Understanding the Paradox in the Results of National Cultural Distance Impact on International Joint Venture Performance: A Narrative Review of Prior Literature

,


Introduction
The literature on international joint ventures (IJVs) continues to grow owing to their crucial importance in global market competition (Tower et al., 2019). Since the 1980s, many businesses have entered into IJVs to expand their operations globally and capitalise on market imperfections (Tower et al., 2019). An IJV is "an independent legal, organisational entity jointly owned by two or more parent companies" (Yeheskel et al., 2001, p. 82); a strategic partnership between firms intended to increase competitiveness in a new market (Liu et al., 2020;Ozorhon & Altun, 2017). For its parent companies, an IJV can be an opportunity to combine local resources with in-depth knowledge of the target market, to share both costs and risk (Jin & Wang, 2021), to learn new skills, and to benefit from technologies employed in the venture (Lu & Ma, 2015). Despite their popularity, many IJVs suffer from instability and poor performance, leading to a high rate of failure (Liu et al., 2020;Sim & Ali, 2000).
Scholars of IJV have identified numerous critical factors affecting IJV performance among partners in cross-national alliances (Tower et al., 2019). National cultural distance (NCD) between partner firms is conventionally considered one of the main contributors to unsatisfactory IJV performance because this can reduce resource commitment and hinder communication (Jin & Wang, 2021;Sarkar et al., 2001;Tower et al., 2019;Triki & Mayrhofer, 2016;). National culture (NC), defined as "the collective values, norms, and priorities that are common to the members of a nation" (Ren et al., 2009, p. 821), is stable (Hofstede, 1983), has an unconscious and deep-rooted influence on people's beliefs and behaviours (Liu et al., 2020), and is reflected in firms' values (Hennart & Zeng, 2002). Therefore, IJV partners must carefully consider the impact of NCD during both pre-formation partner selection and post-formation management (Tower et al., 2019;Yeheskel et al., 2001). The international business literature has traditionally used Hofstede's (1983) five dimensions (individualism versus collectivism; large versus small power distance (PD); strong versus weak uncertainty avoidance (UA); masculinity versus femininity; and long-term versus short-term orientation (LTO)) and Kogut and Singh's (1988) Composite Index (KSCI) to measure NCD between countries. Meanwhile, some researchers use country of origin to Despite this inconclusive body of results, scholars broadly agree on the negative correlation between NCD and venture performance (Nielsen, 2007). NCD is a country-level variable and hard to control, and regardless of venture type (intrafirm JVs, cross-national JVs, traditional IJVs, and trinational IJVs), high NCD means high risk (Cui et al., 2019). This means many firms are drawn to IJVs as a means to mitigate the negative effects of NCD (Kogut & Singh, 1988). Interestingly, Meshi and Riccio (2008) found that NCD is both a leading factor pushing companies to enter into IJVs and a leading cause of IJV dissolution. This apparent paradox creates a dilemma, leaving managers torn between waiting until they have found a compatible partner firm and seizing new opportunities quickly despite the challenges posed by cultural difference (Meschi, 1997). To the authors' knowledge, the current study is among the few responding to the need for reviewing the factors contributing to the inconsistency in results of past works regarding the relationship between NCD and IJV performance.
Through an in-depth review of prior literature on the relationship between NCD and IJV performance, this paper will contribute to the previous literature and guide future research by explore these apparent inconsistencies, providing new insights into the causes of these contradictory results, and trying to fill this knowledge gap. Theoretically, the relationship between NCD and IJV is not direct, and many moderators and mediators can better explain this relationship such as cultural intelligence (Cui et al., 2019), trust between partners (Ng et al., 2007), and conflict management approach (Liu et al., 2020). For instance, adopting cooperative conflict management can weaken the negative impact of NCD on IJV performance (Liu et al., 2020). This study will also provide valuable practical conclusions, which will help managers to reduce the negative impact of NCD, achieve the best possible outcomes for their IJVs, and improve overall venture performance. For instance, recruiting managers who are culturally intelligent can play a vital role in weakening the detrimental effect of NCD. This focus on the literature review leaves little room for empirical analysis. To correct this limitation, empirical analysis should be conducted in future studies to test the authors' assumptions and increase the study's validity. Additionally, NCD is not only the factor affecting IJV performance; therefore, scholars should take into account other determinants such as organizational culture, upon which many scholars agree on its durability and significance compared to NCD.
In the following section we will further explain the method used to select the literature reviewed in this paper. After that, we will conduct a narrative review to compare and contrast past research and identify the academic gap. We will then analyse and discuss the implications of this research for decision-makers, managers, and academics, and we will indicate further research paths. To conclude, we will summarise the paper's key findings.

Methodology
This paper offers a narrative review of prior research published on the impact of NCD on IJV performance. As a narrative review, it will identify patterns, gaps, and disagreements in the existing literature and provide foundations for future research (Ferrari, 2015). However, such a review is not reproducible, its assumptions may be subjective, and its selection and assessment tendencies are unknown (Ferrari, 2015). To select studies for assessment in this paper, we used keyword searches for "national cultural distance," "international joint venture," "IJV performance," and "cultural differences," which are directly related to our topic. We included literature that includes and concerns both the dependent variable and independent variable. We considered only research-oriented studies whose authors empirically analysed connections between the independent variable (NCD) and dependent variable (IJV performance) and are listed in the Association of Business Schools' list of ranked journals (ABS list) in the fields of international business, management, and strategy. We excluded methodological, theoretical, and conceptual papers, books, and published literature reviews, including meta-analyses. Based on the definition of "IJV" presented earlier and Table 2, we selected only studies that refer to traditional IJVs (host-country-and home-country-based firms) because these involve partners' heterogeneous experiences and diverse national cultural backgrounds (Beamish & Kachra, 2004). We did not define a date range, instead focusing on highly cited articles and incorporating even the earliest works on our topic such as Park and Ungson (1997) when studying the positive relationship, Meschi (1997) and Hennart and Zeng (2002) when studying the negative relationship, Barkema and Vermeulen (1997) in explaining the mixed relationship, and Fey and  for the non-relationship between NCD and IJV performance.  Makino and Beamish (1988;801) Owing to the aforementioned intricacies, a simple keyword search was insufficient. As a result, we then examined the full titles, keywords, and abstracts of 58 articles. Following this cross-check, we selected 39 papers for analysis to conform with the study's purpose and time constraints. Then we divided the literature based on the relationship between the study's variables (positive, negative, mixed, and no relationships) to compare and contrast with the results with the same table and between tables to define the reasons behind the inconsistencies and provide new insights for managers and academics who are interested in boosting IJV performance in the global economy.

Literature Review
Prior literature on this subject has produced contradictory findings; different scholars have concluded that NCD has either a positive, negative, mixed, or insignificant impact on IJV performance (Tower et al., 2019). We have organised our selected literature into those four camps. All the authors we examined conducted their research through questionnaires or databases, they all made use of deductive methodology and mid-level theory, and they almost all explored dyadic IJVs with precisely two parent partner companies (see Tables 3, 4, 5, and 6). In studies employing questionnaires, these were primarily sent to IJV managers to evaluate satisfaction with the venture in mas.ccsenet.org Vol. 17, No. 1; question (Li et al., 2010) and its goal achievement (Ng et al., 2007), as well as their perceptions of NCD (Liu et al., 2020). Databases were used to quantify IJV longevity (Pangarkar, 2003), KSCI (Fey & Beamish, 2001), and Hofstede's scores (Malik & Zhao, 2013).
In each stream, we will contrast papers critically to provide a better understanding on the topic and determine literature gaps that can be addressed by further research.

The Positive Relationship between NCD and IJV Performance
Researchers in this section agree that NCD between partners could increase IJV performance (Table 3). Most authors have used transaction costs theory (TCT) to explain the favourable association between NDC and IJV performance (Gao et al., 2017;Ng et al., 2007;Park & Ungson, 1997). According to TCT, strategic alliances such as IJVs reduce costs and share risks, particularly when firms lack host-country knowledge (Kwok et al., 2019). Scholars in this section have found that greater NCD directly correlates with better IJV performance (Table 3). Hu and Chen (1996) observed that 10% of Sino-U.S. ventures were successful, compared to 6.40% of Sino-European ventures, 3.58% of Sino-Japanese ventures, and 5.33% of Chinese-Hong Kong ventures. Similarly, Park and Ungson (1997) found that JVs between Japanese and U.S parents survived longer than between two U.S. parents. They argued that Japanese managers have stronger trust, longer-term orientation, and an intent to retain their good mas.ccsenet.org Vol. 17, No. 1; reputation, reducing the likelihood of IJV dissolution. Li et al. (2010) concluded that trust between Chinese firms and partners with significant NCD helps to avoid conflict and foster strategic cooperation. Conversely, Ng et al. (2007) argued that trust does not moderate the relationship between NCD and venture performance. They found that NCD contributes to reaching financial targets from the perspective of non-Chinese senior executives, and nonfinancial goals from the perspective of local Chinese partners. Koch and Koch (2018) found that although NCD negatively impacts intangible trust (feelings or beliefs about the goodwill of the other), it in fact increases tangible trust (concrete measurement of the other's competence-effective deployment of resources; weighing of benefits versus risks; evaluation of reliability based on past performance). This is crucial for IJV survival in that building tangible trust between partners helps achieve IJV goals and increases venture longevity (Koch & Koch, 2018). They suggested that with clear communication regarding readiness to devote additional resources in future if required, a foreign partner can overcome the detrimental impact of NCD on intangible trust. Table 3 were large, focusing on IJVs with Chinese partners operating in the manufacturing sector. Fadol and Sandhu's (2013) study was the only one conducted in the United Arab Emirates, and the only one to centre on the gas and oil industry. Different scholars have measured NCD using variables such as country of origin (Hu & Chen, 1996), KSCI (Koch & Koch, 2018;Li et al., 2010;Park & Ungson, 1997), Hofstede's (1991) five dimensions (Fadol & Sandhu, 2013), and a combination of KSCI and partner perception to examine both country-level and firm-level NCD (Ng et al., 2007). IJV performance was measured both objectively and subjectively (Tower et al., 2019). Objective metrics focus either on venture success, using profit to categorise financial performance (Hu & Chen, 1996), or venture longevity, measured in years and months (Park & Ungson, 1997) or in quarters (Koch & Koch, 2018). Subjective metrics focus on managers' self-reported satisfaction with their ventures and the achievement of these ventures' goals (Fadol & Sandhu, 2013;Li et al., 2020;Ng et al., 2007). According to Geringer and Hebert (1991), scholars make use of this kind of subjective data because they are easily obtained and tend to reflect financial success. Such data are particularly useful in analysing IJVs based in China, where financial reports and operational data are generally not available to the public .

Most samples in
The articles in this stream highlight the positive impact of NCD on IJV performance and emphasise the role of trust in strengthening partner relationships.

The Negative Relationship between NCD and IJV Performance
In this section, researchers argue that NCD is a leading factor contributing to IJV instability and failure (Table 4). Scholars in this stream have attempted to use various theories to explain the negative influence of NCD on IJV performance. Hennart and Zeng (2002), for instance, used real options theory to describe IJV dissolution. According to this theory, when unanticipated changes in the business environment lead parents to re-evaluate venture value and decide to exercise their options, one parent will take over the other's stake (Hennart & Zeng, 2002). Gao et al. (2017), Liu et al. (2020), and Jin and Wang (2021) used TCT to demonstrate that IJVs with NCD partners are difficult to manage, which leads to higher transaction costs (e.g., negotiation costs) and conflict (Hennart & Zeng, 2002). Pak et al. (2009) and Jin and Wang (2021) concluded that partners can perceive cultural differences as risks, which negatively impacts the performance of their joint ventures. Analysing a database of ventures between 1980 and 1994, Hennart and Zeng (2002) found that the average longevity of Japanese-Japanese JVs was greater than that of Japanese-American IJVs. They claimed that cultural differences can hurt communication, conflict resolution, and partner agreement on common goals, increasing the likelihood of venture dissolution. Park et al. (2009) reported that NCD between Koreans and non-Korean partners hinders the learning process, which similarly leads to poor IJV performance. In two surveys of IJV managers in the Turkish construction sector, Ozorhon et al. (2010) and Ozorhon and Altun (2017) focused on cultural harmony rather than cultural distance. Complementing other studies in this stream, they concluded that increased cultural similarity between partners correlates with increased stability, venture performance, and partner satisfaction (Gao et al., 2017).
Other scholars argue that the relationship between NCD and IJV performance can be moderated by cultural intelligence and appropriate conflict management (Table 2). According to Cui et al. (2019), cultural intelligence can mitigate the negative effects of NCD on IJVs in China. Liu et al. (2020) found that cooperative JV management, focusing on a "win-win" approach, is a reliable means to achieve mutually acceptable outcomes, particularly when partners come from different cultural backgrounds. Meschi (1997) argued that the longer an IJV lasts, the fewer NCDs between partners appear. He proposed that as IJV partners get to know each other over time, the intensity and impact of NCD in the venture declines. A cross-sectional survey of 200 Chinese-foreign IJVs by Jin and Wang (2021) demonstrated that NCD negatively moderates the relationship between resource complementarity and IJV performance. Jin and Wang concluded that NCDs make complementary resource integration difficult and adversely affect the IJV performance.
Scholars in this stream use diverse theories, contexts, and variable measurements, but overall they describe a common, negative impact on IJV performance when choosing a partner from a different NC.

The Mixed Relationship between NCD and IJV Performance
Investigating the direct relationship between NCD and IJV performance, these researchers all reported inconsistent and contradictory findings. For example, Makino and Beamish (1998) found that heterogeneous NCD between partners can both increase the IJV's financial performance and decrease its survival rate (a result of high management complexity). Most scholars in this stream examined the impact of Hofstede's (1993) five dimensions separately ( Table 5) and found that different NCD dimensions have different impacts on venture performance (Pothukuchi et al., 2002). Other researchers used different theories-for instance, Malik and Zhao (2013) used the knowledge-based view to examine whether and how cultural diversity influences the duration of learning alliances in a high-tech setting. Other theories employed in this stream include the resource-based view (RBV) (Yeheskel et al., 2001), TCT (Jiang et al., 2011), the culturalist approach (Meschi & Riccio, 2008), and resource dependence (Kang et al., 2018). Individualism and masculinity differences enhance venture longevity at certain quantiles. PD differences reduce longevity at the median quantile. UA and LTO differences do not affect longevity across all quantiles.
According to Barkema and Vermeulen (1997), differences in UA and LTO harm IJV survival. They found that differences in PD have a smaller negative impact on survival, whereas differences in individualism and masculinity are not detrimental. Yeheskel et al. (2001) found that NCD between partners in individualism and UA improves IJV effectiveness, whereas differences in PD and masculinity are a hindrance (though not a significant one). By contrast, Pothukuchi et al. (2002) concluded that differences in masculinity have a positive effect, whereas differences in individualism, UA, and PD reduce partner satisfaction with IJVs. They explained that differences in masculinity may have led to admiration among Indian executives, who attributed strong IJV performance to their foreign partners' desire for organisational success. Jiang et al. (2011) partially supported Pothukuchi et al. (2002); they found that differences in UA have a detrimental effect on anticipated IJV duration, whereas differences in PD are insignificant. Zhao and Malik (2013) argued that distances based on masculinity, UA, and LTO extend alliance duration, whereas distances based on PD and individualism reduce it. Nevertheless, they found that composite differences based on all five factors were an overall boon to alliance duration, particularly for learning partnerships that value a diversity of knowledge, for example biopharmaceutical alliances (Malik & Zhao, 2013).
Using quantile regression, Tower et al. (2019) demonstrated that higher distances in UA and LTO do not affect IJV longevity across all quantiles, distances in PD have a negative effect at the median quantile, and distances in masculinity and individualism enhance IJVs' life in certain quantiles. Most importantly, Meschi and Riccio (2008) found that the impact of NCD on IJV survival fluctuates during the IJV's life cycle. They argued that NCDs have no impact on IJV during the first period, when partners enjoy a "honeymoon effect." In the second stage, the NCDs appear, hindering cooperation and increasing conflict, until finally the IJV matures, and these differences and tensions between cultures are taken on board and smoothed out (Meschi & Riccio, 2008).
To measure IJV performance, some scholars surveyed respondents to assess their satisfaction with the venture (Table 5). However, the majority calculated IJV longevity objectively, either in years (Barkema & Vermeulen, 1997), months (Pothukuchi et al., 2002), weeks (Malik & Zhao, 2013), or days (Tower et al., 2019). Because researchers in this stream concluded that each NCD has a different impact on IJV performance, most papers in Table 3 relied on Hofstede's dimensions to determine each dimension's effect on IJV performance. Notably, Meschi and Riccio (2008) instead used KSCI, whereas Ozorhon et al. (2008) combined KSCI with perceptual differences based on Hofstede's five measures to evaluate overall satisfaction and project performance.
Because scholars in this stream studied different NCD dimensions separately, comparing these results is difficult. This is complicated further by the use of so many different measurements (for both variables), different research instruments, and different samples and sectors.

No Relationship between NCD and IJV Performance
In this stream, researchers found that NCD does not affect IJV performance (Table 6). Many used RBV and TCT to clarify the relationship-or lack thereof-between NCD and IJV performance. RBV posits that NCD between partners increases heterogeneity and reduces the redundancy of resources, increasing the IJV's competitiveness and its longevity (Mohr et al., 2016). TCT, by contrast, holds that the cost of safeguarding against opportunistic behaviour increases when IJV partners are culturally distant (Mohr et al., 2016). Many studies have shown that NCD does not affect partners' satisfaction with the venture (Bener & Glaister, 2010;Gong et al., 2007); IJV learning (Anh & Baughn, 2011); the achievement of abnormal returns (Merchant, 2002); IJV longevity (Triki & Mayrhofer, 2016); or the ability to meet agreed targets (Dasi-Rodrigues & Pardo-Del-Val, 2015). Different scholars have provided four possible explanations for this apparent lack of impact. Some conclude simply that the role of NCD in alliance formation and success has been overemphasised (Pangarkar, 2003). Others hold that problems caused by differences in organisational climate have wrongly been blamed on NCD (Fey & Beamish, 2001). Still others posit that foreign partners become more familiar with local cultures over time, decreasing the significance of NCD (Beamish & Kachra, 2004;Meschi & Riccio, 2008). Finally, some find that partners are generally willing to adjust their cultural characteristics to accommodate each other during the venture period (Mohr et al., 2016).  Furthermore, many scholars in this stream name other variables influencing IJV performance more significantly than NCD. These include trust (Luo, 2002), the number of partners (Beamish & Kachra, 2004;Gong et al., 2007), and organisational culture (Bener & Glaister, 2010;Damanpour et al., 2012;Fey & Beamish, 2001). Luo (2002) found that trust among partners in Sino-foreign strategic alliances improves their performance despite NCD between parties. Several other researchers conclude that harmony in organisational culture has more impact than differences or similarities in NC. Among IJVs in Russia, for example, Fey and  found that NCD between partners had no impact on venture performance, whereas dissimilarities in organisational culture were actively detrimental. Similarly, in the Indian context, Damanpour et al. (2012) argued that practices rooted in organisational culture contribute more to IJV success or failure than do social beliefs and values rooted in national culture. In the same vein, Kwok et al. (2019) noted that organisational compatibility improves venture performance for Sino-foreign IJVs, whereas NCD between partners has no impact. In other words, when these academics used NCD as a control variable for IJV performance, they found its impact was negligible ( Table 6).
Scholars in this stream agree that NCD has no effect on IJV performance, and that other factors-trust and organisational culture, for example-can be far more impactful (Table 6).

Analysis and Critical Discussion
The rise of globalisation has pushed MNCs to extend their operations overseas (Tower et al., 2019), often using IJVs to pursue strategic goals (Jing & Wang, 2021). When researching ways to improve the performance of these ventures, scholars of business and management have turned again and again to NCD as a particularly significant variable (Tables 3, 4, 5). Nevertheless, academic findings on the relationship between NCD and IJV performance have been fragmented and inconsistent (Hennart & Zeng, 2002). In the previous section of this paper, only authors of six studies out of 39 found that NCD between partner firms tended to increase IJV performance (Table 3), whereas 10 scholars argued the opposite (Table 4). A further 10 concluded that not all NCD dimensions have the same impact on IJV performance (Table 5). Contradicting this prevailing view of the significance of NCD, 13 study teams found no meaningful relationship whatsoever (Table 6). We will now explore potential reasons for these inconsistencies and offer guidance for future research.
Most scholars conducted their research using one of two theories, TCT and RBV. For example, some researchers used TCT to demonstrate that partners from different NCs can experience greater mistrust and opportunistic behaviour, which in turn increases uncertainty and safeguarding costs (Hsieh et al., 2010). Others used RBV to show that NCD between partners can create a competitive advantage, for example by decreasing resource redundancy, because partners from different NCs tend to acquire different resources (Mohr et al., 2016). Most researchers focus on Western and emerging markets; only Fadol and Sandhu (2013) and Triki and Mayrhofer (2016) conducted their research in the Arab world. Fadol and Sandhu (2013) reported that data unavailability and consequent reliance on Western publications presented significant barriers to conducting research in the Arab context. As an example, Hofstede (1983) generalised the NCD scores for the entire Arab world based on results collected in just a few countries. Therefore, we suggest that future researchers should measure respondents' perceived NC differences through questionnaires to test the relevance of Hofstede's scores to their chosen national contexts.
The sample sizes for all studies were large enough to be properly representative. Most authors used single-side respondents, mainly IJV managers, because their evaluations correlate strongly with their parent firms' (Geringer & Hebert, 1991). They also generally have the most relevant knowledge, and if not, they can forward questionnaires to those more knowledgeable about the topic under study (Anh & Baughn, 2011;Jin & Wang, 2021). Fey and  reduced their sample size from 40 to 24 IJVs in Russia to allow for multiple responses from all three parties making up an IJV (Russian parent, foreign parent, IJV managers). However, this smaller sample makes it more difficult to generalise from their results. It is difficult to collect multiple responses in this way without either compromising the sample size or allowing the project to grow unmanageably large. As a workaround, we suggest surveying only one of the three parties in question, ideally the IJV managers themselves (Nielsen, 2007).
As shown in the earlier tables, scholars used different measures of IJV performance depending on study purpose and conditions. Objective measures include financial criteria such as profitability, and operational measures such as survival and determination rate (Liu et al., 2020). However, objective financial data are often unreliable (Fey & Beamish, 2001), hard to obtain (Bener & Glaister, 2010), or one-dimensional (Pothukuchi et al., 2002). Additionally, financial returns are not always a definitive measure of venture performance: IJV termination does not always mean failure, nor does profitability always mean success (Barkema & Vermeulen, 1997;Geringer & Hebert, 1991). Partner firms typically have other motives when venturing: organisational learning, for instance (Fey & Beamish, 2001;Ozorhon, 2008). Furthermore, Delios and Beamish (2001) and Lu and Ma (2015) demonstrated that managerial assessment of IJV profitability correlates negatively with venture dissolution, and that this reverse correlation is statistically significant. In other words, subjective managerial assessments of goal achievement and overall IJV satisfaction can serve as a meaningful alternative to objective data (Liu et al., 2020) where such data are not available. Authors of 20 out of 39 papers examined herein made use of subjective measures because these correlate well with objective data (Geringer & Hennart, 1991). Such measures can demonstrate the extent to which an IJV has achieved its overall objectives, including financial objectives (Bener & Glaister, 2010). However, self-reported measures of performance are often biased and may not reflect actual performance (Nielsen, 2007). According to Ozorhon et al. (2008), objective and subjective measures are complementary, not alternative. We conclude that wherever possible, academics should seek a multidimensional approach to measuring IJV performance to mitigate the aforementioned limitations (Shenkar, 2001). Ideally, governments should also encourage and incentivise firms to disclose financial information and similar data to academics. This would support more effective research, stronger venture performance, and, ultimately, economic and industrial development at the country level (Park et al., 2012).
The authors of studies in Table 5 argue broadly that each NCD dimension has a different impact on IJV performance, overwhelmingly relying on Hofstede's (1983) scores. Between 1967and 1978,000 questionnaires in 40 countries, Hofstede found that NC is stable, changes slowly, and is not disappearing. Barkema and Vermeulen (1997) showed similarly that cultural values remained stable over a 30-year period. However, more recent scholars argue that NCD may no longer be relevant, and that Hofstede's dimensions are in fact an obstacle to their research for a variety of reasons (Malik & Zhao, 2013). First, cultural boundaries differ from national ones (Barkema & Vermeulen, 1997). Second, NC is an intricate concept that cannot be summarised in a simple survey. In other words, because of the complexity of human culture, Hofstede's dimensions cannot offer a truly comprehensive overview of NCD (Barkema & Vermeulen, 1997). Finally, the framework is now decades-old (Malik & Zhao, 2013). For these reasons, some researchers measured firm-level NCD using respondents' perceptual differences based either on Hofstede's dimensions (Cui et al., 2019;Liu et al., 2020;Ozorhon et al., 2010) or on other questionnaire items (Anh & Baughn, 2011;Bener & Glaister, 2010;Luo, 2002;Mohr et al., 2016). Here we conclude that scholars should compare firm-level data (self-reported measures) with national-level data (Hofstede's scores and KSCI) to test the relevance of Hofstede's framework in the 21st century (Lu et al., 2006;Shankar, 2001).
To explore more specific aspects of the relationship between NCD and IJV performance, six of the 39 papers introduced moderating dimensions such as trust (Li et al., 2010;Ng et al., 2007), venture type (Beamish & Kachra, 2004), cultural intelligence (Cui et al., 2020), and approaches to conflict management (Liu et al., 2020), all of which can improve the relationship between NCD and IJV performance. According to Li et al. (2010), trust between Chinese firms and their partners strengthens the positive effect of NCD on IJV performance. However, Luo (2002) argued that trust has the same impact regardless of the NCD between partners, at least in terms of financial performance. We therefore advise managers to prioritise building trust with their partners, a crucial component of venture survival and goal achievement. After introducing IJV type (intrafirm, cross-national, or traditional IJV) as a moderator, Beamish and Kachra (2004) reported that NCD does not impact the performance of traditional IJVs because as foreign firms become more familiar with local cultures over time, they adjust their cultural characteristics to suit their partners (Mohr et al., 2016). Similarly, Meschi (1997) noted that the longer a venture lasts, the smaller the cultural differences between the partners. Thus when evaluating potential partner firms, managers should not overestimate the negative effect of NCD on performance because this will diminish over time. Instead, they should focus on the competitive advantage to be gained through a partner's resources. Examining still another factor, Cui et al. (2019) found that members' cultural intelligence reduces the negative impact of NCD on venture cooperation. Consequently, IJV partners should seek to assemble teams with relevant international work experience and provide cross-cultural training, particularly for new staff, because these measures can reduce conflict and improve cooperation when venturing with partners from different NCs. Finally, Liu et al. (2020) advised that to mitigate the detrimental impact of NCD on IJV performance, managers should adopt cooperative ("win-win") conflict management. Here we stress the importance of constructive resolutions to conflicts, and advise managers to avoid more competitive "win-lose" scenarios. Because NCD between partners can harm IJV performance, managers need to consider the aforementioned measures to mitigate this potentially negative impact and improve their ventures' outcomes.
Because most scholars agree that NCD has a direct impact on IJV performance, only a few included mediator variables to study more indirect effects (Damanpour et al., 2013;Dasi-Rodriguez & Pardo-Del-Val, 2015;On et al., 2013;Pak et al., 2009). Dasi-Rodriguez and Pardo-Del-Val (2015) found that parent firms' nationalities do not mas.ccsenet.org Vol. 17, No. 1; impact the achievement of cooperation agreement aims, but that NCD does interfere with the creation of a hybrid (or "third") culture. Such hybrid cultures in turn tend to increase partners' satisfaction with the alliance (Dasi-Rodriguez & Pardo-Del-Val, 2015). In a similar vein, Pak et al. (2009) concluded that heterogeneity in cultural backgrounds did not lead to partner conflicts, but did discourage partners' learning processes, leading to poor performance. Whatever the direct impact of NCD on venture performance, we conclude that firms should be mindful of these more indirect effects, and that managers should establish clear communication channels. These channels can facilitate both learning exchanges between partners and the cultivation of a hybrid culture.
This paper, like any other, has limitations that should be examined by further researchers. First, we did not use a questionnaire, database, or any other primary data to examine the relationship between NCD and IJV performance empirically. We instead relied on prior research and conducted a narrative review. Second, we studied only the impact of NCD, even though many scholars argue that the impact of organisational culture on IJV performance is just as substantial-or even more so (Damanpour et al., 2012;Fey & Beamish, 2001;Li et al., 2010;Ozorhon et al., 2017;Pothukuchi et al., 2002). Pothukuchi et al. (2002), for instance, found that "the presumed negative effect of partner dissimilarity on IJV performance originates more from differences in organizational culture than from differences in national culture" (p. 258). Third, the majority of IJVs examined in this paper are traditional and dyadic. Future scholars should centre on other types of IJV, particularly those with more than two parent companies, because these ventures experience higher costs and greater management complexities (Hennart & Zeng, 2002).

Conclusion
After comparing 39 academic articles, we found three principal causes for the contradictory findings of the prior literature: lack of consensus on measures of IJV performance (Liu et al., 2020;Nielsen, 2007;Pothukuchi et al., 2002); inconsistency in the conceptualisation and measurement of NCD (Tower et al., 2019); and the omission of moderating/mediating variables in the relationship between NCD and IJV performance (Liu et al., 2020;Pothukuchi et al., 2002). We identified limitations in earlier research particularly regarding measures for evaluating the key variables (NCD and IJV performance) and suggested appropriate solutions. We conclude that the impact of NCD on IJV performance is not straightforward and that researchers should introduce moderator and mediator variables in pursuit of a more nuanced understanding. We also acknowledge this study's lack of empirical analysis as its greatest limitation.
In the field of IJV management, we offer some practical guidance. We advise policy-makers to encourage firms to disclose financial data to academics. This will enable more precise research, which in turn will lead to improved IJV performance. We counsel managers not to be dissuaded by NCD because foreign partners often have valuable resources and can provide a competitive advantage in new markets. In addition, we advise that firms select culturally intelligent managers, provide cross-cultural training for IJV staff, and implement cooperative conflict resolution, all of which can decrease the negative impact of NCD. Crucially, we advise that managers build trust with their partners because this improves performance and diminishes opportunistic behaviour regardless of NCD.