Requirements, Principles, and Performance of Corporate Federalism: A Case of MNC-SME Alliance

Abstract

This study aims to define and propose a corporate federalism model. In addition, it empirically examines the relationships between the requirements, principles, and performance of corporate federalism, as applied to a strategic alliance among a multinational corporation (MNC) and small- and medium-sized enterprises (SMEs). The study uses survey questionnaires to gather information. The data were collected from associates of 171 SMEs in a strategic alliance with a designated MNC. Structural equation modeling was used to analyze data in order to explore proposed relationships. The findings of the study suggest that corporate federalism creates commitment toward the alliance among participants. Among the requirements of corporate federalism, interpersonal relationships represent the most significant factor for the successful practice of corporate federalism in the given sample. Based on the findings, theoretical and practical implications for implementing corporate federalism in strategic alliances are discussed.

1. Introduction

In the last few decades, there was a spate of strategic alliances in the form of business collaboration among firms. However, not all alliances are successful. The failure rate of strategic alliances is reported to be around 50–60% [1], much higher than that of internal ventures or corporate buyouts. When alliances fail, they also lead to a loss of potential revenues, uncompensated transfers of information and technology, operational difficulties, loss of proprietary information and reputation, and much more [2,3]. Hence, it is especially important to maintain the sustainability of strategic alliances.

Existing studies typically suggest pre-formation efforts, such as appropriate partner selection [4], governance structural design [5], or contract negotiation [6] to maintain the sustainability of strategic alliances. Although these pre-formation efforts are important, a post-formation mechanism is also needed to create a sustainable alliance.

One group of researchers recommended corporate federalism as the post-formation mechanism to achieve continuous commitment from the participating organizations within strategic alliances [7,8,9,10,11]. Corporate federalism applies the functions of political federalism to private sector organizations [7]. The three functions of political federalism applied to corporate federalism are the executive function, the judicial function, and the legislative function [10]. Thus, corporate federalism offers management a framework for a strategic alliance, as the shared power and authority between the central organization and their various operating units are regulated through these three functions [11,12]. Moreover, the corporate federalism framework resolves the conflict between the goals of each participating organization and of the strategic alliance as a whole [10,13].

Corporate federalism can be used to mitigate opportunistic behaviors and increase commitment within a strategic alliance by enforcing principles of subsidiarity, interdependence, and coordinated control among the lead and partner organizations [8,10]. Certain criteria must be met for these principles to be effective within strategic alliances. Strategic alliances are often formed by organizations across industry lines, national borders, and sizes [7,11] (p. 456). To manage this diversity in strategic alliances, researchers emphasize the need to integrate these organizations through shared culture, vision, and objectives [10]. Other than these shared values, corporate federalism in strategic alliances also requires strong interpersonal relationships among the members of participating organizations, as well as responsibility and accountability from each organization and person within, to compensate for the lack of formal authority over different organizations [11] (pp. 442–446). Once a strategic alliance meets these requirements, the principles of corporate federalism can be effectively applied throughout the alliance, and each participating organization can show increased commitment toward the alliance.

Based on these key concepts, this study aims to propose and validate a customized corporate federalism model for strategic alliances. The proposed model firstly evaluates the extent to which participating organizations fulfill the requirements of corporate federalism in strategic alliances. Then, the model examines the relationships between the levels of requirements fulfillment to the levels of execution of the principles of corporate federalism. Finally, the model evaluates the performance of corporate federalism in terms of the participating organizations’ increased commitment toward the goals of the alliance. Based on the test results, this study proposes measures for the leaders of the alliances to successfully implement corporate federalism within their knowledge-based networks.

2. Literature Review

2.1. Principles of Corporate Federalism

Like political federalism, corporate federalism refers to the dynamics of shared power and authority among central organizations and the various operating units using three functions: executive, judicial, and legislative [7,10]. In other words, corporate federalism applies the functions of political federalism to private sector organizations. Handy [8] explained the conversion of the three functions of political federalism to the principles of corporate federalism. The executive function, including reservation of powers, is converted into the principle of subsidiarity. The judicial function is converted into the principle of interdependence. Finally, the legislative or monitoring/controlling function is converted into the principle of coordinated controls. The following sub-sections describe how the checks and balances and separation of powers are achieved within the private sector using these three principles.

Principle of subsidiarity. The principle of subsidiarity assumes that matters ought to be decided at the lowest level of competence [12]. Under this principle, each individual or participant is assumed to have the autonomy to make decisions for which they possess the competence [10]. This concept of reverse empowerment or authority at the lower levels is also found in management by objectives (MBO) and self-control. MBO and self-control require upward communication from subordinates on decision-making for specific work assignments they are responsible for [13]. For subordinates to be empowered, they should be well versed in the overall mission and strategy of the organization [14]. MBO and self-control utilize the empowerment of lower levels to ensure their personal freedom and responsibility to the organization [13,15]. Similarly, corporate federalism offers subsidiarity to the lower levels within the organizational structure [10]. Drucker’s concept of MBO and self-control is the managerial philosophy that allows subsidiarity to work by maximizing individual freedom at the lower levels while making provision for responsibility toward the overall mission and strategy of the whole organization.

The empowerment of lower echelons is similar to the principle of subsidiarity in corporate federalism [16]. In strategic management, Hui et al. [16] operationalized empowerment as a managerial practice for providing employees with primary control, which results in an increased sense of self-control and self-efficacy. Studies consistently showed that empowerment is strongly associated with job satisfaction [17].

Principle of interdependence. Being a sub-unit of a federation seems to run counter to ensuring sovereignty for the individual units. Sovereignty, by definition, signifies independent authority. Maciariello and Linkletter [10] offered the following difference between a federation and a confederation to answer the question of divided sovereignty within a federation: “A federation is different from a confederation, where the individual states yield no sovereignty to the center and try to need nothing from their neighbors” [10] (p. 644).

In a federation, individual units enjoy power and authority to govern, but this is coupled with the responsibility to accomplish the goals and objectives of the organization [18] (pp. 3–9). A federation is not simply a loose group of independent states [10]. Each state in a federation has clear purposes critical to itself (normally expressed in a charter) and yields certain functions to the federation as a whole and shares certain resources with the units within the federation. In other words, there is interdependence among the entities within a federation, and this interdependence is the glue that holds the components together [8].

In international business and strategic management, interdependence is often discussed as an important factor for organizations working together. Spekman et al. [19] reported that interdependence within an alliance reduces participating organizations’ opportunistic behaviors since the maltreatment of other organizations would have negative consequences on the entire alliance. Other studies (e.g., Reference [20]) reported that interdependence among participating organizations leads to overall success of the alliance since interdependence contributes to increased trust with other organizations within the alliance. Trust or reliance on each other is important for the performance of the entire alliance, as an organization’s behavior is dependent on the predictability of behaviors of alliance partners [21].

Principle of coordinated controls. The final principle of corporate federalism that Maciariello and Linkletter [10] reported is the importance of implementing coordinated controls. Controls refer to measurements or information [11] (p. 321). Large quantities of data do not make much sense if there is no process to measure and draw information from them. Measurements create accountability for performance of the leaders managing participating organizations. The ability to objectively measure their own performance makes the leaders of sub-units independent and yet responsible to the center. Such an idea of objective evaluation of efforts and performance can be effectively maintained by employing new technologies such as blockchain-based information systems [22]. Previous studies provided various types of measurements for different levels of performance—financial measurements, such as profitability, growth, and cost position [23], and operational performance as measured by stability, longevity, and survival of the alliance [24]. As an overall measure of the success of strategic alliances, the satisfaction of the participating organizations with the alliance is an important performance measure [25].

Corporate federalism can be practiced within a group of entities within the same organization or as a system of organizations. In either case, effective implementation of the three principles is necessary for the success of participating organizations. This study proposes antecedent requirements from the literature on strategic management to implement these principles of corporate federalism in strategic alliances.

2.2. Requirements of Corporate Federalism

With a rapidly changing business environment, firms form strategic alliances to maintain their competitiveness in the market. The boom in strategic alliances raised the interest of top executives and other board members in terms of sustainability of firms. Studies investigated the antecedents of corporate federalism and found that factors such as cultural integration, shared vision, objective specificity, interpersonal relationship, and relationship responsibility [10] can motivate corporate federalism.

Maciariello and Linkletter [10] suggested the need for integrating the diversity of cultures as one of the requirements of corporate federalism in strategic alliances. A strategic alliance is a systems organization consisting of various independent organizations [26]. The systems organization can be formed by organizations in the same industry or different industries, can vary in size, and can even be derived from different sectors of society [11] (p. 456). Furthermore, strategic alliances across national borders are also common [27]. Given the diversity among participating organizations, cultures within a strategic alliance are bound to be diverse. Cultural diversity refers to the variety in corporate, industry, sectoral, and national cultures (e.g., Reference [28]). Some studies pointed out the negative outcomes of cultural diversity on the performance of strategic alliances (e.g., Reference [29]). Many researchers are currently engaged in studies on cultural integration or cultural assimilation among different organizations that work together. This study reviews cross-border cultural integration in strategic alliances.

A strategic alliance is also a relational device that binds different organizations in a common group; thus, participating organizations tend to have different expectations from the strategic alliance [11] (p. 457). Even in MBO and self-control, Drucker and Maciariello [11] recommended that managers should have a thorough understanding of the mission and strategy (p. 246). For a strategic alliance, which integrates various organizations of different sizes from different industries, sectors, and countries, aligning the vision, purpose, mission, strategy, and goals throughout the participating organizations is an important step [30]. Tsai and Ghoshal [30] defined shared vision as “the collective goals and aspirations of participating organizations inside the strategic alliance” (p. 467). Maciariello and Linkletter [10] argued that the alliance must have a vision that is shared by all participating organizations at all levels for the effective implementation of corporate federalism. Having a common vision for the organization can be referred to as formulating a valid theory of business [11]. The theory of business is applicable to all organizations, as this is how an organization intends to create value for its customers [10].

As one of the requirements for successful corporate federalism in the strategic alliance, Maciariello and Linkletter [10] suggested converting the vision into clear objectives for each participating organization. According to the Drucker Management System [10], the vision (the theory of business) is converted into specific objectives in MBO, which are then converted into work assignments for individuals within the organizational structure [11] (p. 104).

In strategic alliances, employees from different organizations report to superiors in other organizations. Thus, the strategic alliance has limited formal authority by itself [11]. Therefore, lack of formal authority must be replaced by strong interpersonal relationships [10]. While Maciariello and Linkletter [10] argued that a strong interpersonal relationship is one of the essential requirements for corporate federalism to work in strategic alliances, they also specified the explicit characteristics of a strong interpersonal relationship that would make it constructive. Favoritism, corruption, and policy violations are some of the negative outcomes of strong interpersonal relationships among participants of strategic alliances that should be guarded against [31]. Constructive interpersonal relationships among the participants of strategic alliances are important for successful outcomes.

In a strategic alliance, authority and responsibility are loosely defined. Thus, each participating organization must voluntarily assume responsibility for the success of the alliance [10]. Strategic alliances are made effective through other organizations. Therefore, a successful alliance, requires taking relationship responsibility [11] (p. 493).

2.3. Relationship between Requirements of Corporate Federalism and Corporate Federalism

Cultural integration and corporate federalism. One of the main purposes of forming strategic alliances is to acquire complementary knowledge, which increases innovation capability and leads to higher productivity [32]. With increased capabilities from collaboration, each participating organization can perform more functions than it could on its own. In other words, effective knowledge transfer within an alliance can offer subsidiarity for the participating organizations.

Many studies reported that, when the cultural distance between two parties is shorter, cultural blending becomes easier [33]. This helps develop common values for the alliance and maintains or increases trust within the alliance [34]. In addition, Chua et al. [35] reported high interdependence within an alliance when trust is high. Therefore, when there is cultural integration in a strategic alliance, participating organizations trust each other more and, consequently, increase interdependence on each other.

Zaheer et al. [36] found that organizations forming strategic alliances with other organizations from similar cultural backgrounds share lower information asymmetry than those from different cultural backgrounds. Tung and Verbeke [37] reported that participating organizations facing lower information asymmetries engage in more effective control functions within the alliance. Therefore, in an international strategic alliance, participating organizations with a high cultural integration would experience low information asymmetries and, consequently, have more effective control functions. Thus, we propose the following hypotheses:

Hypothesis 1 (H1).

Within a strategic alliance, cultural integration and subsidiarity are positively associated.

Hypothesis 2 (H2).

Within a strategic alliance, cultural integration and interdependence are positively associated.

Hypothesis 3 (H3).

Within a strategic alliance, cultural integration and coordinated controls are positively associated.

Shared vision and corporate federalism. The vision shapes each organization’s behavior, dictates its decisions on what to do and what not to do, and defines what the organization considers meaningful results [11] (p. 85). In other words, understanding the vision of the entire alliance allows each organization with respective competencies to visualize the possible contributions and assistance it can provide to other organizations in achieving the common goal. Therefore, a shared vision within a strategic alliance can facilitate the principle of subsidiarity among participating organizations.

An alliance that shares a common vision maintains a system of effective knowledge transfer. Each organization benefits from knowledge sharing, which can increase interdependence among the organizations. Therefore, when participating organizations can share the vision of the whole alliance, the alliance can have an effective knowledge transfer system, which is followed by increased interdependence among participating organizations.

Vision sharing is the process of creating common grounds in a strategic alliance [30]. When each organization within an alliance shares the same perspective, they will evaluate performance using the same measurement standards [38]. In other words, with common measurements within an alliance, the participating organizations will coordinate controls more effectively. These points are reflected in the following hypotheses:

Hypothesis 4 (H4).

Within a strategic alliance, shared vision and subsidiarity are positively associated.

Hypothesis 5 (H5).

Within a strategic alliance, shared vision and interdependence are positively associated.

Hypothesis 6 (H6).

Within a strategic alliance, shared vision and coordinated controls are positively associated.

Objective specificity and corporate federalism. The MBO methodology requires participants to communicate upward, downward, and sideways to set and accomplish objectives [10]. MBO also relies on participants’ self-control to align individual needs with the organizational goals [39]. An organization’s self-control is an important precursor [39] to upward communication from participating organizations to the headquarters. Relying on the self-control of each participating organization for setting specific objectives in a strategic alliance maintains autonomy or individuality in decision-making and operational processes for each sub-unit. Therefore, setting specific objectives can support each organization’s subsidiarity in a strategic alliance.

Specific objectives for each organization are set to meet performance expectations for the whole alliance [40]. They also become the basis for organizing and allocating resources of participating organizations and are a prerequisite for work assignments [10]. The clarification of work assignments and organizational structure through objective specificity will allow participating organizations to determine which other organizations they can depend upon [40]. Therefore, objective specificity of participating organizations would help increase interdependence within the alliance.

When organizations have clear and specific objectives, they can convert their objectives into work assignments according to these key areas [10]. In other words, the measurement or information pertaining to their work assignments can be explicitly represented by the participating organizations’ specific objectives for evaluation [11] (p. 266). Such measurable information is an important prerequisite of what Handy [8] called the legislative function of control in corporate federalism. Thus, we present the following hypotheses:

Hypothesis 7 (H7).

Within a strategic alliance, objective specificity and subsidiarity are positively associated.

Hypothesis 8 (H8).

Within a strategic alliance, objective specificity and interdependence are positively associated.

Hypothesis 9 (H9).

Within a strategic alliance, objective specificity and coordinated control are positively associated.

Interpersonal relationship and corporate federalism. Kelley [41] (p. 95) identified responsiveness, trust, and attribution as the attitudes arising from positive interpersonal relationships within a group. These attitudinal traits of positive interpersonal relationship in personal relationship theory can be extended to the constructive interpersonal relationship among participating organizations of successful strategic alliances [42]. Firstly, local partners’ independence and flexibility because of responsiveness within strategic alliances signifies their subsidiarity. In other words, a constructive interpersonal relationship within a strategic alliance strengthens foreign headquarters’ responsiveness, which then maintains the local partners’ subsidiarity.

Secondly, trust helps defuse conflicts in strategic alliances, as argued by Nooteboom et al. [43]. If an organization encounters unexpected behaviors from its partner organizations that could be interpreted as good or bad, trust reduces the likelihood of a negative interpretation. These traits of trust are dominant signals that can lead to constructive interdependence in strategic alliances [43]. Therefore, a constructive interpersonal relationship can lead to trust in the strategic alliance and, consequently, to interdependence among participating organizations.

Thirdly, the concept of attribution can be extended to strategic alliances [43]. When managers of different organizations in an alliance have constructive interpersonal relationships, they can have attributions toward others that allow them to evaluate each other’s actions. Such evaluations based on attributions (measurement or information) can be interpreted as the controls for monitoring each organization in a strategic alliance. Therefore, a constructive interpersonal relationship within a strategic alliance leads to the maintenance of positive control functions among participating organizations. Thus, we propose the following hypotheses:

Hypothesis 10 (H10).

Within a strategic alliance, interpersonal relationship and subsidiarity are positively associated.

Hypothesis 11 (H11).

Within a strategic alliance, interpersonal relationship and interdependence are positively associated.

Hypothesis 12 (H12).

Within a strategic alliance, interpersonal relationship and coordinated controls are positively associated.

Relationship responsibility and corporate federalism. Relationship responsibility assumes the importance of individuality of other organizations within the alliance [11] (p. 493). Each participating organization has its own strengths, ways to get things done, and values. Each organization works in its own way. By assuming relationship responsibility within a strategic alliance, the loosely connected organizations can not only hold together in an alliance but also maintain their individuality. By keeping the individuality of each organization intact, things can be done at the local partners’ level without much monitoring and intervention from foreign headquarters. In other words, relationship responsibility leads to enhanced subsidiarity by allowing power and authority to function at the appropriate level in each organization. Furthermore, by understanding the strengths of other organizations, each of them can depend on others for the competencies they lack.

Another dimension of relationship responsibility is communication. Just as relationship responsibility asks for an acknowledgment of others’ functional information, such as strengths, performance modes, and values, it also involves effectively communicating an organization’s own functional information to others [11] (p. 495). This functional information becomes a benchmark for participating organizations to evaluate each other’s performance. Therefore, each organization’s assumption of relationship responsibility can lead to more effective coordinated controls. Thus, we propose the following hypotheses:

Hypothesis 13 (H13).

Within a strategic alliance, relationship responsibility and subsidiarity are positively associated.

Hypothesis 14 (H14).

Within a strategic alliance, relationship responsibility and interdependence are positively associated.

Hypothesis 15 (H15).

Within a strategic alliance, relationship responsibility and coordinated controls are positively associated.

2.4. Performance of Corporate Federalism

Existing studies of strategic alliances report commitment toward the alliance from the participating organizations as the major success factor (e.g., Reference [44]). This study seeks to evaluate if the successful maintenance of the principles of corporate federalism within a strategic alliance is related to an increase in commitments from the participating organizations.

Firstly, subsidiarity, which is reverse empowerment [10], sustains the individuality of each organization within the strategic alliance. By sustaining subsidiarity within the alliance, the goals of each participating organization can be protected without sacrificing the alliance goals. This reduces the risk of a participating organization being unfairly exploited for its participation in the alliance and, consequently, increases their commitment toward the alliance.

Secondly, when there is effective interdependence, participating organizations tend to stick (commit) to each other [10]. Das and Teng [34] argued that interdependence between alliance partners is crucial for the alliance performance as it leads to increased commitment. Interdependence also leads to successful strategic alliances by achieving benefits for individual organizations that can only be attained within the alliance by a high commitment from each of the participating organizations [45,46].

Thirdly, corporate federalism can be achieved in strategic alliances by using constructive coordinated controls [10]. Constructive controls create accountability for performance of the leaders in charge of the whole alliance, as well as of the individual units [10]. The leaders’ accountability adds procedural justice and fairness in the performance of the alliance [44]. Luo [47] concluded that fairness in work procedures has a strong effect on individuals’ attitudes toward institutions and commitment to organizations. Thus, we propose the following hypotheses:

Hypothesis 16 (H16).

Within a strategic alliance, subsidiarity and commitment are positively associated.

Hypothesis 17 (H17).

Within a strategic alliance, interdependence and commitment are positively associated.

Hypothesis 18 (H18).

Within a strategic alliance, coordinated controls and commitment are positively associated.

3. Method

3.1. Research Model

The model of corporate federalism in strategic alliances is shown in Figure 1, which graphically summarizes the theoretical logic and hypothesized relations of the constructs reviewed in this study.

3.2. Sample

The sample for this study was selected from organizations currently participating in a strategic alliance. These organizations are small- and medium-sized enterprises (SMEs) in an alliance with a multinational corporation (MNC) in high-technology industry. The MNC selected for this study is a well-known Korean electronics brand for smartphones, computers, televisions, home appliances, and other high-tech products. The SMEs that form strategic alliances with the MNC are in different industries and vary in sizes.

We chose this strategic alliance between an MNC and SMEs because it uses the structure of corporate federalism in its strategic alliances. The structure of corporate federalism is formed between a central organization and its various independent operating units [10]—a central organization and SMEs from various industries and countries, in this case. The SMEs are independent operating units of various sizes.

Additionally, one characteristic inherent to high-technology industries adds additional rigor to the proposed model. Strategic alliances in high-technology or knowledge-based industries display a higher risk in knowledge appropriation than alliances in low-technology or asset-based industries [45]. With this higher risk, high-technology organizations tend to show opportunistic attitudes and low commitment levels within strategic alliances [37]. Thus, testing the proposed model on an increase in commitment levels among participating organizations in such an opportunism-prevalent industry adds more rigor to the research and increases the importance of a valid model.

3.3. Survey Method and Data Collection

We asked chief executive officers or strategic alliance coordinators of the selected SMEs to complete an online survey. The survey was designed based on the nine constructs introduced earlier (Figure 1) using a five-point Likert scale. The initial survey was in English. Later, it was translated into Korean using parallel translation/double translation [46]. The survey was sent to 300 alliance partners located in Korea. Questionnaires were excluded because of missing data. The final sample size was 171.

3.4. Development of Measures

The questionnaire was based on nine constructs and 28 measurement items adapted from previous studies. Luo’s [42] questions measuring cultural distance within a strategic alliance were adopted to measure cultural integration. Shared vision was measured by changing “goals” to “vision” in the survey items developed by Tsai and Ghoshal [30]. Objective specificity was measured by modifying Sawyer’s [44] questions that measure goal clarity. The word “goals” in these items was changed to “objectives” and the subject of the questions were changed from “individuals” to “organizations”. To measure interpersonal relationships, three items were selected from a personal relationship scale [47]. Relationship responsibility was measured by modifying three questions on relationship responsibility: “What are his/her strengths?”; “How does he/she work and perform?”; and “What are his/her values?” [11] (pp. 493–495). Subsidiarity was measured by modifying measurement items of self-determination, a similar concept to subsidiarity, from Hui et al. [16]. To measure interdependence, measurement items developed by O’Donnell [21] to measure interdependence within an alliance were directly adopted. Coordinated controls were measured using three modified items from the measurement items for governance dimensions of strategic alliances by Schreiner et al. [25]. To measure commitment, Luo’s [42] measurements of commitment in strategic alliance are adopted with some modification.

Table 1. Constructs and measurement items. HQ—headquarter.

Constructs	Measures
Cultural integration [42]	In this alliance, local partners feel small cultural distance with the HQs. In this alliance, local partners feel small cultural distance with other local partners. In this alliance, managers from local partners feel small cultural distance with the managers from the HQs and other local partners.
Shared vision [30]	Both parties in this relationship are enthusiastic about pursuing the collective vision. Both parties are committed to improvements that may benefit the relationship as a whole, and not only the individual parties. The parties share the same ambition and vision.
Objective specificity [44]	I understand our duties and responsibilities for the objectives of the alliance. I understand how our work relates to the overall objectives of the alliance. I understand the expected results of my organizations’ work for the alliance.
Interpersonal relationship [47]	I do not believe my counterpart would cheat on me even if they could get away with it. I listen carefully to my counterpart and help him/her solve problems. I share and discuss my problems with my counterpart.
Relationship responsibility [13]	My firm understands strengths of other partners in this strategic alliance. Other organizations in this strategic alliance understand performance mode of my firm. My firm communicates values to other organizations in this strategic alliance.
Subsidiarity [16]	Local partners have autonomy in determining how they do their jobs. Local partners can decide on their own how to go about doing their work. Local partners have opportunity for independence and freedom in their jobs.
Interdependence [21]	The activities of the HQs and local partners influence the outcomes of the alliance. Each partner in this alliance depends on the HQs and other local partners. Work in one partner is connected to the work of the HQs and other local partners.
Coordinated controls [25]	For coordinating alliance-related activities, we established internal processes. We meet regularly to adapt our working procedures to the HQs and other partners. We adjusted our incentive systems to serve the goals of the alliance.
Commitment [42]	The degree of the HQs’ continuous commitment to this alliance is high. The degree of local partners’ continuous commitment to this alliance is high. The HQs and local partners take specific steps to help each other for alliance success.

summarizes the items used to measure the nine constructs listed in the proposed research model.

3.5. Control Variables

Depending on their firm sizes, strategic alliance partners may behave differently in the alliance. In the resource-based view, larger firms tend to be less proactive in sharing their resources than smaller firms, as they have the option of internally developing required resources [48]. In addition, larger and smaller firms have different purposes of forming an alliance. Larger firms typically purport to gain access to tacit knowledge of small firms, while small firms aim to benefit from financial and marketing resources of the larger firms [49]. As purposive behaviorism suggests, smaller and larger firms may behave in a different manner because of the differences in their alliance purposes [50]. Thus, we used firm size measures of the SMEs as the control variable to test if firm sizes (small-sized vs. medium-sized enterprises) influence the relationships between the principles and the performance. There is no generally accepted definition of SMEs. A common approach is the quantitative definition, which uses measures such as number of employees, sales turnover, and total assets. Based on the method suggested by Chu [51] and the United Nations Conference on Trade and Development [52], this study used the number of employees as the criterion to identify SMEs. A company with fewer than 300 full-time employees was recognized as an SME. Specifically, one with 150 full-time employees was recognized as a small-sized company; otherwise, it was considered a medium-sized company. Firm size was constructed to indicate small-sized company (1) or medium-sized company (0).

4. Results

4.1. The Sample Characteristics

A total of 171 Korean SMEs were included in this study. The respondents were executives or alliance coordinators at various levels within the SMEs. The details are tabulated in

Table 2. Organizational characteristics and rank of respondents (n = 171).

Classification		Frequency	Percentage
Alliance duration (number of years)	5 to 9	9	5%
	10 to 14	74	43%
	15 to 19	72	42%
	20 or more	16	9%
Industry sector	Electronics and components	25	15%
	General manufacturing	39	23%
	Logistics	75	44%
	Retail	32	19%
Size (number of employees)	Fewer than 150	74	44%
	150 to 299	97	56%
Respondent rank	Executive level and up	29	17%
	Director level	42	25%
	Manager level	47	27%
	Assistant manager level	53	31%

.

4.2. Findings from the Sample

Means, Standard Deviations, and Correlations of Constructs. The correlations between the five independent variables (cultural integration, shared vision, objective specificity, interpersonal relationship, and relationship responsibility) were all below 0.50. Furthermore, the correlations among the three mediating variables (subsidiarity, interdependence, and coordinated controls) were relatively low, ranging from 0.16 (p < 0.05) to 0.36 (p < 0.001). Although exploratory factor analysis was conducted on all variables, these relatively low correlations among the independent variables and mediating variables implied that multicollinearity was not serious [53] (see

Table 3. Descriptive statistics and correlation matrix (n = 171).

	1	2	3	4	5	6	7	8	9
1. Cultural integration	(0.83)
2. Shared vision	0.37 ***	(0.80)
3. Objective specificity	0.24 **	0.34 ***	(0.82)
4. Interpersonal relationship	0.29 ***	0.47 ***	0.49 ***	(0.80)
5. Relationship responsibility	0.17 *	0.49 ***	0.41 ***	0.45 ***	(0.82)
6. Subsidiarity	0.25 **	0.45 ***	0.26 **	0.51 ***	0.40 ***	(0.82)
7. Interdependence	0.26 ***	0.40 ***	0.19 *	0.40 ***	0.34 ***	0.36 ***	(0.82)
8. Coordinated controls	0.24 **	0.31 ***	0.36 ***	0.36 ***	0.26 **	0.16 *	0.26 **	(0.83)
9. Commitment	0.41 ***	0.51 ***	0.38 ***	0.61 ***	0.47 ***	0.50 ***	0.46 ***	0.39 ***	(0.80)
Mean	3.53	3.63	4.11	4.28	3.70	3.85	3.66	3.89	4.08
Standard deviation	0.86	0.84	0.64	0.58	0.63	0.64	0.76	0.63	0.72

Note. * p < 0.05, ** p < 0.01, *** p < 0.001; Diagonal entries in parentheses are Cronbach’s α values.

).

Cronbach’s α for each set was computed to examine the reliability of measures for all nine sets of multi-item scales. All the scales demonstrated acceptable reliability, with the highest Cronbach’s α observed for interpersonal relationship measures (0.86) and the lowest Cronbach’s α observed for coordinated controls measures (0.70) (see Table 3). Furthermore, composite reliability scores for all constructs (ranging from 0.72 to 0.93, as shown in

Table 4. Confirmatory factor analysis results (n = 171).

Construct	Item	Factor Loading	SE a	Standard Loading	t-Value	CR b	AVE c
Cultural integration	CI1	1.00	-	0.75	-	0.84	0.65
	CI2	1.07	0.11	0.79	9.70
	CI3	1.18	0.11	0.83	9.92
Shared vision	SV1	1.00	-	0.82	-	0.83	0.62
	SV2	0.89	0.08	0.77	10.05
	SV3	0.85	0.08	0.74	9.71
Objective specificity	OS1	1.00	-	0.61	-	0.83	0.64
	OS2	1.21	0.17	0.77	7.13
	OS3	1.12	0.15	0.75	7.10
Interpersonal relationship	IR1	1.00	-	0.74	-	0.93	0.79
	IR2	0.99	0.10	0.79	9.82
	IR3	1.10	0.10	0.79	10.43
	IR4	1.27	0.12	0.85	10.59
Relational responsibility	RR1	1.00	-	0.68	-	0.86	0.68
	RR2	1.09	0.13	0.75	8.03
	RR3	1.11	0.13	0.77	8.15
Subsidiarity	SB1	1.00	-	0.63	-	0.80	0.58
	SB2	1.21	0.17	0.71	7.00
	SB3	1.29	0.18	0.69	6.89
Interdependence	ID1	1.00	-	0.65	-	0.77	0.53
	ID2	0.93	0.14	0.65	6.70
	ID3	1.21	0.16	0.79	7.31
Coordinated controls	CC1	1.00	-	0.65	-	0.72	0.47
	CC2	0.87	0.16	0.59	5.44
	CC3	0.94	0.17	0.58	5.37
Commitment	CM1	-	0.74	-	-	0.82	0.61
	CM2	0.11	0.71	8.56	8.56
	CM3	0.12	0.72	8.71	8.71

Note. p < 0.001 for all loadings; ^a standard error, ^b composite reliability, ^c average variance extracted.

) were well above 0.70.

Validity. The validity of the nine constructs was initially tested with exploratory factor analysis (EFA) using SPSS V.20. The Kaiser–Meyer–Olkin (KMO) measure of sample adequacy was 0.87, which can be interpreted as meritorious [54], indicating that an adequate number of items were available to measure each construct. As suggested by Fabrigar et al. [55], principal component analysis was used for initial data reduction. From the 28 variables, nine factors were extracted, for which the eigenvalues ranged from 8.76 to 0.91. Then, as the latent variables were expected to be correlated based on the correlation matrix (Table 3), ProMax rotation was conducted for each factor to be defined by the subset of measured variables with higher loadings relative to the other variables [56]. ProMax rotation revealed factor loadings above 0.60 for all variables on their expected factors. The nine-factor exploratory factor analysis was conducted and loadings less than 0.40 were omitted. The highest omitted value was 0.27. Next, the convergent and discriminant validity of each measurement item was tested through confirmatory factor analysis. Average variance extracted (AVE) values were all above 0.50 except for coordinated controls, which was slightly less at 0.47 (see Table 4). Another convergent validity test was performed to ensure that all items loaded on their respective constructs were statistically significant with t-values 5.37 or greater (p < 0.01) (see Table 4). The discriminant validity of all measurement items in this study was confirmed as the lowest AVE value was 0.47 (see Table 4), the highest squared Pearson correlation coefficient was 0.37, and the highest Pearson correlation was 0.61 (see Table 3).

Model-fit. The model fit indices were examined against the common goodness-of-fit benchmarks. Gefen [57] suggested there is a reasonable model fit if a χ² to degrees of freedom (df) ratio is less than 3:1. Hair et al. [58] recommended GFI(Goodness of Fit Index), NFI(Normed Fit Index), CFI(Comparative Fit Index), and IFI(Incremental Fit Index) values to be greater than or equal to 0.90, AGFI(Adjusted Goodness of Fit Index) value to be greater than or equal to 0.80, and RMSEA(Root Mean Squared Error of Approximation) value to be less than or equal to 0.08 for a research model to have an acceptable fit. The model fit indices depicted from the hypothesized model with additional correlations between exogenous latent factors were χ²:df = 437.40:340 = 1.29:1, GFI = 0.86, AGFI = 0.82, NFI = 0.81, IFI = 0.95, CFI = 0.95, and RMSEA = 0.04. The RMSEA favored parsimony as it was sensitive to the number of estimated parameters in the research model [59]. The RMSEA indicated model parsimony as 0.04 was far below the cut-off point of 0.08 [59]. Also, as only GFI and NFI were slightly below standard, the hypothesized model could be considered adequate to represent the phenomenon in this study. The hypothesized model and the path coefficients are displayed in Figure 2.

Hypothesis test results. This section presents the results of hypotheses tests among the Korean SMEs and a Korean MNC. Although in reverse order of the suggested hypotheses, it is conceptually clearer to explain the relationships between variables from dependent to mediating variables and mediating to independent variables.

H18, which suggests a positive relationship between coordinated controls and commitment, was supported in the proposed research model with a coefficient of 0.35 (p < 0.001). H17, which suggests a positive association between interdependence and commitment, was also supported with a coefficient of 0.32 (p < 0.01). H16, which suggests a positive link between subsidiarity and commitment, was the strongest relationship in this model with a coefficient of 0.68 (p < 0.001). Among the suggested positive links between relationship responsibility and the three principles of corporate federalism, the association between relationship responsibility and subsidiarity (H13) was the only statistically significant link with a coefficient of 0.21 (p < 0.05). Among the requirements of strategic alliance, interpersonal relationship had the highest significant relationships with the principles of corporate federalism. H11, the positive association between interpersonal relationship and interdependence, was supported with a coefficient of 0.48 (p < 0.01). The strongest relationship between the requirements and principles was found in H10, with a coefficient of 0.50 (p < 0.001). These hypotheses suggested positive associations between interpersonal relationship and interdependence and between interpersonal relationship and subsidiarity. The relative importance of interpersonal relationship in the Korean sample can also be witnessed from the highest mean value of 4.28 (see Table 3). Objective specificity had a significant relationship with coordinated controls (H9), with a coefficient of 0.40 (p < 0.05). It is worth noting that, for coordinated controls, objective specificity had the only supported relationship between the requirements and coordinated controls. The strong association between coordinated controls and objective specificity could also be found from the high Pearson correlation (in Table 3) between the two constructs at 0.36 (p < 0.01).

One unexpected result was found in H8 between objective specificity and interdependence. Unlike the hypothesis, which positively associates objective specificity with interdependence, they were negatively correlated with a path coefficient of −0.30 (p < 0.10). The Pearson correlation between the two constructs in Table 3 was at 0.19 (p < 0.05), which was the lowest among the correlations between interdependence and other requirement constructs. H5, the positive link between shared vision and interdependence, was supported with a coefficient of 0.18 (p < 0.1). The hypothesis test results are organized in

Table 5. Hypothesis test results.

Path		Coefficient	Result
H1	Cultural integration → subsidiarity	0.05	Not supported
H2	Cultural integration → interdependence	0.15 *	Supported
H3	Cultural integration → coordinated controls	0.12 *	Supported
H4	Shared vision → subsidiarity	0.11	Not supported
H5	Shared vision → interdependence	0.18 **	Supported
H6	Shared vision → coordinated controls	0.06	Not supported
H7	Objective specificity → subsidiarity	−0.17	Not supported
H8	Objective specificity → interdependence	−0.30 *	Supported
H9	Objective specificity → coordinated controls	0.40 **	Supported
H10	Interpersonal relationship → subsidiarity	0.50 **	Supported
H11	Interpersonal relationship → interdependence	0.48 **	Supported
H12	Interpersonal relationship → coordinated controls	0.25	Not supported
H13	Relationship responsibility → subsidiarity	0.21 *	Supported
H14	Relationship responsibility → interdependence	0.24	Not supported
H15	Relationship responsibility → coordinated controls	0.06	Supported
H16	Subsidiarity → commitment	0.68 ***	Supported
H17	Interdependence → commitment	0.32 ***	Supported
H18	Coordinated controls → commitment	0.35 ***	Supported
Control	Firm size → subsidiarity	0.00	Not supported
Control	Firm size → interdependence	0.00	Not supported
Control	Firm size → coordinated controls	0.00	Not supported
Control	Firm size → commitment	0.00	Not supported

Note: * 0.05 significance; ** 0.01 significance; *** 0.001 significance.

.

5. Conclusions

Drucker and Maciariello [11] (p. 37) asserted that open access to knowledge allows individuals to attain success and individual accomplishments lead them to fulfill their responsibilities toward society. The idea of corporate federalism and the results of this study offer a firm grounding for action planning to resolve the conflict between freedom and responsibility.

5.1. Theoretical Implications

Firstly, for the sustainability of a strategic alliance, a customized model for corporate federalism was developed and tested. A research model was developed incorporating the requirements for the practice of corporate federalism in strategic alliances, the fundamental factors of corporate federalism, and the performance of corporate federalism in strategic alliances. Five requirements for corporate federalism in strategic alliances were then reinterpreted to relate to previous studies on strategic management and international business. The independence of these seemingly overlapping concepts was confirmed through statistical analyses.

Secondly, corporate federalism was explained by the three principles of subsidiarity, interdependence, and coordinated controls. The current study theoretically supports corporate federalism by interpreting the three principles through corresponding ideas from the literature on international business and strategic management. Moreover, the statistical soundness of these concepts was confirmed in this study.

Thirdly, the performance of corporate federalism in strategic alliances in terms of its impact on commitment from alliance partners was also considered. One of the most frequently mentioned success factors for strategic alliances is the influence of corporate federalism on commitment. This study analyzed the impact of corporate federalism on the success of strategic alliances.

Lastly, the results serve as empirical evidence in support of Drucker’s assertions that corporate federalism can be adopted from the principles of political federalism. The results also support the theories postulated by later researchers on the usefulness of corporate federalism in strategic alliances. Finally, the results also validate the utility of the customized model for corporate federalism developed in this study.

5.2. Managerial Implications

Firstly, among the principles of corporate federalism, subsidiarity was the strongest predictor of commitment in the alliance between a Korean MNC and Korean SMEs. In other words, to effectively increase the level of commitment within the alliance, leaders of strategic alliances in Korea should allocate their resources to promote subsidiarity of alliance partners.

Secondly, one issue that could be raised by the leaders of both the central organization and the sub-units of the alliance is how to promote subsidiarity among partners within a strategic alliance. Based on the empirical results, the importance of interpersonal relationships in the Korean business environment was consistent with previous studies on the importance of Jeong in Korean business behavior [60]. There could be other reasons for the importance of interpersonal relationship in the Korean sample, such as resource sharing style and the overall purpose of alliance, which are known to be impacted by the firm size. However, firm size as the control variable in this study did not have a significant impact on the relationships. Thus, we develop the discussion of this managerial implication based on cultural issues. Yang [60] (p. 286) defines Jeong as “sharing personalized emotions or feelings of de-differentiating the self and the other”. However, some studies discussed the negative aspects of having strong interpersonal (informal) relationships among the participants of strategic alliances because such relationships could lead to favoritism, corruption, and policy violations [31]. Therefore, according to Drucker and Maciariello [11] (p. 23), what managers do in different countries is the same but how they do it depends on culture. Promoting interpersonal relationships may be the most effective managerial practice for promoting corporate federalism within a Korean strategic alliance, but it also requires alertness to avoid the downside of doing so. Several studies reported the ways of promoting interpersonal relationships among employees in strategic alliances. In fact, in the Korean business environment, teams or colleagues spending time together outside the work place is understood as an extension of work. Social gatherings such as dinners, drinking, and attending each other’s family events are considered critical interpersonal relationship-building occasions [60].

5.3. Limitations and Future Research

Although the data generally support the proposed model, it is necessary to note the limitations of the study. Firstly, this study examined the requirements of corporate federalism in strategic alliances that influence the practice of corporate federalism. However, all these are internal requirements and there are numerous other possible external requirements. The value of this study could be enhanced if future research could uncover the relationships between external factors such as government regulations, economic conditions, and firm types with the principles of corporate federalism.

Secondly, this study successfully explained commitment as a performance factor within corporate federalism. It would be meaningful to split the performance of corporate federalism into two extremes—positive and negative—to examine performance patterns, such as commitment and opportunism. If future research reveals the relationships between the requirements and the two extreme performance factors, the requirements to avoid or demote will be better understood, and not just the requirements to promote.

Thirdly, in this study, samples of strategic alliances in Korea were examined. For further validation of the research model, it is important to apply the model in different economic and cultural environments. Thus, it is important to consider a scoreboard for each country the study is conducted in, so that managers can establish specific objectives in the design of new strategies and these strategies can also be sustainable in the future.

Finally, the current study is a cross-sectional study, which may bare endogeneity problems among the independent variables. Future study on the same sample over time will allow a longitudinal study, which may resolve the endogeneity bias.

Despite its limitations, the strength of this research is that it was conducted in a real business context with highly relevant domain knowledge. Although caution must be exercised to avoid over-generalizing these findings, this research deepens our understanding of the practice of corporate federalism in strategic alliances. We hope this study will stimulate future researchers to employ different approaches in this area.