Determinants of Staff Localization in Headquarters-Subsidiary-Subsidiary Relationships

Abstract

This paper investigates how cultural distance, the local experience of a foreign subsidiary, and the intensity of local competition jointly affect the staff localization of MNEs’ subsidiaries. While previous studies on the effects of cultural distance have mainly focused on the gap between home and host countries, we extend the existing “home-host” country perspective to the home-intermediary-host country relationship. This study regards Korea as an intermediary country and utilizes 520 observations from a unique survey conducted by the Export-Import Bank of Korea from 2006 to 2013. The results suggest that the impact of cultural distance on staff localization is a function of local experience and competitive environment in the home-intermediate-host relationship structure. This paper makes a theoretical contribution to our understanding of the behavior of multinational corporations by expanding the cultural distance perspective between the home and host countries explored in previous research to the home-subsidiary-subsidiary structure.

1. Introduction

The impact of distance between home and host countries on strategies adopted by multinational enterprises (MNEs) is well established in international business literature [1,2]. Distance has been found to affect organizational processes and consequences [3], including location choice [4,5], entry choice [6], corporate political activities [7,8], transfer of knowledge [9], and firm performance [10]. Although previous studies have focused primarily on the distance between home and host countries [11,12], the investment situation of MNEs is more complicated. When expanding into rather “distant” host countries, understanding the impact of cultural differences is vital for corporate sustainability [13], and the invariable international norms and practices do not necessarily suit each subsidiary. Researchers suggest that attempts to across national borders are likely to be more successful, with senior managers often leaving scope for local interpretation to tackle challenges they face in host countries [14]. And in practice, an MNE can invest indirectly in its foreign subsidiary in a host country through another foreign subsidiary in a third (neither home nor host) country, termed an intermediate country. While this home-ubsidiary-subsidiary investment (illustrated in Figure 1) by MNE is frequently observed [15], it has been largely understudied in the literature.

To address this research gap, we developed an integrated model that considers the effects of cultural distance between home and intermediate countries on the local responsiveness of MNEs. Our research question is to study the impact of cultural distance on staff localization and the contingent relationship of local experience and competitive environment in the home-intermediate-host relationship structure. We propose that cultural distance is a determinant of localization even in the parent-subsidiary-subsidiary context. Specifically, we focus on localization in human resource management in multinational subsidiaries. In addition, we explore whether or not the impact of cultural distance on staff localization is contingent on subsidiary capabilities and the competitive environment.

This study makes both theoretical and practical contributions. First, it builds on previous research that focused on the cultural distance between home and host countries. We broaden the scope of discussion by exploring the cultural distance between intermediate and home countries, providing a theoretical contribution in opening the black box of MNEs’ investment options. Second, we examine the foreign subsidiary itself as a unit of analysis by addressing the impact of subsidiary-level conditions on localization. While many studies of MNE subsidiaries deal with institutional differences, only a few address subsidiary-level constructs [16]. Thus, our findings deepen extant understanding of MNE localization and their subsidiaries’ entrepreneurial actions.

Our conclusions should be of interest to multinational parent firms interested in employing expatriates, granting autonomy to their subsidiaries, and developing market strategies. Because MNE subsidiaries often lack knowledge of local market conditions, they are likely to employ local employees to improve their local competitiveness. In addition, organizational experience over time can improve a subsidiary’s market knowledge, which results in parent firms’ granting more autonomy [17]. In contrast, increased competition in culturally dissimilar markets can cause subsidiaries to orient toward a globally integrated strategy using expatriate employees [18,19]. We suggest that the impact of cultural distance on staff localization depends on the MNE subsidiary’s local experience and the intensity of competition in the host country. Specifically, we argue that while a subsidiary’s local experience can be a positive moderating factor, the intensity of competition in the host market negatively moderates the effect of cultural distance on localization of staffing. Panel data of 520 MNE subsidiaries over five years provides general support for our hypotheses.

2. Background and Hypothesis Development

2.1. Local Responsiveness and Staff Localization of MNEs

Multinational corporations must choose between globally integrated or locally responsive strategic orientations. The former involves control of foreign subsidiaries within the parent firm, while the latter separates them from domestic operations and requires adaptation to local conditions [20,21,22,23,24]. Most MNEs attempt to improve their global competence by establishing networks with overseas subsidiaries while dealing with institutional obstacles associated with local market heterogeneity. Whether an MNE succeeds depends mainly on the extent to which it responds to local conditions [25,26].

Many previous studies examined the determinants of a global integration strategy [27,28]. However, some investigated the determinants of local responsiveness by shedding new light on subsidiary roles and explaining subsidiary autonomy from headquarters [29,30,31,32]. In practice, most MNE subsidiaries respond to local adaptation by hiring a local chief executive officer or employees, creating a joint venture with local firms, and using local embedded marketing [33]. Subsidiaries with local experience tend to cope effectively with problems specific to the location, which leads to superior performance [34,35], but acquiring local market knowledge through market transactions can be challenging [36]. For example, local employees can provide unique knowledge about social customs, local business practices and customer orientation [37]. Likewise, MNEs are more likely to use host-country nationals (HCNs) rather than parent-country national (PCN) expatriates or third-country national (TCN) expatriates to respond to local pressures [38].

Subsidiaries’ motivations for staff localization vary. For example, MNE subsidiaries in China conform to government pressures such as legal restrictions [39]. In addition, they must deal with socio-cultural differences, particularly when operating in culturally dissimilar markets such as emerging economies. For instance, “guanxi” is a locally accepted practice that fosters connections that build formal institutional support in China [40,41]. Also, MNEs are often aware of the considerable cost of using expatriates rather than local employees [41]. In addition, the effectiveness of expatriates relies on contextual variables, including cultural distance [42], which can also encourage the hiring of local staff.

2.2. Cultural Distance

International business (IB) scholars have introduced the concept of distance between countries in the form of cultural, institutional, and psychological dimensions to explain the relationship between MNEs and their subsidiaries [42]. Cultural distance refers to the differences that individuals from different countries may have in observing certain behaviors, affecting the transfer of work practices and methods from one country to another [43]. This may increase the costs of obtaining information and interrupt communications, making it difficult for local subsidiaries to integrate, apply their own routines, and adjust products [7].

For organizations engaged in international activities, the cultural distance may be crucial. For example, it is commonly believed that a greater cultural distance means a larger cultural difference, which explains why a foreign firm in culturally dissimilar markets deals with institutional obstacles attributable to the liability of foreignness [44]. Hofstede et al. [43] argued that this is because culture is the basis of institutional arrangements. More precisely, culture can be regarded as a part of the informal system of an environment, which is the basis of formal systems [45]. When multinational companies enter an institutional setting with different rules, they must meet social expectations, show social responsibility in the host country, and establish social legitimacy. Difficulty obtaining social legitimacy is a function of the cultural distance between the country of origin and the host country [46]. This makes multinational companies more inclined to invest in countries with cultures similar to those of their home countries [47]. Many researchers have analyzed the effects of cultural distance on strategic decisions such as location choice [48], degree of ownership [49], entry mode [1,6], transfer of practices [10,50], and even performance [51].

2.3. Hypothesis Development

Once a host country has been targeted, entry strategies must be matched with the cultural distance to that country to enhance competitive advantages resulting either from a small cultural distance or the ability to mitigate the negative impact of a large distance. The existing cultural distance is a measure of the difference between home and host countries, and heterogeneity is assumed due to characteristic differences of the countries. However, the investment situation of MNEs can be complicated. In practice, there are many foreign subsidiaries in a host country in which their individual parent firms indirectly invest through a subsidiary in another country, requiring us to define this third country as an intermediate country. A subsidiary in an intermediate country mediates between its headquarters in the home country and the subsidiary in the host country. Therefore, the subsidiary in the host country faces pressure from two sides: one from the intermediate subsidiary and the other from the parent firm [16]. The home and host country effects still affect staff localization even when the subsidiary in the host country receives investments from the intermediate firm rather than the parent firm.

Facing the dual pressures of globalization and local response, multinational companies may choose a competitive strategy, such as a global strategy or multi-domestic strategy, and then select staffing practices for their subsidiaries appropriate to the strategy [52]. An MNE often balances its strategic needs by responding to both global integration and local responsiveness. In terms of global integration, greater cultural distance leads to greater use of PCN expatriates because different cultures result in uncertainty, risk, and information asymmetry between the headquarters and its subsidiaries [53,54]. In terms of local responsiveness, greater cultural distance leads to more employment of local staff [41,54].

In the former case, because the subsidiary makes decisions and conducts management operations under the control of headquarters and therefore benefits from the competency of the resource-rich headquarters as a member of a multinational company, how the competency of headquarters is applied at the local level is essential. In the latter case, because it is a foreign company operating at a local level, exploration competency, which manages competition with local companies and relationships with governments and develops local competency that headquarters do not have, is also important. Subsidiaries in mediating countries, which are under pressure to use both competencies, can go abroad to reduce costs or explore global markets, regardless of resource utilization at headquarters or independent exploration activities of subsidiaries [23]. Although the subsidiary makes local investment decisions on its own, the subsidiary cannot ignore the influence of headquarters because it operates within the management system and culture of headquarters. If headquarters invests in the host country through a subsidiary in a mediating country-a bypass strategy-the influence of headquarters cannot be ignored [16]. Regardless of the presence of a subsidiary in the mediating country or the level of autonomy of the subsidiary, the cultural distance between home and host countries can affect the localization of the subsidiary in the host country. We can therefore expect that:

Hypothesis 1 (H1).

Cultural distance between home and host country is positively related to the employment of local managerial staff.

Foreign subsidiaries are embedded in the local environment, which requires responsiveness to the local context, as a bypass strategy to overcome heterogeneity between the home and host country [55]. However, if the cultural distance between the home and intermediate countries is considerable, a subsidiary in the intermediate country also has to overcome the liability of foreignness [16]. As a result, headquarters may apply new institutions through localization. When a country such as China has undeveloped systems or relies heavily on informal transactions, a coercive isomorphism can motivate to increase localization [40].

Previous research has explored the importance of congruence between mitigating cultural conflicts in MNEs and staff localization strategies [52]. The unique Chinese social culture and internal labor market have been identified as the core factors determining Chinese human resource management practices [56]. Successful MNEs are likely to mitigate cultural conflicts between home and target countries by developing staff localization strategies that accurately interpret the highly heterogeneous environment and lead to appropriate actions [5]. To avoid cultural friction between local norms and a company’s practices, many firms, and those from western countries, in particular, will tend to invest temporarily where the culture and social norms are similar (and the cultural distance is relatively smaller), as is the case in Korea and Japan [3,57]. Drawing on staff localization strategies identified in previous research [58], we hypothesized that cultural distance between home and intermediate countries would be positively related to employment for local managerial staff:

Hypothesis 2 (H2).

Cultural distance between home and the intermediate country is positively related to the employment of local managerial staff.

Foreignness is a burden because foreign companies are less familiar with the local market and business environment than their local counterparts [2]. As far as China is concerned, foreign companies need to employ more than a certain proportion of local employees. If they have local managers or invest in local companies or state-owned companies, it is usually easier to set up companies in the registration process. If the cultural distance between home and host countries is large and there is great heterogeneity, it may be challenging to overcome external responsibility. Subsidiaries can use their experience, ability and understanding of the local business environment to offset external responsibilities. With the expansion of cultural distance, the ability of relatively independent subsidiaries will have a greater offset effect. If local subsidiaries can avoid relying on the headquarters or subsidiaries of intermediary countries, the headquarters can accept greater localization [59,60]. Therefore, if the cultural distance is large, the subsidiary will actively hire local people, and with the improvement of the subsidiary’s experience and ability, the company will be more likely to hire local managers who can make crucial decisions.

Because a subsidiary can gain internal competency through operational experiences by connecting local customers, the subsidiary can make decisions based on its experiences rather than the guidelines of headquarters when those guidelines are a poor fit for local conditions. It has superior alternatives, even if a subsidiary follows unilateral guidelines established by headquarters in the early stages of business development. This will give a subsidiary an independent role in the local market rather than one that depends on the direction from headquarters [22,28]. Local business experiences can allow a subsidiary to exercise its absorptive capacity to capture new business opportunities and introduce new management activities and performance standards. Therefore, the empirical capacity of a subsidiary can be a factor in its ability to increase localization.

Hypothesis 3 (H3).

A subsidiary’s local experience positively moderates the impact of the cultural distance between home and host country on the employment of local managerial staff.

From two perspectives, the local competitive environment may be seen as an external responsibility. First, if competition in a local market is fierce, and overseas subsidiaries compete aggressively with local companies, performance may suffer directly. From an uncertainty perspective, the competitive strategies in the home country may not be effective in the local country, and it is possible to spend additional resources on developing new competition strategies such as analyzing competition structure and conducting marketing [61]. These factors may make it challenging to achieve high levels of performance [62]. However, intense competition is not necessarily a negative factor as it could positively affect innovation in the local country. Cultural distance and local competitive environment may promote localization, and both aspects are based on the characteristics of the local market. However, a combination of the two could raise the pressure from global integration rather than generate a local response.

The existing competition environment is regarded as a foreignness liability comparable to cultural distance. To address it, a foreign company can improve connections with local customers by hiring local employees and seeking guanxi relationships with governments or suppliers by localizing managerial positions [40]. The cultural distance at the national level and the intense competition at the industrial level can be understood through the liability of foreignness [16]. As foreign companies feel their limitations as a foreign company, the stronger the competition, the more prominent localization becomes.

This paper deals with the control of subsidiaries, which receive knowledge and competence from headquarters (illustrated in Figure 2). Companies operating in an intermediate country can utilize the subsidiary’s capabilities in that country. When competition with local companies is intense, foreign companies can acquire local knowledge and build competency by hiring local staff. This can involve simply imitating the competitive strategies used by local competitors. Multinational companies can also increase the number of expatriates to transplant their management methods because, as a first-mover, they can develop innovative marketing and effective management strategies that differ from local competitors [63].

Hypothesis 4 (H4).

The intensity of local competition negatively moderates the impact of cultural distance between home and host country on the employment of local managerial staff.

3. Sample and Methods

3.1. Data Source

We utilized 11,002 observations made by the Export-Import Bank of Korea from 2006 to 2013 of the investment statuses of Korean companies operating in China. Focusing only on companies that placed their headquarters in foreign countries, we used the Stata 14 program as an analysis tool after data processing. In the data processing, after excluding data with missing values of the home country or selecting China as the home country, 868 items remained. We identified multiple business-investment cases within one company, leaving 676 items. We then removed all except the largest subsidiaries, leaving 636 items after correcting errors in the original materials. Finally, we removed every observation with a home country of Labuan, Mauritius, Bahamas, Cayman Islands, International Finance Corporation, Virgin Islands, Cyprus, and Malta, which are known as paper-company countries, and then selected 570 items (205 companies by each home country) as analysis subjects. As these constituted unbalanced panel data that do not have every object (i) depending on time (t), removing items with missing values or multicollinearity in the panel data analysis process left 520 items as analysis subjects and 186 panel groups (subsidiaries). After data preprocessing, we revised the observation years to the period between 2006 to 2010. The minimum observation value of each year per group was 1 year, and the maximum observation value was 5 years. The average observation value was 2.7 years.

3.2. Measurement

3.2.1. Dependent Variable

This paper calculated the proportion of local managers of total managers by using the number of local managers and the number of Korean managers suggested in the survey.

3.2.2. Explanatory Variables

By using the four cultural dimensions developed by Hofstede et al. [43] and Kogut and Singh [1] of power distance (the solidity of social class; PDI), individualism-collectivism (IDV), masculinity-femininity (task orientation-human orientation, MAS), and uncertainty avoidance (UAI), we determined cultural distances for each country. The cultural distance between home and host and cultural distance between home and intermediate country are included in the analysis. The cultural distance calculation formula was derived using formula (1). For example, China’s PDI, IDV, MAS, and UAI values were 80, 20, 66, and 30, respectively, and Korea’s values were 60, 18, 39, and 85, respectively. The cultural distance between two countries is calculated as

$$Cultural Distanc{e}_j={{\displaystyle \sum }}_{i=1}^4\sqrt{{\left(I_{ij}-I_{iu}\right)}^2}$$

(1)

where $I_{ij}$ and $I_{iu}$ are the values of the standardized characteristic i corresponding to countries j and u, respectively.

We established an investment period, which is a substitute for the experience capability of a subsidiary. We then calculated the investment period with investment age using the initial year when the subsidiary of the intermediating country invested in the local subsidiary. The intensity of competition considered a liability of foreignness at the industrial level utilized a survey item to describe competitive relationships with local companies. The answer sheet consisted of a 5-point scale of very poor, poor, neutral, good, and very good, with very good assigned a value of 1 point and very poor 5 points. The larger the observation value, the stronger the competition.

3.2.3. Control Variables

This paper measured control variables of subsidiaries in the mediating country (Korea), such as listed corporations, company age, sales, and whether they engage in manufacturing. Listed corporations were divided into listed corporations, designated listed corporations, externally audited corporations, generally listed corporations, designated KOSDAQ corporations, listed KOSDAQ corporations, closure corporations, and merged corporations, with a value of 1 given to listed corporations, designated listed corporations, designated KOSDAQ corporations, and listed KOSDAQ corporations, and 0 assigned to the others. We used the year of establishment of the Korean corporation, and natural-log values were used for sales. In the case of manufacturing companies, because most categories were divided into manufacturing and service from the survey items, we coded the manufacturing industry as 1 and the service industry as 0.

For control variables of the local-country (China) subsidiaries, we selected an asset scale, whether the manufacturing industry exists, and whether independent investment exists. We utilized a survey item related to the total assets of a local subsidiary for the asset scale and used an industry code for the local corporations to determine whether a manufacturing industry exists. Both control variables of the intermediating and local countries were used after referencing prior research.

The controlling factors of local experience of local companies take advantage of the number of business years and the reverse causality in the basic research model. Converting the business period into the empirical competency according to entering time is problematic because the amount of experience over time and its quality can differ among companies. Local experience and knowledge tend to increase over time, but the amount of the increase depends on the subsidiary’s ability to adapt to local conditions. For example, when investing in local companies, if a company enters a market through a joint venture or syndicated investment with local companies, the entering company can adapt more rapidly because it can learn the capabilities of the local company. However, in the case of a single investment, it is difficult to assume that the entering company can accumulate a comparable amount of experience during the same 1-year period because the company does not immediately recognize the local environment and practices. We, therefore, controlled for whether a single investment exists and applied the experience of subsidiaries as a controlling variable. For whether a single investment exists, we coded a single investment as 1 and the others as 0, out of each observation value of a single investment, joint venture, and syndicated investment, using the survey items related to the investment types.

3.3. Methodology

We conducted a panel analysis on 520 observations using Stata 14, a statistical software package. Our analysis involved 186 headquarters-intermediating subsidiaries and local subsidiaries out of the Export-Import Bank of Korea. After conducting a model test to determine the panel data, we conducted correlation analysis in

Table 1. Illustration of “Home-Subsidiary-Subsidiary” Investment.

		Mean	SD	1	2	3	4	5	6	7	8	9	10	11
1	LOCAL	0.74	0.34	1.00
2	LISTED (IM)	0.3	0.46	0.00	1.00
3	AGE (IM)	25.76	12.98	0.16 ***	0.34 ***	1.00
4	SIZE (IM)	25.7	1.61	0.03	0.51 ***	0.28 ***	1.00
5	MNF (IM)	0.87	0.34	0.07 *	−0.07 *	−0.02	−0.20 ***	1.00
6	SIZE (HS)	16.64	1.34	−0.01	0.23 ***	0.16 ***	0.54 ***	0.06	1.00
7	MNF (HS)	0.89	0.32	0.11 **	−0.14 ***	0.05	−0.36 ***	0.45 ***	0.09 *	1.00
8	WOS (HS)	0.65	0.48	−0.11 ***	0.00	−0.22 ***	−0.13 ***	−0.12 ***	−0.33 ***	−0.07	1.00
9	CD (HM-IM)	68.33	14.17	0.01	0.01	−0.18 ***	−0.05	−0.08 *	−0.12 ***	−0.04	0.09 *	1.00
10	CD (HM-HS)	75.51	14.67	0.14 ***	0.11 **	0.03	0.07	−0.08 *	−0.16 ***	−0.03	−0.04	0.31 ***	1.00
11	LOCEXP	6.78	4.1	0.05	0.15 ***	0.26 ***	0.14 ***	0.02	0.21 ***	−0.01	−0.10 **	−0.03	−0.20 ***	1.00
12	CMPTN	2.74	0.76	0.07 *	0.04	0.23 ***	−0.08 *	0.04	−0.14 ***	0.01	0.01	−0.12 ***	−0.03	0.13 ***

* p < 0.1, ** p < 0.05, *** p < 0.01; (HM): Home, (IM): Intermediary, (HS): Host; LOCAL: Staff Localization; CD: Cultural distance; LOCEXP: Local experience; CMPTN: Competition intensity; LISTED: Listed firm; AGE: Firm age; SIZE: Firm size (Firm asset); MNF: Manufacturing; WOS: Wholly owned subsidiary.

and panel data analysis in

Table 2. Results of the GLS Regression Analysis.

DV: LOCAL	Model 1	Model 2	Model 3	Model 4	Model 5
LISTED (IM)	−0.043	−0.048	−0.054	−0.044	−0.050
	(0.061)	(0.060)	(0.059)	(0.060)	(0.059)
AGE (IM)	0.004 **	0.004 **	0.004 *	0.004 **	0.004 **
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
SIZE (IM)	0.030	0.027	0.028	0.025	0.026
	(0.020)	(0.020)	(0.020)	(0.020)	(0.020)
MNF (IM)	0.041	0.043	0.033	0.042	0.031
	(0.071)	(0.071)	(0.070)	(0.071)	(0.069)
SIZE (HS)	−0.031 *	−0.026	−0.029	−0.024	−0.027
	(0.019)	(0.019)	(0.019)	(0.019)	(0.019)
MNF (HS)	0.226 ***	0.215 ***	0.236 ***	0.213 ***	0.236 ***
	(0.075)	(0.075)	(0.075)	(0.075)	(0.074)
WOS (HS)	−0.077	−0.069	−0.071	−0.071	−0.073
	(0.050)	(0.050)	(0.049)	(0.050)	(0.049)
CD (HM-IM)	0.001	0.000	0.001	−0.000	0.000
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
CD (HM-HS)		0.002	0.001	0.002	0.001
		(0.001)	(0.002)	(0.002)	(0.002)
LOCEXP			0.010 *		0.010 *
			(0.005)		(0.005)
CD (HM-HS) * LOCEXP			0.037 **		0.042 ***
			(0.015)		(0.015)
CMPTN				−0.002	−0.003
				(0.020)	(0.020)
CD (HM-HS) * CMPTN				−0.036 **	−0.042 **
				(0.017)	(0.017)
Constant	0.128	0.046	0.045	0.062	0.065
	(0.475)	(0.478)	(0.470)	(0.489)	(0.479)
N Chi2	520	520	520	520	520
	23.39861	24.96160	34.43298	29.64265	41.08551

* p < 0.1, ** p < 0.05, *** p < 0.01; Standard error in parentheses.

.

Panel data for various objects have been researched at several time points using variables of the object (i) over time (t). Independent variables and error of basic assumptions of ordinary least squares (OLS) regression analysis violate a condition that the problem of omitted variable bias is not severe. It was, therefore, necessary to control unit-specific unobserved heterogeneity. To resolve the heterogeneity problem, fixed effects and the possibility of random effects should be determined. The fixed effect is such that the original characteristics of an object are assumed as a fixed parameter with independent variables; it then controls the variable by estimating or removing it. Because this parameter is estimated and a characteristic of an object that cannot change with time, individual property effects should be eliminated through assumptions such as within transformation, first deviation, and least squares dummy variables. However, the possibility of random effects assumes that an individual original characteristic is given randomly, meaning it is not related to the independent variable but is regarded as a random variable. Although the fixed effects focus on the variance in the panel object (i), the random effects estimate the weighted average of the between effect and the fixed effect. The weight will be determined depending on the relative size of the error term between effects and the fixed effects and on the number of observation values for each object.

Hypothesis testing to determine the model to provide estimates includes F, Hausman, and Breusch-Pegan tests. F tests determine the model between the fixed effects and the joint OLS. The Hausman test determines the fixed effects and the random effects. The Breush-Pegan test determines the random effects and the joint OLS. We selected an appropriate model between the fixed and random models through the Hausman test, considered the degree of appropriateness between this model and the joint OLS, and then determined a final model.

The Hausman test produced a p-value of 0.8313, which means we could not reject the null hypothesis that there is no significant difference between the two regression coefficients and that there is no difference between the two regression coefficients. Because estimates of the random model do not differ from the estimates of the unbiased fixed model, both models were deemed unbiased, but the random model was selected because it was more efficient [8]. In the case of the Breush-Pegan test, we rejected the null hypothesis because the p-value was 0.0000. The heterogeneity between individuals was 0, and the random model was again selected because the random model results were significant.

Since the panel data include characteristics of time series data, autocorrelation was possible. In the random-effects model determined through model tests, a generalized least squares estimation was utilized if autocorrelation occurred.

4. Results

Model 1–5 in Table 2 report the relevant regression results of the relationship between cultural distance and localization level. We tested for the presence of multicollinearity in our analyses by examining variance inflation factors (VIF). All VIFs (max: 2.50; mean: 1.47) were below 10, confirming that multicollinearity is not a severe problem in our study. The coefficient of cultural distance (Home-Host) in model 5, the overall model is not statistically significant (p = 0.05). Therefore, Hypothesis 1 is not supported. However, we considered Hypothesis 1 as supported because it was marginally significant. Because the estimated coefficient for the localization level of cultural distance (Home-Intermediary) is not statistically significant either in model 2 or model 5, Hypothesis 2 is not supported. Because the cultural distance is fixed as the difference value between countries, we expect to find no variance for each year on panel data with time-series data. The overall estimated coefficient in the model could have less statistical significance compared with cross-sectional data. In addition, because this study fixed the intermediating and local countries like Korea and China, respectively, the variance with the home countries also represented a certain value, and the coefficient in the model had less statistical significance.

Next, model 3 shows the moderating effects of the local subsidiary’s experience, as measured by the investment period, are statistically significant at a 0.05 level, with an estimated coefficient of 0.037. This influences the relationship between cultural distance (Home-Host) and localization level and supports the moderating effect of hypothesis 3: when the cultural distance between home and the local country is long, the level of localization will increase. Subsidiaries in the mediating country can control this relationship depending on their experience operating in the local country. As it mediates the control to a positive direction, the longer cultural distance becomes and the more experience subsidiaries have, the greater the localization level becomes.

In model 4, the intensity of competition is statistically significant since its mediating effect on the localization level has a mediated coefficient value of −0.036 at the 0.05 level. Therefore, Hypothesis 4 was supported. This means that the basic model—the longer the cultural distance between home and local countries, the greater the localization level—would change depending on competitive intensity. Therefore, the mediating effects were controlled in a negative direction: if the cultural distance is great, the localization level increases, but if competition intensifies, the level of localization decreases.

Figure 3 suggests that the joint effect between longer cultural distance (Home–Host) and local subsidiary’s experience makes a firm more likely to be localized. This result supports Hypothesis 3. Similarly, Figure 4 demonstrates that in a local country with a long cultural distance (Home-Host), the more intense the competition, the lower the localization level of firms. This result supports Hypothesis 4.

5. Discussion

5.1. Implications

Based on prior research, the more heterogeneous two countries become due to long cultural distances, the greater the localization level. We reviewed existing theories on headquarters-subsidiary-subsidiary investment structure and determined that this theory requires revision. Although we did not verify new determinants of localization, this paper is the first to develop and explain a new and expanded investment structure for headquarters-subsidiary-subsidiary relationships by utilizing variables that prior studies on multinational companies had used. We provide a theoretical contribution that expands our understanding of the effects of cultural distance on multinational companies by applying a new investment structure. In addition, we suggest the relationship between an organization and its environment differs according to local conditions [64]. By integrating international management strategies [20,21,22,23] and organizational learning [65], we provide an explanation for behavior associated with dual pressures involved in headquarters-subsidiary-subsidiary investment structures.

Our research of the behavior of multinational companies concerning actual headquarters-subsidiary-subsidiary investment structures suggests practical implications. Whether the investment structure covered in this study involved independent subsidiaries or bypass-investment strategies made at headquarters remains unknown, the cultural distance between home and local countries still affected the localization in the headquarters-subsidiary-subsidiary investment structure. Regardless of whether a subsidiary in the intermediating country entered a foreign nation or headquarters used a bypass investment strategy through a subsidiary in the mediating country, the effects of headquarters were at least partially maintained.

We verified the effects of a local subsidiary’s experiences by controlling investment types. Prior research considered entry mode as a determinant of localization. Still, we regarded it as a control variable because a joint venture or syndicated investment allows a foreign company to adapt more rapidly to local conditions and acquire different qualitative and quantitative local experiences over the same business period (1 year), compared with the single-investment strategy. We found that the subsidiary’s experience further expanded localization despite this control, contrary to prior findings. This means that, when entering a foreign market with a considerable cultural distance, it is important to hire local managerial staff and ensure local autonomy, rather than embrace a global integration strategy over time and encourage subsidiaries to have independent competency. MNEs should design localization strategies by taking cultural differences that characterize subsidiaries located in different contexts into account. Our results further support the role of staff management in reducing internal barriers concerning the nature of work within MNEs. Taking MNEs entering the Asian market as an example, the performance of American companies is generally better than that of European companies, which is directly proportional to the number of local talents. For instance, VIACOM, one of the three big media giants in the world, has introduced MTV programs and consumption patterns to China successfully because of its employment of Li Yifei, who is familiar with China’s market environment, policies and regulations, and is well versed in western thinking and corporate culture.

Finally, considering the aspect of local competition, we verified endogenous characteristics that originate in specific companies based on national factors because the cultural distance is a variable at the national level. However, it has a structure that companies could not know about the external industrious environment where companies operate the business. Because it is necessary to consider the liability of foreignness (in the form of cultural distance) and the liability of foreignness at the industrial level, there is room to determine if localization can increase when the liability of foreignness increases considering competition intensity. According to prior research, the liability of foreignness increases because of the lack of local competence, indicating that localization should increase because local competition becomes more critical as it intensifies. However, local competition can lead to price competition and increase marketing and promotional costs. From the perspective of the competitive superiority of a company, MNEs that lack local capabilities may have successful business models and other advantages, such as advanced management systems and extensive human resources, compared with local companies. Therefore, multinationals can introduce new competitive methods that utilize the management system of headquarters, including innovative marketing strategies, efficient production management, and the dispatching of skilled expatriates familiar with strategies of headquarters to create differentiated competitive structures.

5.2. Limitations

This study has several limitations. First, the measurement of cultural distance can be problematic. The effects of cultural distance on business operations or performance can differ depending on the local situation, and cultural differences may act as a liability of foreignness. As the world becomes globalized and information technology spreads, people experience reduced levels of heterogeneity, reducing the value of time-invariant variables according to scoring at the national level. Because cultural distance is a time-invariant variable, examining only two cultural distances acquires a high priority. Because this study dealt with relationships between three countries, three variables on cultural distance should have been derived. Still, the differences between intermediating and local countries were excluded due to multicollinearity problems with the other variables. Further research on panel data analysis that considers all three cultural distances using variables that change every year is necessary.

Second, whether the empirical capability of subsidiaries is based on the competence of headquarters or established by subsidiaries, which headquarters cannot have, remains unknown. Although we controlled for a single investment type to calculate the ability of subsidiaries from the time of investment, further research is needed to reveal when the knowledge transfer occurs from headquarters or subsidiaries with independent capabilities. This could be applied to all empirical research on the abilities of subsidiaries of MNEs. The international management strategies and organizational learning of multinational companies also warrant further study.

Third, because this study treated the localization level as a dependent variable, we only observed behaviors of multinational companies at the system level. We did not research performance and therefore could not determine how the localization level can affect the survival of actual companies. Because we utilized panel rather than cross-sectional data, reverse causality cannot be ruled out. Because a company’s performance through localization is the final concern of multinational companies, further empirical analysis on the effects of localization on the performance of a company in the headquarters-subsidiary-subsidiary structure would have theoretical value.

Fourth, considering every intermediating country in our study was Korea, and every local country was China, some potentially influential factors in the research model may have been overestimated or underestimated. However, this limitation is inherent to any study of theoretical and practical issues in the investment structure of headquarters-subsidiary-subsidiary relationships.

5.3. Future Research

Future studies may identify other factors and conditions to advance research on the relationship between MNEs’ localization strategy and cultural distance. First, since the structure of MNE investment (i.e., the percentage of the parent MNE’s ownership in the focal investment) affects MNE expatriate staffing strategy, future studies are encouraged to collect and test parent MNE ownership. Second, our sample includes Chinese (i.e., host) and Korean (i.e., intermediary) subsidiaries headquartered in other countries, such as Japan. However, it is still unclear whether this model can be applied to the influence of the cultural distance between host and intermediary countries on staffing localization. Thus, we call for future studies to include the cultural distance between host and intermediary countries. Third, other mechanisms in addition to cultural distance may correspond to the staffing localization. Future research should consider whether different types of distance, such as institutional distance, influence MNE staffing localization.