MAS countries will evince higher rates of innovation.

3. Method

The current paper examines whether in the age of globalizationcultural aspects can still predict a countrys level of innovation.The paper examines data for 1998, 2003, and 2007 in 35 countries:

PowerDistance

(PDI)

R&D Expense

Individualism(IDV)

Masculinity(MAS)

UncertaintyAvoidance

(UAI)

H1b

H1a

InnovationIndicators:

Patents

Journal Articles

High-technologyExports

H2aH2b

H3a

H4a

H3b

H4b

Fig. 1. Research model.

K. Efrat / Technovation 34 (2014) 122014

33 OECD nations (Iceland was excluded for lack of cultural data),Singapore, and Hong Kong. I aimed to control for the level ofeconomic development, in line with previous ndings whichsuggest a strong linkage between countries level of economicdevelopment and rate of cultural change (Freeman, 2002). There-fore, only countries regarded as economically developed wereincluded in the study. The three aforementioned years werelikewise chosen to control for the potential inuence of economicfactors. Economic changes, no less than the level of economicdevelopment, are known to have a potential impact on thetendency to innovate (Inglehart and Baker, 2000). Since one ofmy goals was to learn whether globalization has inuencedinnovation by eliminating the impact of national culture, culturalaspects were examined over a decade. Two of the years for whichdata were studied were chosen in order to avoid any economicinuences: 2003 because that year witnessed the end of theeconomic changes that followed the high-tech downfall of 19992001; and 2007 because that was the nal year clear of any of theeconomic pressures associated with the present economic crisis.Since the study investigated the longitudinal impact of culturalaspects, I marked 1998 as the rst year on the continuum, givingus a decade of data. One issue is worth clarifying in this regard.Culture tends to change at a very slow rate (Hofstede, 2001). Also,by controlling for the potential impact of economic changes,cultural foundations in terms of norms and values tend to be longlasting (Inglehart and Baker, 2000). In choosing to examine thesespecic three years on the continuum, I was aiming to identifysignicant cultural impacts, such that would be manifested in allthree years, not to examine differences in cultural impact betweenthe three years.

The current study is based on a variety of secondary sources.

3.1. Measures of innovation inputs

As noted earlier, NIS has a major impact on a countrys ability toinnovate (Fagerberg and Srholec, 2008; Balzat and Hanusch, 2004).NIS represents the national systems dedicated to enhancing rmsand individuals ability to turn thoughts and ideas into solid productsand services. These systems are funded by national organizations.The amount of national funds invested in creating innovation musttherefore be considered a predictive variable. Data on nationalinvestments in innovation were retrieved from the World Bankdatabase (2010) as presented in its website. Two economic indicatorswere used for each country in the present study. The rst, used as acontrol variable, is gross national income (GNI) per capita (in US

dollars) divided by the midyear population. The second is theamount of money invested in innovation as a percentage of thecountrys GDP (see Appendix for complete data on each country).

The second group of factors, representing national culture,consists of the cultural dimensions discussed earlier: PDI, IDV,UAI, and MAS. Updated national scores for these dimensions weretaken from The Hofstede Center website (2013) (see Table 1 fordescriptive statistics and correlations).

3.2. Measures of innovation outputs

Three measures were used to capture innovation, each refer-ring to a different aspect of innovation: patents, scientic andtechnical journal articles, and high-technology exports. All threemeasures were also obtained from the World Bank database.The selected variables represent the reciprocal interactionbetween industry and academia that lies at the core of innovation(Etzkowitz and Leydesdorff, 2000). Scientic articles represent oneend of this scale, being pure products of academia, whereas high-technology exports represent the opposite end measured in termsof the industrys output. Patents are located in-between, gener-ated partly by academia and partly by industry. Incorporating allthree aspects of innovation fully captures in my view the culturalimpact on innovation.

Previous studies have used patents as a measure of innovation(Jaffe and Trajtenberg, 2005; Hu, 2004; Shane, 1993; Acs et al.,2002; Acs and Audretsch, 1988; Seltzer and Bentley, 1999). In linewith previous studies, the patent variable is measured as the ratiobetween the total number of patents granted in a country and thecountrys population.

Previous studies have also used scientic and technical journalarticles as a measure of innovation (Chen et al., 2007; Fagerberget al., 2005). Here, again, the current study measures the ratio ofthe total number of articles published by a countrys researchers tothe countrys population.

The third measure is given by a countrys high-technology exportsas a percentage of its total exports. Again, previous studies have usedthis indicator as a measure of innovation (Kedia et al., 1992; Fagerberget al., 2005; AtuaheneGima, 1996; Song and Parry, 1997).

4. Results

I tested the hypotheses using structural equation modeling(SEM) analysis with AMOS software. The exibility of SEM analysis

Table 1Descriptive statistics and correlations of the variables.

Mean (s.d.) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 PDI 48.54 (20.72) 12 IDV 58.03 (21.65) .56nn 13 UAI 65.54 (24.83) .32 .32 14 MAS 51.26 (24.15) .23 .10 .12 15 R&D Invest.(98) 1.49 (.83) .49nn .29 .12 .06 16 R&D Invest.(03) 1.68 (1.02) .55nn .27 .25 .22 17 R&D Invest.(07) 1.82 (1.09) .50nn .07 .24 .24 18 High-tech exp.(98) 17.22 (12.39) .17 .09 .58nn .00 .45n 19 High-tech exp.(03) 17.41 (11.05) .20 .05 .50nn .02 .38n 110 High-tech exp.(07) 15.32 (9.54) .10 .00 .53nn .02 .34 111 Patents (98) 633.54 (722.6) .04 .12 .30 .15 .41n .40n 112 Patents (03) 671.57 (766.3) .08 .09 .23 .19 .39n .45nn .94nn 113 Patents (07) 695.78 (919.5) .08 .20 .22 .10 .53nn .50nn .91nn .97nn 114 Articles (98) 465.96 (329.6) .57nn .59nn .53nn .28 .71nn .31 .23 .20 .14 115 Articles (03) 512.28 (303.6) .63nn .49nn .56nn .29 .77nn .34n .24 .24 .20 .94nn 116 Articles (07) 567.1 (307.6) .61nn .46nn .58nn .35n .71nn .33 .23 .23 .19 .92nn .98nn

(PDI-Power distance; IDV-Individualism; UAI-Uncertainty avoidance; MAS-Masculinity).* po .05, **po .01

K. Efrat / Technovation 34 (2014) 1220 15

makes it more suitable to explaining the relationships amongmultiple variables. The three models tested, one for each year,were signicant. All t indices were very good (see Table 2).

All the hypotheses were examined for each of the three years(1998, 2003, 2007). Results for all three years are presented below.

I rst examined the potential mediating impact of the invest-ment in innovation on the three chosen aspects of innovation. H1asuggested that investment in innovation mediates the impact ofPDI on innovation. PDI was found to have a negative impact on theinvestment, while investment in innovation was positively corre-lated with all three aspects of innovation, therefore conrmingH1a.

None of the other three cultural dimensions were found to haveany impact on investment in innovation. H2a, H3a, and H4a werethus not supported.

I then turned to examine the direct impact of each of the fourcultural aspects on the three innovation indicators.

The model for the year 1998 explained 48% of the variance inhigh-technology exports, with UAI ( .62, po .01) acting asstrong predictor. As for patents, the model explained 39% of thevariance, with IDV (0.42, po .05), MAS (0.25, po .10), andUAI ( .39, po .01) signicantly impacting innovation. Themodel also explained 84% of the variance in the third innovationindicator (scientic articles), with IDV (0.43, po .01), MAS( .28, po .01) and UAI ( .19, po .05) strongly impactinginnovation (For detailed results see Table 3).

Similar ndings were obtained for 2003, where high-technology exports (R20.36) were signicantly inuenced byUAI ( .50, po .01); patents (R20.39) with IDV (0.37,po .05), MAS (0.35, po .05), and UAI ( .26, po .10); andscientic articles (R20.81) with IDV (0.24, po .05), MAS( .19, po .05), and UAI ( .23, po .01).

The ndings for 2007 largely followed this pattern. The modelwas signicant for high-technology exports (R20.39) withrespect to UAI ( .55, po .01); for patents (R20.42) withrespect to IDV (0.46, po .01), MAS (0.31, po .05), andUAI ( .23, po .10); and for scientic articles (R20.74) withrespect to IDV (0.28, po .05), MAS ( .26, po .01) and UAI( .28, po .01) (see Table 3 for detailed results).

In terms of the hypotheses, the results were as follows. H1bwas not substantiated, as PDI did not explain any of the innovationfactors examined. H4b was substantiated, as UAI was found tohave a negative impact on innovation. As for H2b, the impact ofIDV on innovation was ambivalent: its impact on scientic articleswas as expected, but its impact on patents was contrary to thehypothesis. Precisely the opposite results were found for MAS(H3b): the impact of MAS on patents was as expected, but itsimpact on scientic articles was contrary to the hypothesis.

Post-hoc analysis: With the above results in place, I conducted apost-hoc analysis to check for potential interactions between thevarious cultural dimensions. The three modied models (1998,2003, and 2007) showed very good t indices (see Table 4 forresults).

As noted earlier, UAI was found to have a consistently negativeimpact on the various aspects of innovation, while IDV and MASwere more ambivalent. In light of these results, I created two newconstructs based on the interaction between UAI and IDV (Con-struct 1), and between UAI and MAS (Construct 2). I then ran thethree models, one for each year, using the two new interactionconstructs. The results were interesting. As mentioned earlier,patents were impacted negatively by UAI and IDV and positivelyby MAS. When using the interaction constructs, however, I foundthat the interaction between UAI and MAS eliminated the negativeinuence of UAI, resulting in a positive impact of Construct 2 onpatents. A similar result was obtained with regard to scienticarticles. As noted earlier, scientic articles were impacted nega-tively by UAI and MAS and positively by IDV. The interactionbetween UAI and IDV eliminated, however, the negative impact ofUAI, resulting in a positive impact of Construct 1 on scienticarticles (see Table 5 for detailed results).

5. Discussion

The present study had two main parts. First, almost two decadesafter Shane (1992, 1995), I set out to examine whether the impact of

Table 2Results of model t indices for the path analysis models.

CMIN (Chi-square) DF P CFI NFI RMSEA

Model 1Year 1998 2.996 6 .392 1.00 .980 .000Model 2Year 2003 2.982 6 .811 1.00 .979 .000Model 3Year 2007 4.356 6 .629 1.00 .968 .063

(CFIComparative t index; NFINorm t index; RMSEARoot square error ofapproximation).

Table 3Constructs weights.

Model 1 1998 Model 2 2003 Model 3 2007

Export(R2 .48)

Patents(R2 .39)

Articles(R2 .84)

Export(R2 .37)

Patents(R2 .41)

Articles(R2 .82)

Export(R2 .39)

Patents(R2 .42)

Articles(R2 .74)

R&D expense .32n .45nn .76nnn .28 .45nn .64nnn .36nn .51nnn .48nnn

UAI .62nnn .39nnn .19nn .50nnn .26n .23nn .54nnn .23n .29nnnIDV .42nn .43nnn .37nn .24nn .46nnn .28nnMAS .25n .27nnn .34nn .19nn .30nn .26nn

n po .1.nn po .05.nnn po .01.

Table 4Results of model t indices for the path analysis modelsmodied models.

CMIN (Chi-square)

DF P CFI NFI RMSEA

Modied Model 1Year1998

1.740 6 .942 1.00 .984 .000

Modied Model 2Year2003

4.481 6 .612 1.00 .960 .000

Modied Model 3Year2007

6.328 6 .387 .996 .940 .040

(CFIComparative t index; NFINormed t index; RMSEARoot square error ofapproximation).

K. Efrat / Technovation 34 (2014) 122016

certain aspects of national culture on the motivation to innovatewere mediated by the investment in innovation, bearing in mindthat these two decades have witnessed rapid globalization. Thesecond part, which follows from the rst, was to explore differencesin cultural inuence on several elements of innovation. This wasfollowed by a post-hoc analysis establishing potential interactionsbetween the various aspects of national culture.

Before I discuss the results, one issue regarding cultural aspectsmust be addressed. The use of Hofstedes cultural dimensions maybe seen as controversial (Venaik and Brewer, 2010). The mainpoints presented by its critics concern (1) the use of quantitativemeasures, (2) the fact that Hofstedes data were collected from theworkers of a particular organization (IBM), raising questions aboutthe studys generalizability, and (3) the time elapsed since the rstround of data collection. In their comprehensive work, Magnussonet al. (2008) compare Hofstedes cultural dimensions with othercultural frameworks. Observing that culture may be even morestable than Hofstede himself had argued, they conclude thatHofstedes framework has a strong convergent validity comparedto more contemporary cultural frameworks such as Schwartz(1994) and House et al. (2004). Hence, Hofstedes cultural dimen-sions are often used to explain either direct or mediated relation-ships (Mueller et al., 2013; Tihanyi et al., 2005).

Let us turn to the results, starting with the rst part of thestudy. Of the four cultural aspects, only PDI was found to bemediated by investment in innovation, indicating that investing ininnovation in high-PDI countries will neutralize the potentiallynegative impact of PDI on innovation. This in itself is an importantnding. While this nding helps establish investment in innova-tion as the foundation of enhanced innovativeness, it does not, initself, entail that culture is of no importance, but rather conrmsthat PDI coincides and interacts with other forces which inuencethe tendency to innovate at both the rm and the national level(Grinstein, 2008).

Turning to the second part of the study, though culturaldimensions proved to impact innovativeness, the present studyrevealed some shifts in this impact over time. Overall, three of thefour cultural aspects were shown to impact the three dimensionsof innovation. My hypothesis expected low-PDI characteristics tohave a positive effect on innovation, following some strongevidence in support of this expectation (Shane et al., 1995, 1993;Grinstein, 2008. But despite consensus in the literature, thendings revealed no such impact, suggesting that the relevanceof PDI has diminished over time. One possible explanation has todo with the shift from a nation-based to a corporations-basedeconomy and with organizational techniques to facilitate innova-tion. In earlier research, many high-PDI characteristics were linkedto barriers to innovation at the national level (Ahmed, 1998;Martins and Terblanche, 2003; Shane, 1993). One major changein recent years has been the growing prominence of MNEs in theglobal arena (Fiegenbaum and Lavie, 2000). Whereas in the past

innovation was measured at the national level, nowadays MNEsprovide the necessary basis for its creation. PDI characteristicsmeasured at the national level are therefore no longer suitable forunderstanding barriers to innovation. Furthermore, since mostcountries examined in the present research most of them OECDmembers share similar regime principles, the differences amongthem with respect to PDI are relatively insignicant.

Other hypotheses referred to the impact of IDV and MAS oninnovation. In brief, I conjectured that high IDV and high MASwould favor innovativeness. The ndings were quite surprising,however: while some aspects of innovation were found to behigher in cultures scoring high on IDV or MAS, others were higherin low-IDV and low-MAS cultures.

With respect to IDV, these ndings may resemble previousndings (Morris et al., 1994, 1993; Eisenberg, 1999) which suggesta curvilinear relationship between entrepreneurship and indivi-dualism/collectivism, whereby a high level of individualism hin-ders innovation. The mean IDV score for the countries in thepresent study was 58 higher than the mean of 43 reported byHofstede (2001) in his original survey indicating a relatively highlevel of individualism in the countries in our survey. The presentndings support the conjecture presented earlier that a certainlevel of collectivism will encourage innovation.

Another issue that can further our understanding of collecti-visms impact on innovation is networks, which can be under-stood as a byproduct of globalization. The need to survive and tooutperform competitors, on the one hand, and the diminishingrelevance of geographical borders, on the other, have led to thegrowth of networks. Networks have proven to facilitate innovationby sharing resources such as unique knowledge (Chen, 2004;Ritter and Gemunden, 2004). Networks by their nature emphasizethe centrality of sharing and collaboration, two foundations ofcollectivism. Since networks have been gaining a foothold in theglobal arena shifting the spotlight to the advantages of colla-boration, especially in terms of producing innovation it is only amatter of time before a similar shift on the spectrum of individu-alism/collectivism occurs.

The counter-hypothetical results of the MAS dimension canalso be explained as related to collaboration and networking.As noted earlier, collaboration and networking have both becomecentral aspects of management theories, especially in connectionwith innovation, and are both motivated by forces of globalization(Dakhli and De Clercq, 2004; Marino et al., 2002; Vecchi andBrennan, 2009; Kali and Reyes, 2007; Rycroft and Kash, 2004).Both features appear to be highly important in the global arena,granting a competitive advantage to rms that have them andenhancing their innovative capabilities (Coviello and Munro,1997). Both can also be linked with femininity as described byHofstede (2001). Femininity values relationships as a means toachieving cooperation and well-being; relationships, on this view,help us meet present goals by creating an environment more

Table 5Constructs weightsmodied models.

Modied model 1 1998 Modied model 2 2003 Modied model 3 2007

Export(R2 .41)

Patents(R2 .39)

Articles(R2 .74)

Export(R2 .31)

Patents(R2 .38)

Articles(R2 .79)

Export(R2 .33)

Patents(R2 .38)

Articles(R2 .67)

UAIIDV interactionconstruct

.49nnn .62nnn .36nnn .40nn .52nnn .21nn .43nn .53nnn .26nn

UAIMAS interactionconstruct

.02 .33n .57nnn .02 .41nnn .47nnn .04 .34nn .60nnn

n po .1nn po .05.nnn po .01.

K. Efrat / Technovation 34 (2014) 1220 17

conducive to achieving them. Networks can be viewed as aprogressive form of relationships, aimed at creating value for theirparticipants. Previous ndings support this explanation by con-cluding that femininity is linked to the tendency to form alliances(Marino et al., 2002) and to the ability to produce decision-supportsystems (Vecchi and Brennan, 2009). Since the ability to formpartnerships is associated with femininity, and since these colla-borations have proven to contribute to innovation, the MASdimension had a signicant impact on innovation. Its impactvaried, however, with respect to different aspects of innovation.Femininity was strongly related to scientic articles but inverselyrelated to patents. This can be explained by the different motiva-tions associated with each of the two aspects of innovation.Patents often involve higher levels of innovation, namely disrup-tive innovation or at times even innovation breakthroughs; theytherefore possess little resemblance to previous innovation. Suchinnovation is strongly associated with the characteristics of mas-culinity. By contrast, scientic papers must rely on previousndings and existing knowledge frameworks. In addition, theymore often involve collaboration between several scholars(Emrouznejad et al., 2008).

As for the nal hypothesis concerning uncertainty avoidance: aspredicted, this dimension had a negative impact on some aspects ofinnovation, suggesting higher rates of innovation in low uncertainty-avoiding countries. The rationale presented earlier therefore holds,

substantiating the claim that certain constructs embodied in thecultural foundation of low-UAI countries openness to change,willingness to take risks, abilities prevailing over seniority becomeoperational in an innovative climate.

Turning nally to the post-hoc results: My aim in the post-hocanalysis was to identify any potential interaction between thevarious cultural aspects. Such interaction was indeed found. UAIalone had a negative inuence on all aspects of innovation, but apositive impact when combined with either one of the other twocultural dimensions, IDV and MAS. This suggests that while theimpact of UAI is probably consistently negative, its impact isweaker than that of each of the other two dimensions. Anyassessment of the impact of UAI should therefore consider IDVand MAS as well.

6. Conclusion

The present study aims to answer the question whether today,deep into the age of globalization and following the shift fromnation-based to rm-based innovation motivators, culture stillmatters. The ndings show that while investment in innovation isa major facilitator of innovation, cultural aspects, specically Indi-vidualism, Masculinity, and Uncertainty Avoidance, still motivateinnovation. This impact appears to have been stable and rm across

Table A1Table of original data.

Countries PDI IDV MAS UAI 1998 2003 2007

R&Dinvestments

Patents Articles High-techexport

R&Dinvestments

Patents Articles High-techexport

R&Dinvestments

Patents Articles High-techexport

Australia 36 90 61 51 1.51 .0957 .0764 11.28 1.80 .1085 .0750 13.77 2.16 .1273 .0846 11.15Austria 11 55 79 70 1.78 .0028 .0051 10.66 2.26 .0287 .0577 15.30 2.53 .0340 .0581 11.32Belgium 65 75 54 94 1.86 .0092 .0551 8.00 1.88 .0068 .0607 8.52 1.90 .0058 .0665 7.48Canada 39 80 52 48 1.76 .1123 .0733 15.47 2.03 .1175 .0743 14.04 1.90 .1216 .0843 14.33Chile 63 23 28 86 .50 .0184 .0065 3.67 .67 .0150 .0088 3.51 .67 .0228 .0104 6.59Czech Rep. 57 58 57 74 1.14 .0426 .0219 7.91 1.25 .0350 .0276 13.23 .15 .0087 .0356 14.13Denmark 18 74 16 23 2.04 .0322 .0889 17.63 2.58 .0357 .0929 19.62 2.56 .0340 .0958 16.90Estonia 40 60 30 60 .57 .0334 .0219 11.63 .77 .0444 .0261 20.45 1.11 .0046 .0374 11.52Finland 33 63 26 59 2.87 .0520 .0875 21.97 3.43 .0419 .0939 23.75 3.47 .0381 .0943 21.48France 68 71 43 86 2.14 .0279 .0521 21.93 2.17 .0271 .0488 19.47 2.04 .0268 .0481 18.86Germany 35 67 66 65 2.27 .0699 .0523 14.5 2.52 .0708 .0511 16.37 2.54 .0741 .0539 14.22GreatBritain

35 89 66 35 1.76 .0506 .0788 28.85 1.75 .0530 .0759 25.93 1.82 .0409 .0772 19.51

Greece 60 35 57 112 .53 .0028 .0240 7.97 .57 0.38 .0325 12.18 .57 .0053 .0444 8.02Hong Kong 68 25 57 29 .43 .2241 N.A. 20.90 .69 .1352 N.A. 12.65 .82 .1987 N.A. 19.33Hungary 46 80 88 82 .68 .0155 .0214 20.55 .93 .0474 .0241 25.65 .96 .0078 .0243 25.20Ireland 28 70 68 35 1.24 .0298 .0409 44.14 1.17 .0235 .0420 34.47 1.28 .0212 .0570 27.93Israel 13 54 47 81 3.13 .0846 .1008 19.56 4.34 .0881 .0975 18.10 4.76 .1115 .0922 7.54Italy 50 76 70 75 1.05 .0130 .0352 7.90 1.11 N.A. .0412 7.84 1.18 .0170 .0447 6.63Japan 54 46 95 92 3.00 .3180 .0425 25.75 3.20 .3234 .0448 24.06 3.44 .3101 .0413 18.96Korea,South

60 18 39 85 2.34 .1625 .1625 26.82 2.49 .2479 .0279 32.15 3.21 .3559 .0381 33.45

Luxembourg 40 60 50 70 1.80 .0263 .0263 12.00 1.65 .0053 .0084 11.97 1.62 .0083 .0152 8.85Mexico 81 30 69 82 .38 .0038 .0038 19.17 .40 .0120 .0036 21.31 .37 .0157 .0040 17.09Netherlands 38 80 14 53 1.89 .0191 .0191 29.99 1.76 .0176 .0780 30.89 1.72 .0149 .0867 25.71NewZealand

22 79 58 49 1.05 .1494 .1494 12.19 1.19 .1706 .0695 9.49 1.21 .1855 .0750 9.46

Norway 31 69 8 50 1.63 .1403 .1403 16.24 1.71 .1283 .0685 18.64 1.64 .1412 .0865 18.05Poland 68 60 64 93 .67 .0169 .0169 2.86 .54 .0163 .0172 3.07 .57 .0072 .0187 3.76Portugal 63 27 31 104 .65 .0015 .0015 4.03 .74 .0015 .0230 8.57 1.21 .0026 .0322 8.54Singapore 74 20 48 8 1.81 .1700 .1700 58.84 2.11 .1921 .0714 56.27 2.52 .2168 .0826 46.40Slovak Rep. 104 52 110 51 .78 .0338 .0338 3.63 .57 .0306 .0166 3.71 .46 .0063 .0179 5.35Slovenia 71 27 19 88 1.33 .0209 .0209 4.27 1.27 .0185 .0476 6.26 1.45 .0171 .0633 5.34Spain 57 51 42 86 .87 .0069 .0069 6.68 1.05 .0075 .03832 7.26 1.27 .0078 .0467 5.16Sweden 31 71 5 29 3.55 .0533 .0533 19.56 3.85 .0416 .1080 16.02 3.61 .0319 .1083 15.70Switzerland 34 68 70 58 2.52 .0364 .0364 16.33 N.A. .0300 .1108 22.15 3.00 .0269 .1217 21.67Turkey 66 37 45 85 .37 .0038 .0038 2.20 .48 .0012 .0087 2.06 .72 .0027 .0118 2.04USA 40 91 62 46 2.61 .0859 .0859 33.22 2.66 .1179 .0676 30.81 2.72 .1468 .0675 28.54

K. Efrat / Technovation 34 (2014) 122018

three points in time spanning one decade. In addition, PDI, once astrong indicator of innovation, diminished in its inuence. In sum,my ndings suggest that national culture still possessesa substantial inuence on the ability of rms, either localor foreign, to create and maintain innovation in specic locations.Furthermore, the present study establishes a synergetic relationshipbetween the different dimensions of culture: dimensions whichimpact innovation negatively when operating individually mayimpact it positively when combined with others. Firms shouldtherefore take national culture into account when determining thelocation of organizational units involved in creating any type ofinnovation. More specically, they should consider national culturewhen establishing their expectations regarding the units outputsand when planning the management of such units.

Appendix

See appendix Table A1 here.

References

Acs, Z.J., Anselin, L., Varga, A., 2002. Patents and innovation counts as measures ofregional production of new knowledge. Research Policy 31, 10691085.

Acs, Z.J., Audretsch, D.B., 1988. Innovation in large and small rms: an empiricalanalysis. American Economic Review 78, 678690.

Ahmed, P.K., 1998. Culture and climate for innovation. European Journal ofInnovation Management 1, 3043.

AtuaheneGima, K., 1996. Differential potency of factors affecting innovationperformance in manufacturing and services rms in Australia. Journal ofProduct Innovation Management 13, 3552.

Audretsch, D.B., Lehmann, E.E., 2005. Does the knowledge spillover theory ofentrepreneurship hold for regions? Research Policy 34, 11911202.

Balzat, M., Hanusch, H., 2004. Recent trends in the research on national innovationsystems. Journal of Evolutionary Economics 14, 197210.

Brock, D.M., Barry, D., Thomas, D.C., 2000. Your forward is our reverse, your right,our wrong: rethinking multinational planning processes in light of nationalculture. International Business Review 9, 687701.

Chandler, G.N., Keller, C., Lyon, D.W., 2000. Unraveling the determinants andconsequences of an innovation-supportive organizational culture. Entrepre-neurship Theory and Practice 25, 5976.

Chell, E., 2001. Entrepreneurship: Globalization, Innovation and Development.Cengage Learning.

Chen, D., Lin, W.C., Huang, M., 2007. Using essential patent index and essentialtechnological strength to evaluate industrial technological innovation competi-tiveness. Scientometrics 71, 101116.

Chen, S., 2004. Taiwanese IT rms offshore R&D in China and the connection withthe global innovation network. Research Policy 33, 337349.

Coviello, N., Munro, H., 1997. Network relationships and the internationalisationprocess of small software rms. International Business Review 6, 361386.

Dakhli, M., De Clercq, D., 2004. Human capital, social capital, and innovation: amulti-country study. Entrepreneurship & Regional Development 16, 107128.

Dwyer, S., Mesak, H., Hsu, M., 2005. An exploratory examination of the inuence ofmational culture on cross-national product diffusion. Journal of InternationalMarketing 13, 128.

Eisenberg, J., 1999. How individualismcollectivism moderates the effects ofrewards on creativity and innovation: a comparative review of practices inJapan and the US. Creativity and Innovation Management 8, 251261.

Elenkov, D.S., Manev, I.M., 2005. Top management leadership and inuence oninnovation: the role of sociocultural context. Journal of Management 31,381402.

Emrouznejad, A., Parker, B.R., Tavares, G., 2008. Evaluation of research in efciencyand productivity: a survey and analysis of the rst 30 years of scholarlyliterature in DEA. Socio-Economic Planning Science 42, 151157.

Etzkowitz, H., Leydesdorff, L., 2000. The dynamics of innovation: from NationalSystems and Mode 2 to a Triple Helix university-industry-governmentrelations. Research Policy 29, 109123.

Fagerberg J., Srholec M., and Knell M., The competitiveness of nations, The DRUIDTenth Anniversary Summer Conference, Copenhagen Denmark, 2005.

Fagerberg, J., Srholec, M., 2008. National innovation systems, capabilities andeconomic development. Research Policy 37, 14171435.

Fiegenbaum, A., Lavie, D., 2000. Strategic management of MNCs entry into foreignmarkets: experience of Israel in the 1990s. European Management Journal 18,93105.

Freeman, C., 2002. Continental, national and sub-national innovation systemscomplementarity and economic growth. Research Policy 31, 191211.

Freeman, C., 1995. The National System of Innovation in historical perspective.Cambridge Journal of Economics 19, 524.

Freeman, C., Soete, L., 1997. The Economics of Industrial Innovation. Routledge.

Freund, C.L., Weinhold, D., 2004. The effect of the Internet on international trade.Journal of International Economics 62, 171189.

Grinstein, A., 2008. The effect of market orientation and its components oninnovation consequences: a meta-analysis. Journal of the Academy of Market-ing Science 36, 166173.

Hayton, J.C., George, G., Zahra, S.A., 2002. National culture and entrepreneurship: areview of behavioral research. Entrepreneurship: Theory & Practice 26, 3352.

Hofstede, G., 1994. The business of international business is culture. InternationalBusiness Review 3, 114.

Hofstede, G., 2001. Cultures Consequences: Comparing Values. Sage Publications,Thousand Oaks, CA.

Hofstede, G., 1980. Cultures Consequences: International Differences in Work-Related Values. Sage, Beverly Hills, CA.

House, R.J., Hanges, P.J., Javidan, M., Dorfman, P.W., Gupta, V., 2004. Culture,Leadership, and Organizations: The GLOBE Study of 62 Societies. Sage, Thou-sand Oaks, CA.

Hu, A.G., 2004. Multinational corporations, patenting, and knowledge ow: thecase of Singapore. Economic Development and Cultural Change 52, 781800.

Hurley, R.F., Hult, G.T.M., 1998. Innovation, market orientation, and organizationallearning: an integration and empirical examination. The Journal of Marketing62, 4254.

Inglehart, R., Baker, W.E., 2000. Modernization, cultural change, and persistence oftraditional values. American Sociological Review 65, 1951.

Jaffe, A.B., Trajtenberg, M., 2005. Patents, Citations, and Innovations: A Window onthe Knowledge Economy. The MIT Press.

Jones, G.K., Davis, H.I., 2000. National culture and innovation: implications forlocating global R&D operations. MIR: Management International Review 40,1139.

Kali, R., Reyes, J., 2007. The architecture of globalization: a network approach tointernational economic integration. Journal of International Business Studies38, 595620.

Kedia, B.L., Keller, R.T., Jullan, S.D., 1992. Dimensions of national culture and theproductivity of R&D units. The Journal of High Technology ManagementResearch 3, 118.

Lundvall, B., 2007. National innovation systemsanalytical concept and develop-ment tool. Industry and Innovation 14, 95119.

Magnusson, P., Wilson, R.T., Zdravkovic, S., Zhou, J.X., Westjohn, S.A., 2008. Breakingthrough the cultural cluttera comparative assessment of multiple cultural andinstitutional frameworks. International Marketing Review 25, 183201.

Marino, L., Strandholm, K., Steensma, H.K., Weaver, K.M., 2002. The moderatingeffect of national culture on the relationship between entrepreneurial orienta-tion and strategic alliance portfolio extensiveness. Entrepreneurship Theoryand Practice 26, 145160.

Martins, E.C., Terblanche, F., 2003. Building organizational culture that stimulatescreativity and innovation. European Journal of Innovation Management 6,6474.

Morris, M.H., Avila, R.A., Alien, J., 1993. Individualism and the modern corporation:implications for innovation and entrepreneurship. Journal of Management 19,595612.

Morris, M.H., Davis, D., Allen, J.W., 1994. Fostering corporate entrepreneurship:cross-cultural comparisons of the importance of individualism versus collecti-vism. Journal of International Business Studies 25, 6589.

Mowery, D.C., Oxley, J.E., 1995. Inward technology transfer and competitiveness:the role of national innovation systems. Cambridge Journal of Economics 19,6793.

Mueller, V., Rosenbusch, N., Bausch, A., 2013. Success patterns of exploratory andexploitative innovation. Journal of Management , http://dx.doi.org/10.1177/0149206313484516.

Nachum, L., Zaheer, S., Gross, S., 2008. Does it matter where countries are?proximity to knowledge, markets and resources and MNE location choice.Management Science 54, 12521265.

Nakata, C., Sivakumar, K., 1996. National culture and new product development: anintegrative review. The Journal of Marketing 60, 6172.

Ritter, T., Gemunden, H.G., 2004. The impact of a companys business strategy on itstechnological competence, network competence and innovation success. Jour-nal of Business Research 57, 548556.

Rycroft, R.W., Kash, D.E., 2004. Self-organizing innovation networks: implicationsfor globalization. Technovation 24, 187197.

Schumpeter, J.A., 1942. Capitalism, Socialism, and Democracy. Allen&Unwin,London.

Schwartz, S., 1994. Beyond individualism/collectivism: new cultural dimensions ofvalues. In: Kim, U., Triandis, H.C., Kagitcibasi, C., Choi, C., Yoon, S.C. (Eds.),Individualism and Collectivism: Theory, Methods, and Applications. Sage,Thousand Oaks, CA, pp. 85119.

Seltzer, K., Bentley, T., 1999. The Creative Age: Knowledge and Skills for the NewEconomy. Demos.

Shane, S., 1995. Uncertainty avoidance and the preference for innovation cham-pioning roles. Journal of International Business Studies 26, 4768.

Shane, S., 1993. Cultural inuences on national rates of innovation. Journal ofBusiness Venturing 8, 5973.

Shane, S.A., 1992. Why do some societies invent more than others? Journal ofBusiness Venturing 7, 2946.

Shane, S., Venkataraman, S., MacMillan, I., 1995. Cultural differences in innovationchampioning strategies. Journal of Management 21, 931952.

Song, M.X., Parry, M.E., 1997. The determinants of Japanese new product successes.JMR, Journal of Marketing Research 34, 6476.

K. Efrat / Technovation 34 (2014) 1220 19

Steensma, H.K., Marino, L., Dickson, P.H., 2000. The inuence of national culture onthe formation of technology alliances by entrepreneurial rms. Academy ofManagement Journal 43, 951973.

The Hofstede Center website at: http:/geert-hofstede.com/dimensions.html(accessed on June 1st, 2013).

Thoenig, M., Verdier, T., 2003. A theory of defensive skill-biased innovation andglobalization. American Economic Review 93, 709728.

Tihanyi, L., Grifth, D.A., Russell, C.J., 2005. The effect of cultural on entry modechoice, international diversication, and MNE performance: a meta-analysis.Journal of International Business Studies 36, 270283.

Vecchi, A., Brennan, L., 2009. A cultural perspective on innovation in internationalmanufacturing. Research in International Business and Finance 23, 181192.

Venaik, S., Brewer, P., 2010. Avoiding uncertainty in Hofstede and GLOBE. Journal ofInternational Business Studies 41, 12941315.

World Bank, 2010 at: http:/data.worldbank.org/indicator; (accessed on June 1st,2013).

Kalanit Efrat (Ph.D., Marketing, University of Haifa, Israel, 2008) is alecturer of marketing at the Ruppin Academic Center, Israel.

K. Efrat / Technovation 34 (2014) 122020

The direct and indirect impact of culture on innovationIntroductionLiterature reviewInnovation and the national innovation system (NIS)Innovation in a cultural perspective

MethodMeasures of innovation inputsMeasures of innovation outputs

ResultsDiscussionConclusionAppendixReferences