Systemic Analysis of Reverse Cross-Border M&A: The Heterogeneous Impacts of EMNE Network and Agency Problems Under Host Country Security Review Constraints

Abstract

Against the backdrop of geopolitical restructuring and rapid digital transformation, foreign investment security reviews in host countries have become an increasingly important institutional constraint on cross-border business activities. From a broader organizational perspective, such regulatory mechanisms shape the external environment in which emerging market multinational enterprises (EMNEs) make international investment decisions and conduct cross-border acquisitions. This paper explores the moderating effects of EMNE network and internal agency problems on the duration of reverse cross-border M&A (CBMA) under host country security reviews. Utilizing a Negative Binomial regression model, we empirically analyze 503 reverse CBMAs undertaken by Chinese firms in developed economies from 2003 to 2022. The findings reveal that host country security reviews significantly prolong M&A duration. Notably, business group affiliation weakens this positive relationship, whereas political networking strengthens it. Regarding internal governance factors, Type I agency problems reinforce the delaying effect of foreign investment security reviews, while Type II agency problems weaken it. This study provides practical implications for managers and policymakers seeking to improve firms’ responses to increasingly complex regulatory environments.

1. Introduction

The global digital transformation of business activities has profoundly reshaped the strategic behavior and international expansion patterns of multinational enterprises (MNEs). As an important component of the global economy, emerging market multinational enterprises (EMNEs), particularly Chinese firms, have increasingly adopted reverse cross-border mergers and acquisitions (CBMAs) in developed economies as an important pathway for acquiring strategic assets, technological capabilities, and international competitiveness. According to the World Investment Report 2022, global foreign direct investment increased by 64% in 2021, reaching USD 1.58 trillion, with developing economies contributing more than 25% of the total global investment flow. However, there are obvious differences in the cross-border M&A operation between EMNEs and developed-country MNEs. Data shows that the completion rate of announced cross-border M&A transactions of developed-country MNEs reaches 82%, while that of EMNEs is only 67.5% (Zhou et al., 2016), which fully reflects the institutional and structural barriers existing in the current global digital economic ecosystem [1].

Among external institutional constraints affecting cross-border investment, host country foreign investment security review mechanisms have become increasingly important. Unlike traditional tariff barriers, security reviews represent formal regulatory mechanisms through which host governments evaluate potential national security implications associated with foreign investments. Existing studies suggest that Chinese firms often face heightened scrutiny under review mechanisms such as the Committee on Foreign Investment in the United States (CFIUS), increasing uncertainty and regulatory costs during cross-border acquisitions (Chen et al., 2020; Heinemann, 2012) [2,3]. Such regulatory interventions may not only influence the likelihood of transaction completion but also affect the efficiency and duration of M&A processes.

Prior international business research suggests that firms may rely on relational resources and internal organizational characteristics to cope with external constraints. In the context of EMNE internationalization, increasing attention has been devoted to the role of informal networks in shaping overseas investment outcomes (Shen et al., 2022) [4]. As two core heterogeneous network of EMNEs, political networking and business group form differentiated resource acquisition and risk response, respectively. Political networking refers to relationships between firm executives and government-related actors, which may provide firms with policy information and access to institutional resources (Li, 2019) [5]. Business group affiliation enables firms to access internal capital, managerial experience, and information-sharing mechanisms within affiliated firms, potentially strengthening their international competitiveness (Gaur et al., 2014) [6]. Both political networking and business group have been verified as key driving factors affecting the operational process of EMNE cross-border M&A (Luo et al., 2008) [7]. However, limited research has examined whether these heterogeneous network structures influence firms’ ability to cope with host country security review requirements.

This study aims to fill the research gaps in EMNE internationalization and cross-border M&A research from two dimensions, adhering to the holistic and interdisciplinary research perspective. First, the stable operation of the EMNE cross-border M&A under external institutional constraints relies on the resource support from the home country network (Li & Chang, 2016) [8]. Based on social network theory, enterprise embedded different network can obtain heterogeneous resource endowments, thereby forming differentiated coping strategies for external regulatory constraints. However, the existing literature mostly focuses on the direct impact of EMNE network on investment decision-making, capital operation, and international performance and ignores the interaction effect between heterogeneous network structures and host country security review on M&A operational efficiency (Tan & Meyer, 2010; Castaldi et al., 2019) [9,10]. Although Shen et al. (2022) discussed the value of political and business network ties in the internationalization process of state-owned MNEs, their research focuses on the dispute resolution of overseas enterprises and does not involve the mechanism of network responding to host country investment security review constraints [4]. In high-intensity regulatory environments with complex boundaries, different network configurations will produce completely different adaptation results (Singh et al., 2018; Wu & Ang, 2020) [11,12]. Clarifying this differential mechanism is crucial for constructing a resilient EMNE cross-border digital commerce operation.

Second, this study incorporates internal corporate governance into the overall research framework of cross-border M&A and explores the moderating effect of different agency problems. Agency problems are core internal governance contradictions of modern enterprise complex management, which are divided into two heterogeneous conflicts: principal–agent conflict derived from the separation of ownership and control rights, and principal–principal conflict caused by the asymmetry of residual control rights and residual claim rights. The two types of agency problems represent differentiated stakeholder interest allocation and internal operational friction mechanisms of enterprises, which will inevitably produce heterogeneous impacts on the external adaptation process of enterprise M&A (Bhaumik et al., 2019; Zhou, 2019; Sauerwald et al., 2019; Gama & Bandeira, 2021) [13,14,15,16]. However, prior studies often treat agency problems as a single construct and provide mixed findings regarding their effects on cross-border acquisitions (Breuer et al., 2018; Datta et al., 2020) [17,18]. Distinguishing between different types of agency problems may therefore provide a more nuanced understanding of how internal governance structures shape firms’ responses to external regulatory constraints.

This paper selects 503 reverse CBMA cases of Chinese enterprises investing in developed economies from 2003 to 2022 as the research sample and conducts quantitative analysis to verify the research hypotheses. Existing studies provide important insights into the determinants of cross-border M&A outcomes; however, several limitations remain. First, prior research has primarily focused on the direct effects of host country institutions while paying limited attention to how security review mechanisms influence the operational efficiency of reverse CBMA undertaken by EMNEs. Second, although relational resources such as business groups and political networking have been widely examined in international business research, limited evidence exists regarding their distinct roles in helping firms respond to external regulatory constraints (Zhang et al., 2016) [19]. Third, previous studies often treat agency problems as a single governance construct, overlooking the possibility that different forms of agency conflict may influence firms’ responses to institutional environments in different ways. To address these gaps, this study develops an integrated analytical framework examining how host country foreign investment security reviews influence reverse CBMA and how this relationship varies across different network and governance characteristics. Accordingly, this study addresses the following research questions: How do host country foreign investment security reviews influence the duration of reverse cross-border M&A? How do external network resources, including business group and political networking, moderate the relationship between foreign investment security reviews and reverse CBMA? How do different types of agency problems moderate the relationship between foreign investment security reviews and reverse CBMA?

2. Theoretical Development and Hypotheses

Complex adaptive system theory suggests that organizational outcomes emerge from interactions among heterogeneous components operating under environmental constraints rather than from isolated factors alone. In the context of cross-border M&A, firms operate within a multilayered environment in which external institutions, relational resources, and internal governance mechanisms jointly influence strategic behavior. In this study, the systems perspective serves as an organizing framework. Specifically, we conceptualize host country foreign investment security review systems as external boundary conditions that create institutional constraints for firms’ international operations. Business groups and political networking represent firm-level relational resources that may shape adaptive responses to these constraints, while Type I and Type II agency conflicts capture internal governance conditions affecting organizational behavior. Accordingly, our analytical framework consists of three levels: external institutional conditions, firm relational structures, and internal governance mechanisms. The purpose of introducing this perspective is to provide a structured explanation of how factors operating at different levels influence reverse CBMA duration.

Foreign investment security review mechanisms have become an important institutional arrangement through which host countries regulate cross-border capital flows and strategically sensitive investments. Different from traditional tariff and trade barriers, security review belongs to the institutional intervention of complex social systems. It not only increases the transaction cost of cross-border business activities and restricts the normal iterative operation of enterprise internationalization (Graham & Marchick, 2006), but also has opaque and uncertain implementation characteristics, which easily trigger differentiated and discriminatory regulatory behaviors for foreign-funded enterprises (Heinemann, 2012) and even lead to the termination or suspension of cross-border M&A operations (Friedman et al., 2015) [3,20,21]. Moreover, prior studies suggest that the effectiveness of firms’ responses to institutional obstacles is shaped by broader governance environments, which influence how organizations perceive and react to uncertainty and external constraints (Olarewaju et al., 2025) [22]. Thus, the impact of security review mechanisms may extend beyond formal regulation and interact with firms’ relational and organizational characteristics.

As EMNEs increasingly expand internationally, firms have moved beyond relying solely on standalone internationalization strategies and instead increasingly utilize various forms of organizational and relational resources to adapt to changing investment environments (Li et al., 2019; Shen et al., 2022) [4,23]. Existing studies suggest that relational resources, including political ties, business group affiliations, and strategic partnerships, may help firms reduce information disadvantages, mitigate institutional barriers, and obtain resources necessary for international expansion (Shi et al., 2021) [24]. Among these relational mechanisms, business groups and political networking represent two important but distinct forms of organizational resources. Because they rely on different mechanisms for resource mobilization and information acquisition, they may generate different consequences for firms’ international investment behavior (Aggarwal et al., 2019; Haveman et al., 2017) [25,26].

The international business (IB) literature highlights that business groups enable firms to circumvent external market imperfections, thereby strengthening competitiveness and positively influencing FDI outcomes (Chari, 2013; Castaldi et al., 2019) [10,27]. Affiliation with a business group often correlates with greater accumulated international investment experience compared to standalone firms (Gaur et al., 2014), enhancing capabilities for global diversification and performance optimization (Kim et al., 2010) [6,28]. Conversely, political networking can furnish firms with political capital and regulatory knowledge (Parente et al., 2019), potentially reducing the incidence of international investment disputes and smoothing entry into host markets [29]. However, the efficacy of political ties is contingent upon institutional context: their marginal benefit tends to diminish in developed economies relative to developing ones (Wu & Ang, 2020) and may erode over time (Fan et al., 2013) [12,30]. Importantly, political networking operates as a double-edged sword—while beneficial in certain contexts, it may also deter performance and investment outcomes. For instance, excessive reliance on political ties can undermine the relationship between entrepreneurial orientation and startup performance (Su et al., 2015), create friction with foreign partners, and complicate cooperative negotiations during cross-border investments (Shen et al., 2022) [4,31].

Agency problems introduce another layer of complexity in the CBMA process. Where ownership and control are separated, agency problems emerge—exacerbated in reverse CBMA contexts by heightened informational asymmetry and environmental volatility (Lewis & Bozos, 2019) [32]. Agency challenges are conventionally categorized into Type I (principal–agent conflict) and Type II (principal–principal conflict) (Zhou, 2019) [14]. In Type I scenarios, divergent objectives between owners (seeking firm wealth maximization) and managers (pursuing personal utility maximization) may result in moral hazard or adverse selection, eroding shareholder value. In many emerging markets, concentrated ownership structures give rise to Type II conflicts, wherein controlling shareholders—entrusted with decision-making authority by dispersed minority investors—exploit informational asymmetries for private gain (Bao & Lewellyn, 2017) [33]. The intricacy of global CBMA operations further amplifies both agency types (Krapl, 2015) [34], inflating transaction costs and prolonging deal duration (Lewis & Bozos, 2019) [32]. Risk-management-oriented IB scholarship suggests that such distortions may incentivize managerial behaviors oriented toward rapid, scale-driven expansion—often culminating in over-payment (Donaldson, 1984), impaired post-merger performance (Arvanitis & Stucki, 2015) [35,36], and heightened regulatory scrutiny (Parvinen & Tikkanen, 2007) [37].

Given the increasing use of investment security regulations worldwide, recent research has increasingly focused on how firms respond to institutional barriers in cross-border investment contexts (Castaldi et al., 2019; Friedman et al., 2015) [10,21]. This study develops an integrated analytical framework incorporating host country institutional conditions, firm relational resources, and internal governance mechanisms to explain variation in reverse CBMA duration. Rather than focusing on isolated determinants, our framework emphasizes how factors operating at different analytical levels jointly shape firms’ cross-border acquisition outcomes.

2.1. Foreign Investment Security Review and Reverse Cross-Border M&A

Foreign investment security review mechanisms constitute an important institutional constraint affecting firms’ cross-border acquisition activities. These mechanisms involve formal procedures through which host governments evaluate whether foreign investment transactions pose potential risks to national security. First, when a foreign investment security review exists in the host country, only when the host government security review is passed can a company’s cross-border M&A proceed smoothly. Developed countries have more stringent foreign investment security reviews. In other words, the host government or officials reviews whether foreign M&A activities in their territory violate their national security issues according to more stringent legal provisions, procedures, or requirements, which makes it more challenging for firms to acquire legitimacy (Dikova et al., 2019) [38]. Host governments intervene or eliminate support for foreign firms’ M&A when they believe a firm’s cross-border M&A is not legitimate (Li et al., 2017) [39].

Second, when an acquirer is subject to a strict security review in the host country, the government pressures policymakers in the name of national economic security so that the acquirer’s total investment share in the country is reduced (Tingley et al., 2015) [40]. The security review requires the acquirer to add a large amount of supporting documents on whether foreign investment affects national security (Xie et al., 2017) [41]. A large amount of information is collated, and the lengthy review process lengthens the duration of corporate CBMA. For instance, the CNOOC (China National Offshore Oil Corporation) proposed acquisition of Unocal encountered extensive political and regulatory scrutiny in the United States despite multiple commitments designed to reduce national security concerns. The prolonged review process ultimately contributed to the withdrawal of the acquisition proposal. Such cases illustrate how security-related concerns may substantially delay acquisition completion.

Third, the vague definition of a security review itself in the host country leads to the potential requirements of security reviews that may be difficult for firms to meet when they merge and make acquisitions (Li et al., 2017) [39]. This forces firms to spend more time researching the potential requirements contained in the security review, increasing the time for firms to decipher the CBMA process and related laws and regulations, which in turn leads to a longer M&A duration. Based on the above analysis, our research argues that the duration of CBMA is prolonged when a foreign investment security review mechanism exists in the host country. In sum, we postulate the following:

H1. 

Foreign investment security reviews increase the duration of reverse cross-border M&A.

2.2. The Moderating Role of EMNE Networks

EMNE network research suggests that firms frequently rely on external relational resources to address market imperfections and environmental uncertainty. Internal firm resources alone are often insufficient for coping with institutional barriers and information asymmetry in international operations. Different forms of network relationships may therefore provide firms with distinct informational resources, legitimacy advantages, and adaptive capabilities (Castaldi et al., 2019; Wu & Ang, 2020) [10,12]. As an important relational carrier of enterprise social connection, networks link different organizational actors to form an interconnected and interactive complex system (Tichy et al., 1979) [42]. Following Shen et al. (2022), this study distinguishes between two forms of network relationships among EMNEs: business group and political networking [4].

Business group refers to cooperative arrangements among legally independent firms connected through economic and social relationships (Elango & Pattnaik, 2007) [43]. Business groups may help firms overcome external constraints through internal resource sharing, information exchange, and risk diversification. Firms affiliated with business groups can access internal knowledge regarding host country regulatory practices, policy requirements, and accumulated international investment experience (Carney et al., 2017; Purkayastha et al., 2018) [44,45].

First, business groups may provide firms with experience accumulated from prior international operations. Through information sharing among affiliated firms, group members can obtain knowledge about host country regulations, security review requirements, and practical M&A experiences, thereby improving preparation and reducing delays associated with regulatory uncertainty (Popli et al., 2017; Fan et al., 2013) [30,46]. At the same time, the business group network has a risk buffer function. Firms facing regulatory pressure may obtain financial, legal, and informational support from affiliated firms, which facilitates adaptation to host country requirements and reduces transaction uncertainty (Shi et al., 2021) [24]. Enterprises can rely on the group’s experience reserve to complete pre-review material preparation and operational scheme optimization in advance, efficiently respond to the constraints of security reviews, and shorten the M&A operational cycle. In addition, the business group has reputation spillover effects. Excellent group network operation endows member enterprises with higher credibility and market visibility (Buchuk et al., 2014) [47], which can effectively alleviate the institutional vigilance and political concerns of the host country regulatory (Chari, 2013) [27]. The high legitimacy of the business group can help member enterprises break through the outsider disadvantage, build trust connections with host country institutional systems and market subjects, reduce additional review barriers, and accelerate the operational progress of reverse CBMA. Based on the above analysis, this study proposes the following:

H2a. 

Business groups weaken the impact of host country security reviews on reverse cross-border M&A.

Political networking differs from business group because it represents firms’ relationships with governmental actors and political institutions (Parente et al., 2019; Dong et al., 2013) [29,48]. Although political connections may provide access to information and policy support, their effectiveness depends on institutional contexts and may become less beneficial in developed economies.

First, strong political networking may reduce organizational transparency. Firms relying heavily on political relationships often adopt more complex organizational arrangements and resource allocation strategies, making their operational structures more difficult for external stakeholders to evaluate (Balakrishnan et al., 2019; Kim & Zhang, 2016) [49,50]. However, the host country security review takes enterprise operational transparency as a core evaluation index. The low transparency will lead the host regulatory system to form negative institutional cognition, reduce the acceptance of M&A behaviors, and increase the difficulty of review (Li et al., 2019) [23].

Second, the over-embedding of political networks forms a closed relational circle, which hinders the iterative update of enterprise external information and operational ideas (Nahapiet & Ghoshal, 1998) [51]. Solidified political networking restricts enterprises from absorbing new market-oriented operational modes and regulatory response strategies, reduces the adaptability of enterprises to the host country’s new review rules, and cannot effectively resolve the institutional pressure brought by security reviews.

Third, high-density political networks will trigger cognitive bias of the host country’s institutional system towards enterprises. Long-term political relational embedding labels the enterprise’s M&A behavior with strong political attributes by the host regulatory system, which breaks the market-oriented operational logic of cross-border business and triggers additional security concerns (Li, 2019; Chung et al., 2015) [5,52]. Meanwhile, excessive dependence on political networking may be perceived as a non-market strategy that relies on political influence rather than market-based capabilities, potentially creating legitimacy concerns in host country environments (Choi et al., 2020) [53]. Such perceptions may weaken trust among regulators and external stakeholders and increase regulatory scrutiny during cross-border transactions. In addition, opportunistic behaviors associated with political relationships may damage firms’ reputational standing and intensify host country concerns regarding transparency and organizational intentions (El Nayal et al., 2021) [54]. These concerns may create additional barriers during the review process and prolong the duration of reverse CBMA. Taken together, political networking may strengthen the positive effect of host country security reviews on EMNE reverse CBMA. Therefore, this study proposes the following:

H2b. 

Political networking strengthens the impact of host country foreign investment security reviews on reverse CBMA.

2.3. The Moderating Role of Agency Problems

Corporate governance plays an important role in shaping firms’ responses to external institutional constraints during cross-border M&A activities. Agency conflicts may influence managerial incentives, decision-making processes, and firms’ ability to cope with regulatory requirements, thereby affecting the relationship between host country security reviews and reverse CBMA (Krapl, 2015; Bao & Lewellyn, 2017) [33,34]. Based on the classification of enterprise governance conflicts by Zhou (2019), this study distinguishes between two forms of agency problems, Type I principal–agent problem and Type II principal–principal problem, and examines their differentiated moderating effects [14].

Type I agency problem is an interest divergence between enterprise owners and managers caused by the separation of ownership and control. Managers who possess operational discretion may prioritize personal objectives over long-term firm value (El Diri et al., 2020) [55]. By contrast, Type II agency problems arise from conflicts between controlling and minority shareholders under concentrated ownership structures, where controlling shareholders may use their decision-making authority to pursue private benefits.

These two forms of agency problems reflect different governance structures and may influence firms’ responses to external regulatory requirements in different ways. Reverse CBMA is characterized by high uncertainty and involves multiple stages such as target evaluation, transaction negotiation, and post-acquisition integration (Pan et al., 2019) [56]. Successfully navigating these processes requires managerial effort and substantial investment in due diligence, information gathering, and strategic planning (Ahammad & Glaister, 2013) [57]. However, under the influence of the Type I agency problem, the evaluation of managers is dominated by short-term quantitative financial indicators, and the long-term work such as M&A risk response and regulatory adaptation cannot be effectively reflected in performance assessment (Gaspar et al., 2005; Saghi-Zedek & Tarazi, 2015) [58,59]. Driven by self-interest, managers will avoid high-cost and long-cycle M&A optimization work, lack the initiative to respond to host country security review constraints, and cannot effectively resolve external obstacles, thus intensifying the negative impact of security reviews on M&A operation.

Meanwhile, severe Type I agency problems will reduce the information transparency of enterprises. The interest divergence between managers and shareholders leads to serious internal information asymmetry problems. Managers will deliberately conceal operational information and delay information disclosure for self-interest, resulting in the opaque operation of the enterprise (Elayan et al., 2008) [60]. The low transparency of the enterprise makes it impossible for the host country’s regulatory system to accurately identify the M&A motivation and operational logic, triggering excessive institutional vigilance and political risk concerns. The host country will further tighten the review standards, increase institutional review links and legal constraints, and significantly prolong the operational cycle of reverse CBMA (Gama & Bandeira, 2021; Li & Fleury, 2020) [16,61]. Accordingly, this study proposes the following:

H3a. 

Type I agency problems reinforce the impact of host country foreign investment security reviews on reverse CBMA.

Although several surveys have shown that agency problems can negatively impact M&A value creation (Yu et al., 2014), these negative effects are limited when Type II agency issues are prominent [62]. This is because Type II agency issues can have positive effects regarding resource marshaling and preparation and corporate transparency to weaken the negative effects of foreign security reviews on M&A.

First, for firms with prominent Type II agency problems, controlling shareholders are more focused on long-term strategic goals and have a strong ability to encroach on the interests of minority shareholders. When controlling shareholders pursue greater control gains, they not only gain access to numerous resources by increasing their shareholdings but also achieve greater resource allocation by increasing their control over the company, such as regarding talent, assets, and markets (Morck et al., 2005) [63]. In other words, by dominating the company’s decisions and controlling the company’s resources, the controlling shareholder can prepare the firm with more resources to pass host country security reviews and successfully complete the M&A transaction and increase the company’s influence and share price (Opoku & Yin, 2021) [64].

Second, firms with more pronounced Type II agency problems may face stronger external monitoring and disclosure pressures, which can increase information transparency under certain conditions. Specifically, when a firm has significant shareholder agency problems, corporate and private information is highly valued by analysts and the market, which means that the firm is subject to strict external monitoring and regulations. Furthermore, in companies with a high level of excess control, the willingness of controlling shareholders to voluntarily disclose information to avoid an unfavorable financing environment is high (Allaya et al., 2022) [65]. Strong external oversight and voluntary disclosure as described above results in companies exhibiting greater transparency, which reduces host country concerns and undue attention to companies, increases the efficiency of the review process and decision-making, effectively avoids emotional bias in review results, and makes it easier for firms to pass foreign investment security reviews in host countries. In summary, we postulate the following:

H3b. 

Type II agency problems weaken the impact of host country foreign investment security reviews on reverse CBMA.

3. Materials and Methods

3.1. Sample and Data Collection

Our study uses a sample of 503 reverse CBMA conducted and completed by China in developed countries from 2003–2022 for empirical analysis. Chinese firms provide an appropriate research context for our study. First, China is the largest developing country, the second largest OFDI “power” in the world (World Investment Report), and a typical representative of reverse cross-border M&A. Second, EMNE networks are particularly prevalent in China, and the agency problem is accentuated due to the unique institutional characteristics of China (Zhou, 2019) [14]. We selected 2003 as the starting year because it represents an important stage in China’s outward investment development. Following the implementation of the Going Global strategy, overseas M&A activities increased substantially and Chinese firms increasingly pursued strategic asset-seeking acquisitions. During the same period, foreign investment security review mechanisms in host countries also evolved considerably, making this period particularly suitable for examining the proposed relationships.

To test the proposed hypotheses, this study collected data from multiple sources. First, we obtained an initial sample of overseas M&A transactions conducted by Chinese firms from the Zephyr database for the period 2003–2022. We then matched the acquiring firms with listed firms in the CSMAR database to identify overseas M&A transactions undertaken by firms listed on the Shanghai and Shenzhen Stock Exchanges. Following Ilhan-Nas et al. (2018), several screening criteria were applied to construct the final sample [66]. Specifically, (1) the acquirer was required to hold more than 10% of the target firm’s equity after the transaction; (2) acquisitions involving firms operating in the financial sector were excluded; and (3) target firms located in tax-haven countries were removed. In addition, only cross-border M&A transactions involving target firms in developed economies were retained. The focus on developed host economies reflects the conceptual setting of reverse cross-border M&A, which generally involves firms from emerging economies acquiring firms in more advanced institutional and technological environments to obtain strategic assets and capabilities. After applying these screening procedures, the final sample consisted of 503 cross-border M&A transactions between 2003 and 2022. To reduce potential endogeneity concerns and mitigate simultaneity bias, all explanatory variables were lagged by one year. In addition, standard data-cleaning procedures were performed. Continuous variables were winsorized at both tails to reduce the influence of extreme observations, and multiple robustness tests were conducted to ensure the reliability of the results.

3.2. Measurements

3.2.1. Dependent Variable

Referring to Lawrence, Raithatha, and Rodriguez (2021), we chose M&A duration as a variable to measure reverse CBMA behavior [67]. The M&A duration selected in this paper indicates the time spent by Chinese firms on reverse CBMA, namely the time span from the announcement date to the completion date of the M&A. The smaller the value, the faster the M&A is completed.

3.2.2. Independent Variables

The independent variable is the foreign investment security review (FISR), which captures whether a host country has established formal laws or regulations related to foreign investment security reviews. Following prior studies, FISR is operationalized as a binary indicator taking the value of 1 if the host country has a formal security review mechanism and 0 otherwise (Jiang, 2023) [68]. We adopt this measure for three reasons. First, the existing literature has not established a widely accepted continuous index for assessing the stringency of foreign investment security reviews across countries. Second, our theoretical focus emphasizes the institutional gatekeeping effect arising from the existence of security review mechanisms rather than variations in regulatory intensity. Third, substantial heterogeneity in legal scope, filing requirements, sector coverage, and implementation procedures across countries makes constructing a comparable cross-country index challenging and potentially subject to subjective weighting bias.

3.2.3. Mediator Variables

We used a dummy variable to measure business groups (BG), which takes the value of 1 if the firm is affiliated with an enterprise group and 0 otherwise (Shi et al., 2021) [24]. First, we specified the conditions required for a group according to the Interim Regulations on Registration and Management of Enterprise Groups issued by the State Administration for Industry and Commerce of the People’s Republic of China in 1998, which stated that the parent company of an enterprise group had a registered capital of at least 50 million RMB and at least five subsidiaries. We extracted the number of affiliates of the enterprise from the Guotaian database and examined whether the final controlling entities had more than 5 companies. Finally, we further reexamined the measurement of enterprise group affiliation through relevant secondary sources (e.g., corporate websites and media reports).

We measured corporate political networking (PN) using a hierarchical weighted approach following You et al. (2010) [69]. First, we manually reviewed executives’ resumes to identify whether executives had previous or current political experience. Political experience included serving as government officials, managers of state-owned banks, presidents or vice presidents of government-supervised business associations, CPPCC members, NPC deputies, model workers, or recipients of government honors. Second, executives with political backgrounds were assigned weighted scores according to the administrative hierarchy level of their political positions. Specifically, political positions at the central, provincial, municipal, and county/district levels were assigned scores of 4, 3, 2, and 1, respectively. Finally, weighted political scores of all executives within the firm were aggregated to construct the firm-level political networking indicator.

We chose the operating expense ratio (Agency_cost1, AC1), which is the ratio of the sum of a company’s sales and management expenses to its main operating income, to measure the Type I agency problem. This indicator captures the excess overhead incurred by executives’ over-employment spending, through which it is possible to evaluate managers’ over-budget spending and other agency costs, with larger values representing more serious agency problems between shareholders and managers.

We adopted the proportion of related-party other receivables to total assets (Agency_cost2, AC2) to measure Type II agency problems. Type II agency problems primarily arise from conflicts between controlling and minority shareholders, where controlling shareholders may engage in tunneling behavior to appropriate firm resources. Specifically, inter-firm lending transactions have been widely regarded as effective indicators of controlling shareholder expropriation because resource transfers disguised as “temporary loans” within other receivables often exhibit strong concealment characteristics. Following prior studies, we identified other receivables associated with related-party transactions involving controlling shareholders, parent companies, and subsidiaries and divided this amount by total assets (Zheng et al., 2014) [70]. Larger values indicate more severe principal–principal agency conflicts. Compared with broader governance indicators, this measure directly captures realized resource appropriation behavior and therefore provides a closer representation of the theoretical mechanism underlying Type II agency conflicts.

3.2.4. Control Variables

Following prior studies, we included a set of control variables that may influence the duration of CBMA. To improve the robustness of the empirical analysis, control variables were introduced at both the firm and country levels. At the firm level, we controlled for the number of executives, state ownership, private ownership, firm size, acquirer R&D intensity, and executive team stability. Firm characteristics and governance structures may influence firms’ decision-making efficiency, resource availability, and ability to manage complex acquisition processes. Acquirer state ownership is a dummy variable equal to 1 if the acquiring firm is state-owned and 0 otherwise, while private ownership is a dummy variable equal to 1 if the acquirer is privately owned and 0 otherwise. Executive team stability is measured as a composite indicator incorporating executive departures, new appointments, and changes in team size. Acquirer R&D intensity is measured as the ratio of R&D expenditure to operating revenue. Firms with stronger technological capabilities and innovation investments generally possess greater absorptive capacity and adaptability in overseas acquisitions, which may facilitate integration processes and improve acquisition efficiency. At the country level, we controlled for the target country’s economic freedom, China’s GDP growth, foreign direct investment (FDI) inflows, and formal institutional distance between China and the host country. Country-level factors may affect the external environment in which cross-border acquisitions take place. In particular, formal institutional distance captures differences in regulatory and institutional environments between China and host countries. Greater institutional distance may increase information asymmetry, legitimacy challenges, and transaction uncertainty during cross-border M&A activities, thereby influencing acquisition duration. Detailed variable names and corresponding measurements are presented in Table 1.

Table 1. Variable Definitions.

	Variable Names	Measurement
DUR	Duration	Number of days between the announcement date and the completion date of the M&A transaction.
FISR	Foreign investment security review	Indicator = 1 if the host country has established a formal foreign investment security review mechanism, 0 otherwise.
BG	Business groups	Indicator = 1 if the acquiring firm is affiliated with a business group, 0 otherwise.
PN	Political networking	Executives with political backgrounds are weighted by administrative rank; firm-level PN is the sum of weighted scores across top executives.
AC1	Type I agency problem	Ratio of the sum of selling expenses and administrative expenses to operating revenue.
AC2	Type II agency problem	Ratio of related-party other receivables to total assets.
SM	Senior management team size	Number of executives in the top management team.
SOE	State ownership	Indicator = 1 if the acquirer is a state-owned enterprise, 0 otherwise.
POE	Private ownership	Indicator = 1 if the acquirer is a privately owned enterprise, 0 otherwise.
SIZE	Firm size	Natural logarithm of total assets of the acquiring firm.
RD	R&D intensity	Ratio of R&D expenditure to operating revenue.
ETS	Executive team stability	Composite indicator constructed from executive departures, new appointments, and changes in team size.
EF	Economic freedom	Economic freedom index of the host country.
GDP	China GDP	Annual GDP growth rate of China.
FDI	Foreign direct investment inflows	Natural logarithm of host country foreign direct inflows.
InsD	Institutional distance	Kogut–Singh formal institutional distance between China and the host country (based on WGI).

3.3. Model

Since M&A duration is measured as the number of days between deal announcement and completion, the dependent variable exhibits non-negative count-like characteristics and duration properties. Previous studies commonly employ Poisson or Negative Binomial models for such outcomes. To determine the most appropriate specification, we conducted a two-step diagnostic procedure. First, we examined the distributional properties of the dependent variable. Descriptive statistics reveal that M&A duration exhibits substantial dispersion and a right-skewed distribution, suggesting the potential presence of over-dispersion and the influence of extreme observations. Second, we formally tested for over-dispersion using a likelihood-ratio test of the dispersion parameter (α). The test rejects the null hypothesis that α = 0 (p < 0.001), indicating that the equidispersion assumption required by Poisson regression does not hold. Therefore, we adopt Negative Binomial regression as the baseline estimation model. By introducing an additional dispersion parameter, the Negative Binomial model accommodates unobserved heterogeneity and provides more reliable estimates when the conditional variance exceeds the conditional mean. The baseline model is specified as follows:

$$\mathrm{l}\mathrm{n}\left({\mathrm{\mu }}_{\mathrm{i}}\right)={\mathrm{\beta }}_0+{\mathrm{\beta }}_1{\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{R}}_{\mathrm{i}}+{\mathrm{\beta }}_2{\mathrm{M}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{o}\mathrm{r}}_{\mathrm{i}}+{\mathrm{\beta }}_3{({\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{R}}_{\mathrm{i}}\times \mathrm{M}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{o}\mathrm{r}}_{\mathrm{i}})+\mathrm{\gamma }{\mathrm{X}}_{\mathrm{i}}+{\mathrm{\epsilon }}_{\mathrm{i}}$$

(1)

In the above specification, FISR denotes foreign investment security review, X_i represents a vector of control variables including SM, SOE, POE, SIZE, RD, ETS, EF, GDP, FDI, and InsD, and ε_i represents the disturbance term.

We used STATA16 software to conduct descriptive statistics and pairwise correlation analysis, and the pairwise correlation analysis is presented in Table 2. As can be seen from Table 2, the correlation coefficients and standard deviations between variables are small, which indicates that the overall stability of the data is strong and can be better for regression statistical analysis. Given the nature of cross-border M&A transactions, acquisition duration may exhibit substantial heterogeneity and right-skewed distribution characteristics because transaction completion often depends on multiple factors, including negotiation processes, due diligence procedures, and regulatory reviews. To assess potential multicollinearity concerns, variance inflation factors (VIFs) were calculated based on the full model. The mean VIF is 1.81 and the maximum VIF is 3.20, both below the commonly accepted threshold of 10, indicating that multicollinearity is unlikely to be a concern.

Table 2. Correlations matrix of variables.

	SM	SOE	POE	SIZE	RD	ETS	EF	GDP	FDI	InsD	DUR	FISR	BG	PN	AC1	AC2
SM	1.00
SOE	0.34 ***	1.00
POE	−0.27 ***	−0.68 ***	1.00
SIZE	0.38 ***	0.31 ***	−0.16 ***	1.00
RD	0.11 *	−0.13 **	0.07	−0.10 *	1.00
ETS	0.01	−0.03	0.04	−0.15 ***	−0.05	1.00
EF	0.07	0.03	−0.04	−0.00	−0.05	0.06	1.00
GDP	0.15 ***	0.22 ***	−0.01	−0.06	−0.15 ***	0.24 ***	0.05	1.00
FDI	−0.17 ***	−0.27 ***	0.09 *	0.05	0.18 ***	−0.14 **	−0.01	−0.80 ***	1.00
InsD	0.20 ***	0.22 ***	−0.15 **	0.08	−0.05	0.02	0.60 ***	0.31 ***	−0.36 ***	1.00
DUR	−0.13 **	−0.01	0.02	−0.20 ***	−0.01	0.09 *	0.24 ***	0.15 ***	−0.08	0.09 *	1.00
FISR	−0.02	0.00	−0.07	0.08	0.04	0.04	0.19 ***	−0.09 *	0.15 **	0.05	0.20 ***	1.00
BG	0.06	0.07	−0.10 *	0.17 ***	0.01	−0.03	0.08	−0.08	0.07	0.06	−0.10 *	−0.01	1.00
PN	0.33 ***	0.08	−0.08	0.33 ***	0.01	−0.10 *	−0.03	0.03	−0.08	0.08	−0.05	0.04	0.05	1.00
AC1	0.07	−0.15 ***	0.18 ***	−0.22 ***	0.56 ***	−0.00	−0.07	−0.02	0.03	−0.09 *	0.08	0.05	0.02	0.02	1.00
AC2	0.03	0.16 ***	−0.07	0.06	−0.11 *	0.02	0.02	0.10 *	−0.11 *	0.07	−0.04	0.03	0.06	0.01	−0.01	1.00
Mean	16.43	0.22	0.63	22.5	4.57	0.93	75.61	107.07	20.54	3.46	385.1	0.86	0.98	1.68	0.18	0.01
S.D.	3.93	0.41	0.48	1.17	5.17	0.07	6.41	1.52	0.12	1	342.42	0.35	0.13	2.21	0.13	0.02
N	503	503	503	503	503	503	503	503	503	503	503	503	503	503	503	503

Note: * p < 0.05; ** p < 0.01; *** p < 0.001.

4. Results

4.1. Hypothesis Testing Results

To test the proposed hypotheses, we employed Negative Binomial regression. The estimation results are reported in Table 3. Model 1 presents the baseline model including only control variables. Model 2 adds the main explanatory variable, foreign investment security reviews (FISR). The coefficient of FISR is positive and significant (b = 0.838, p < 0.001), suggesting that the presence of host country security review mechanisms significantly prolongs the duration of reverse cross-border M&A. This finding supports H1.

Table 3. Regression result.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
SM	−0.031	−0.024	−0.024	−0.027	−0.034 *	−0.037 *
	(0.020)	(0.019)	(0.019)	(0.020)	(0.020)	(0.020)
SOE	0.193	0.276	0.249	0.314	0.269	0.334
	(0.241)	(0.242)	(0.241)	(0.241)	(0.240)	(0.241)
POE	0.035	0.095	0.068	0.108	0.073	0.159
	(0.192)	(0.191)	(0.192)	(0.191)	(0.193)	(0.192)
SIZE	−0.120 *	−0.167 **	−0.155 **	−0.164 **	−0.156 **	−0.162 **
	(0.066)	(0.067)	(0.068)	(0.069)	(0.068)	(0.067)
RD	0.005	−0.001	−0.001	0.001	−0.012	−0.000
	(0.015)	(0.014)	(0.014)	(0.015)	(0.017)	(0.014)
ETS	0.116	0.157	0.100	0.409	0.304	−0.072
	(0.970)	(0.961)	(0.957)	(0.967)	(0.941)	(0.959)
EF	0.044 ***	0.038 ***	0.039 ***	0.038 ***	0.039 ***	0.038 ***
	(0.014)	(0.014)	(0.014)	(0.014)	(0.014)	(0.013)
GDP	0.225 **	0.214 **	0.234 **	0.172 *	0.179 **	0.250 ***
	(0.093)	(0.091)	(0.091)	(0.093)	(0.090)	(0.092)
FDI	1.235	0.751	0.860	0.428	0.300	1.266
	(1.003)	(0.968)	(0.963)	(0.980)	(0.943)	(1.041)
InsD	−0.087	−0.060	−0.065	−0.082	−0.079	−0.043
	(0.095)	(0.095)	(0.095)	(0.098)	(0.094)	(0.093)
FISR		0.838 ***	0.996 ***	1.029 ***	1.138 ***	0.766 ***
		(0.207)	(0.212)	(0.206)	(0.211)	(0.212)
BG			6.284 ***
			(1.829)
BG × FISR			−6.777 ***
			(1.902)
PN				−0.371 ***
				(0.117)
PN × FISR				0.384 ***
				(0.124)
AC1					−8.067 ***
					(1.873)
AC1 × FISR					8.887 ***
					(1.970)
AC2						38.381 **
						(19.193)
AC2 × FISR						−49.309 **
						(19.588)
LR chi2	29.18	43.05	48.80	51.24	56.49	53.00
Pseudo R2	0.0044	0.0064	0.0073	0.0076	0.0084	0.0079
N	503	503	503	503	503	503

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; and Standard errors in parentheses.

Model 3 introduces the moderating effect of business groups (BG). The interaction term between FISR and business groups is significantly negative (b = −6.777, p < 0.001), indicating that business groups weaken the positive relationship between security reviews and M&A duration. This finding supports H2a. Model 4 examines the moderating role of political networking (PN). The interaction term between FISR and political networking is significantly positive (b = 0.384, p < 0.001), suggesting that political networking strengthens the impact of security reviews on M&A duration. Therefore, H2b is supported.

Model 5 tests the moderating role of Type I agency problems (AC1). The interaction coefficient between FISR and AC1 is significantly positive (b = 8.887, p < 0.001), indicating that Type I agency conflicts strengthen the delaying effect of security reviews on M&A duration. Thus, H3a is supported. Model 6 investigates the moderating effect of Type II agency problems (AC2). The interaction coefficient between FISR and AC2 is significantly negative (b = −49.309, p < 0.05), indicating that Type II agency conflicts weaken the impact of security reviews on M&A duration. Therefore, H3b is supported.

To facilitate interpretation of interaction effects in economic terms, we additionally estimated average marginal effects based on the negative binomial regressions. Rather than directly interpreting interaction coefficients, marginal effects quantify the expected change in M&A duration associated with foreign investment security reviews under different organizational conditions. Specifically, for standalone firms, security reviews increase expected M&A duration by approximately 695 days, whereas the increase declines to approximately 234 days for business-group-affiliated firms. Moreover, we calculated the marginal effect of foreign investment security reviews on expected M&A duration at different levels of political networking. The results reveal substantial heterogeneity across political networking levels. At a low level of political networking (−1 SD), foreign investment security reviews increase expected M&A duration by approximately 68 days. At the mean level of political networking, security reviews increase expected M&A duration by approximately 273 days (p < 0.01). At a high level of political networking (+1 SD), the effect further increases to approximately 370 days (p < 0.01). At a low level of Type I agency problems, no significant effect was observed. At the mean level, security reviews increase expected M&A duration by approximately 291 days (p < 0.01). At a high level of Type I agency problems, the effect further rises to approximately 430 days (p < 0.01). At a low level of Type II agency problems, foreign investment security reviews increase expected M&A duration by approximately 438 days (p < 0.01). At the mean level, the effect decreases to approximately 226 days (p < 0.01). At a high level, the effect gradually attenuates and eventually becomes indistinguishable from zero.

Appendix A.1 provides subgroup descriptive statistics that further illustrate structural differences between group-affiliated and standalone firms, as well as firms with different levels of agency costs, supporting the heterogeneity interpretation of our main results.

4.2. Additional Regression Results

Given the unique characteristics of the CFIUS review process, the U.S. security review of CBMA by Chinese firms is extremely strict, and the U.S. places China on the list of “countries of special concern” and requires the Department of Commerce to provide a separate report on Chinese investments in the U.S. to CFIUS biennially. This substantially increases the barriers faced by Chinese investment in the U.S. Furthermore, in the sample of reverse M&A in this paper, the proportion of target firms located in the United States is relatively large, so excluding the sample of Chinese firms acquiring U.S. firms and running regressions is conducive to a more reasonable examination of the impact of EMNE networks and two types of agency problems on security reviews and CBMA. The stability of host country composition after removing U.S. observations is reported in Appendix A.2, confirming that the subsample retains representativeness of major developed host economies. Table 4 shows the results of the regressions excluding the sample of M&A in the U.S. The regression coefficients of the six models in the table are consistent and significant with the findings of the previous study, indicating that the previous regression results are robust.

Table 4. Excluding U.S. sample (DV = Duration).

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
SM	−0.025	−0.016	−0.016	−0.024	−0.027	−0.034
	(0.025)	(0.024)	(0.024)	(0.024)	(0.025)	(0.025)
SOE	0.077	0.147	0.121	0.265	0.079	0.243
	(0.315)	(0.315)	(0.313)	(0.318)	(0.314)	(0.315)
POE	0.230	0.299	0.271	0.304	0.269	0.362
	(0.257)	(0.256)	(0.256)	(0.254)	(0.262)	(0.255)
SIZE	−0.132	−0.188 **	−0.181 **	−0.221 **	−0.184 **	−0.165 *
	(0.086)	(0.088)	(0.088)	(0.094)	(0.088)	(0.087)
RD	0.005	−0.003	−0.003	−0.001	−0.016	0.002
	(0.020)	(0.020)	(0.019)	(0.020)	(0.022)	(0.020)
ETS	0.593	0.697	0.615	1.555	0.931	0.254
	(1.244)	(1.228)	(1.222)	(1.276)	(1.175)	(1.249)
EF	0.028 *	0.027 *	0.028 *	0.026 *	0.025	0.029 *
	(0.015)	(0.015)	(0.015)	(0.015)	(0.015)	(0.015)
GDP	0.134	0.131	0.154	0.078	0.090	0.176
	(0.110)	(0.109)	(0.108)	(0.113)	(0.106)	(0.110)
FDI	0.270	−0.144	−0.011	−0.352	−0.720	0.409
	(1.190)	(1.154)	(1.150)	(1.156)	(1.077)	(1.258)
InsD	0.084	0.083	0.072	0.053	0.074	0.072
	(0.116)	(0.116)	(0.116)	(0.117)	(0.116)	(0.113)
FISR		0.719 ***	0.884 ***	1.012 ***	1.230 ***	0.646 ***
		(0.237)	(0.242)	(0.249)	(0.247)	(0.243)
BG			6.129 ***
			(1.931)
BG × FISR			−6.695 ***
			(2.132)
PN				−0.451 **
				(0.189)
PN × FISR				0.513 ***
				(0.197)
AC1					−11.448 ***
					(2.355)
AC1 × FISR					12.341 ***
					(2.534)
AC2						35.039 *
						(21.259)
AC2 × FISR						−47.369 **
						(21.894)
LR chi2	24.75	32.86	37.36	39.80	47.86	40.008
Pseudo R2	0.0056	0.0075	0.0085	0.0091	0.0109	0.0091
N	344	344	344	344	344	344

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; and Standard errors in parentheses.

Secondly, to further examine the moderating mechanism of political networking, we performed subgroup analyses based on tercile splits of the political networking measure, omitting the middle tercile to create distinct high and low political embeddedness groups. The results indicate that the positive moderating effect of political networking is more pronounced among firms with high political embeddedness, providing additional support for the robustness of our findings (Table 5).

Table 5. High/low political embedded subgroup regressions (DV = Duration).

	Low Political Embeddedness	High Political Embeddedness
	Model 1	Model 2
SM	−0.039	−0.054
	(0.032)	(0.033)
SOE	0.039	0.629
	(0.385)	(0.415)
POE	−0.025	0.562
	(0.306)	(0.360)
SIZE	−0.082	−0.129
	(0.103)	(0.125)
RD	−0.003	0.019
	(0.017)	(0.034)
ETS	0.126	0.619
	(1.546)	(1.864)
EF	0.048 **	0.028
	(0.019)	(0.028)
GDP	0.191	0.132
	(0.142)	(0.161)
FDI	0.501	1.018
	(1.525)	(1.982)
InsD	−0.232	0.096
	(0.145)	(0.175)
FISR	0.287	2.315 ***
	(0.325)	(0.401)
LR chi2	13.94	38.51
Pseudo R2	0.0043	0.0188
N	241	158

Note: *** p < 0.01, ** p < 0.05; and Standard errors in parentheses.

Thirdly, to further examine the moderating mechanism of agency problems, we performed subgroup analyses based on tercile splits of the agency problems measure, omitting the middle tercile to create distinct high and low agency problems groups. The results indicate that the reinforcing effect of Type I agency problems on the relationship between foreign investment security reviews and M&A duration is more pronounced among firms with higher levels of Type I agency problems, whereas the delaying effect of foreign investment security reviews on M&A duration becomes weaker among firms with relatively high Type II agency problems than among firms with relatively low Type II agency problems. This pattern is consistent with our theoretical argument and provides additional support for the robustness of the baseline findings (Table 6).

Table 6. High/low agency problems subgroup regressions (DV = Duration).

	Low AC1	High AC1	Low AC2	High AC2
	Model 1	Model 2	Model 3	Model 4
SM	−0.043	0.004	0.002	−0.040
	(0.038)	(0.032)	(0.034)	(0.044)
SOE	0.256	0.141	0.158	0.846
	(0.401)	(0.429)	(0.365)	(0.550)
POE	0.046	0.084	−0.019	0.716
	(0.332)	(0.280)	(0.309)	(0.449)
SIZE	−0.139	−0.257 **	−0.221 **	−0.211
	(0.140)	(0.126)	(0.111)	(0.146)
RD	−0.015	−0.018	−0.015	0.005
	(0.063)	(0.016)	(0.024)	(0.030)
ETS	−0.166	0.286	−1.330	0.211
	(1.783)	(1.407)	(1.483)	(1.963)
EF	0.029	0.045 *	0.045 **	0.030
	(0.022)	(0.027)	(0.023)	(0.026)
GDP	0.298 **	0.772 **	0.417 **	0.043
	(0.129)	(0.364)	(0.170)	(0.161)
FDI	2.673	1.986	3.582 *	−0.922
	(1.692)	(2.073)	(2.125)	(1.920)
InsD	−0.082	0.030	−0.189	−0.011
	(0.177)	(0.148)	(0.182)	(0.186)
FISR	0.476	2.121 ***	0.980 ***	0.525
	(0.357)	(0.394)	(0.347)	(0.425)
LR chi2	15.83	31.742	20.532	13.435
Pseudo R2	0.007	0.014	0.009	0.006
N	168	167	169	166

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; and Standard errors in parentheses.

Furthermore, we estimated alternative model specifications to examine the robustness of our findings. Given the duration-related nature of M&A completion, we additionally employed Cox survival models as an alternative estimation approach. The results reported in Table 7 are fully consistent with the baseline negative binomial estimations in both direction and statistical significance, providing additional support for the robustness of the proposed hypotheses.

Table 7. Alternative model (DV = Duration).

	Model 1	Model 2	Model 3	Model 4	Model 5
SM	0.032 **	0.033 **	0.037 ***	0.038 ***	0.039 ***
	(0.014)	(0.014)	(0.014)	(0.014)	(0.014)
SOE	−0.091	−0.070	−0.116	−0.088	−0.127
	(0.165)	(0.166)	(0.166)	(0.166)	(0.166)
POE	0.013	0.037	0.006	0.030	−0.007
	(0.132)	(0.133)	(0.132)	(0.134)	(0.132)
SIZE	0.109 **	0.099 **	0.118 **	0.097 **	0.102 **
	(0.045)	(0.046)	(0.046)	(0.046)	(0.045)
RD	0.003	0.003	0.002	0.009	0.002
	(0.009)	(0.009)	(0.009)	(0.011)	(0.009)
ETS	−0.378	−0.347	−0.541	−0.509	−0.277
	(0.648)	(0.648)	(0.655)	(0.650)	(0.648)
EF	−0.030 ***	−0.031 ***	−0.032 ***	−0.031 ***	−0.030 ***
	(0.009)	(0.009)	(0.009)	(0.009)	(0.009)
GDP	−0.149 ***	−0.163 ***	−0.140 **	−0.140 **	−0.170 ***
	(0.055)	(0.055)	(0.056)	(0.055)	(0.056)
FDI	−0.425	−0.493	−0.408	−0.330	−0.817
	(0.643)	(0.643)	(0.645)	(0.645)	(0.686)
InsD	0.019	0.023	0.042	0.029	0.009
	(0.064)	(0.064)	(0.066)	(0.065)	(0.064)
FISR	−0.659 ***	−0.672 ***	−0.705 ***	−0.725 ***	−0.634 ***
	(0.142)	(0.144)	(0.143)	(0.145)	(0.144)
BG		−0.240 *
		(0.139)
BG × FISR		0.301 **
		(0.147)
PN			0.346 *
			(0.179)
PN × FISR			−0.414 **
			(0.188)
AC1				0.329 **
				(0.152)
AC1 × FISR				−0.398 **
				(0.163)
AC2					−0.603 **
					(0.259)
AC2 × FISR					0.691 ***
					(0.263)
LR chi2	67.76	71.45	72.70	73.07	76.77
p	0.000	0.000	0.000	0.000	0.000
N	503	503	503	503	503

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; Standard errors in parentheses.

Finally, we use propensity score matching (PSM) to address potential self-selection bias in FISR. We divided the sample into treatment and control groups according to whether the host country had a formal foreign investment security review mechanism, followed by radius caliper matching. Table 8 reports balance diagnostics before and after propensity score matching. The results indicate that the matching procedure substantially improves covariate balance for most variables. The regression results in Table 9 indicate that FISR significantly increases M&A duration, which is consistent with the baseline results.

Table 8. PSM balance diagnostics.

	Balance Metrics
	Before Matching	After Matching
SM	−15.5	−6.0
SOE	1.2	0.7
POE	−21.7	8.3
SIZE	24.0	−12.6
RD	12.5	−10.8
ETS	10.7	5.9
EF	56.0	11.9
GDP	−25.6	9.9
FDI	36.7	−15.1
InsD	14.7	13.9

Table 9. Propensity Score Matching (DV = Duration).

	Model 1
SM	−0.025
	(0.021)
SOE	0.451
	(0.285)
POE	0.230
	(0.236)
SIZE	−0.183 **
	(0.079)
RD	−0.001
	(0.017)
ETS	−0.429
	(1.085)
EF	0.034 *
	(0.017)
GDP	0.254 ***
	(0.097)
FDI	1.229
	(1.146)
InsD	−0.024
	(0.104)
FISR	0.913 ***
	(0.216)
LR chi2	32.39
Pseudo R2	0.0072
Observations	424

Note: Standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

5. Discussion

5.1. Theoretical Implications

This study contributes to the literature on cross-border M&A, international business, and EMNE internationalization in three ways. Rather than focusing on isolated determinants of cross-border M&A outcomes, this study develops an integrated analytical framework that incorporates external institutional conditions, firm relational resources, and internal governance characteristics. The contribution of this study lies in comparing how different categories of firm characteristics independently shape responses to host country security review environments.

First, this study contributes to the literature on cross-border M&A by extending research on the role of host country security reviews in reverse CBMAs undertaken by EMNEs. Existing studies acknowledge that institutional distance influences cross-border investment outcomes (Dikova et al., 2019), yet limited attention has been paid to the directional aspect of institutional differences, particularly situations in which EMNEs enter developed economies with stronger institutional environments (Konara & Shirodkar, 2018) [38,71]. Our findings suggest that security reviews represent an important institutional barrier in such contexts and significantly prolong M&A duration. This finding refines existing understanding of how institutional environments affect cross-border acquisition processes and responds to calls for greater attention to the direction rather than simply the magnitude of institutional distance (Konara and Shirodkar, 2018) [71].

Second, this study contributes to the literature on EMNE network resources by comparing the heterogeneous roles of business group affiliation and political networking in shaping firms’ responses to security review environments. Existing research provides mixed evidence regarding the effectiveness of different relational resources during internationalization. Our findings demonstrate that business group affiliation and political networking do not generate equivalent effects. Specifically, business group affiliation helps firms mitigate the negative effects of security reviews by facilitating internal resource sharing, coordination, and experience transfer. By contrast, political networking becomes less effective and may even intensify regulatory concerns under stricter institutional environments. These findings contribute to the contingency perspective on EMNE network resources by showing that the value of relational resources depends on the institutional context in which they operate.

Third, this study contributes to the corporate governance literature by distinguishing between two heterogeneous forms of agency conflict. Prior research often treats agency problems as a single construct, producing inconsistent conclusions regarding their effects on international investment behavior (Breuer et al., 2018; Datta et al., 2020) [17,18]. Our findings suggest that different agency conflicts influence firms’ responses to external institutional constraints in distinct ways. Type I agency conflicts amplify the delaying effect of security reviews, indicating that managerial inefficiencies and internal governance frictions may intensify challenges associated with regulatory scrutiny. By contrast, Type II agency conflicts exhibit a weaker constraining effect and may facilitate organizational coordination and resource mobilization under external pressure. These findings provide a more nuanced understanding of how governance structures shape cross-border M&A outcomes.

5.2. Managerial Implications

This study provides practical implications for enterprise managers and policymakers seeking to improve cross-border acquisition effectiveness under increasingly stringent security review environments. Reverse cross-border M&A activities involve substantial uncertainty and regulatory complexity, making it important for firms to align internal governance, network resources, and regulatory adaptation strategies.

For enterprise managers, firms should improve organizational transparency and compliance capabilities when conducting acquisitions in developed economies characterized by stricter institutional environments. Establishing standardized accounting systems, strengthening intellectual property protection, improving ESG practices, and enhancing disclosure quality may reduce information asymmetry and facilitate communication with host country regulators. The findings further suggest that managerial strategies should vary across different types of firms. For state-owned enterprises (SOEs), particular attention should be devoted to legitimacy concerns because SOEs are often subject to stronger political scrutiny in foreign markets. Managers of SOEs should strengthen transparency mechanisms, clarify commercial motivations behind acquisitions, and reduce host country concerns regarding political influence. Building independent governance structures and improving disclosure practices may help alleviate regulatory concerns during security review processes. For private firms, although political concerns may be less pronounced, resource constraints and limited international experience may represent major challenges. Private firms should therefore make greater use of business group resources, strategic alliances, and external professional services to compensate for information disadvantages and improve acquisition preparedness.

In addition, managerial priorities may differ between high-technology and traditional-industry acquirers. High-technology firms are more likely to face stricter security reviews because acquisitions involving strategic technologies, data assets, or sensitive industries often attract greater regulatory attention. These firms should strengthen intellectual property management, proactively communicate with regulators, and conduct more extensive pre-acquisition regulatory assessments. By contrast, firms in traditional industries may place greater emphasis on improving operational efficiency and resource coordination during the acquisition process.

Regarding internal governance, firms should adopt governance mechanisms that reduce managerial opportunism and improve strategic coordination. To mitigate the adverse effects associated with Type I agency conflicts, firms should improve incentive alignment and strengthen monitoring mechanisms in cross-border investment decision-making. For firms characterized by more concentrated ownership structures, controlling shareholders should be encouraged to direct resources toward long-term strategic objectives and acquisition preparation activities rather than pursuing short-term private benefits.

For policymakers, improving transparency and predictability in foreign investment review procedures may help reduce unnecessary uncertainty and transaction costs associated with cross-border acquisitions.

5.3. Limitations and Future Directions

There are certain limitations.

First, this study focuses on reverse CBMA involving acquisitions in developed economies. Future studies may further investigate whether the effects of foreign investment security reviews differ under positive and negative institutional distance conditions and compare their influence across different host country institutional environments.

Second, this study measures foreign investment security reviews using a binary indicator reflecting the existence of a formal security review mechanism in host countries, consistent with prior research and aligned with our theoretical focus on the institutional gatekeeping role of security review mechanisms. Future research may extend this approach by developing more refined measures of review intensity that incorporate dimensions such as regulatory scope, filing thresholds, review procedures, enforcement practices, and temporal changes in regulatory policies. Such efforts may provide additional insights into cross-country differences in regulatory environments, although constructing a theoretically validated and comparable cross-national index remains an important challenge.

Third, future research may further improve the measurement of agency conflicts by adopting multidimensional indicators that capture broader governance characteristics. For example, future studies may integrate ownership concentration, divergence between cash-flow and control rights, board structure, related-party transactions, and other governance attributes to construct more comprehensive measures of agency conflicts. In addition, although the current indicators follow established approaches in the prior literature and align closely with our theoretical constructs, more refined measurements may further strengthen construct validity and provide richer insights into how different governance dimensions shape firms’ responses to external institutional constraints.

Fourth, while the present study focuses on M&A duration as the primary outcome variable and an indicator of transaction process efficiency, future research could extend the analysis to post-acquisition outcomes, including post-merger integration effectiveness, subsidiary performance, innovation outcomes, and long-term value creation. Such extensions may help determine whether the effects of foreign investment security reviews and organizational characteristics persist beyond the transaction completion stage and continue to influence firms’ subsequent strategic and operational outcomes. Examining whether security reviews exert persistent effects beyond the completion stage would provide a more comprehensive understanding of EMNE internationalization processes.

Finally, future research could examine how dynamic institutional environments influence the relationship identified in this study. In particular, host country political cycles, changes in government leadership, geopolitical tensions, and shifts in regulatory priorities may alter the implementation and intensity of security review practices. These dynamic contextual factors may further shape firms’ responses to institutional constraints.

Appendix A.1

To provide additional information regarding sample heterogeneity, we report subgroup descriptive statistics for key firm characteristics. Specifically, the sample is divided according to (1) business group affiliation (group-affiliated firms versus standalone firms), and (2) agency costs (high versus low agency-cost firms). These subgroup statistics provide a more comprehensive understanding of structural differences across firm categories and improve the transparency of empirical analyses.

Table A1. Subgroup Descriptive Statistics (group-affiliated firms versus standalone firms).

	BG Firms		Standalone
	Mean	S.D.	Mean	S.D.
DUR	378.59	342.19	402.93	364.95
FISR	0.85	0.35	0.73	0.46
SM	16.52	4.10	17.20	5.17
SOE	0.23	0.42	0.27	0.46
POE	0.62	0.49	0.73	0.46
SIZE	22.51	1.17	22.28	2.88
RD	4.50	5.15	3.69	4.06
ETS	0.93	0.07	0.94	0.05
EF	75.71	6.45	73.62	6.58
GDP	107.09	1.52	108.80	2.69
FDI	20.54	0.14	20.38	0.25
InsD	3.47	0.99	3.54	1.12

Table A1. Subgroup Descriptive Statistics (group-affiliated firms versus standalone firms).

	BG Firms		Standalone
	Mean	S.D.	Mean	S.D.
DUR	378.59	342.19	402.93	364.95
FISR	0.85	0.35	0.73	0.46
SM	16.52	4.10	17.20	5.17
SOE	0.23	0.42	0.27	0.46
POE	0.62	0.49	0.73	0.46
SIZE	22.51	1.17	22.28	2.88
RD	4.50	5.15	3.69	4.06
ETS	0.93	0.07	0.94	0.05
EF	75.71	6.45	73.62	6.58
GDP	107.09	1.52	108.80	2.69
FDI	20.54	0.14	20.38	0.25
InsD	3.47	0.99	3.54	1.12

Table A2. Subgroup Descriptive Statistics (high versus low agency-cost firms).

	High AC1		Low AC1		High AC2		Low AC2
	Mean	S.D.	Mean	S.D.	Mean	S.D.	Mean	S.D.
DUR	397.55	342.35	360.88	342.34	389.15	350.70	369.34	334.44
FISR	0.87	0.34	0.83	0.38	0.83	0.37	0.87	0.34
SM	16.32	4.48	16.76	3.72	16.46	4.19	16.63	4.07
SOE	0.15	0.36	0.30	0.46	0.25	0.43	0.20	0.40
POE	0.70	0.46	0.55	0.50	0.60	0.49	0.65	0.48
SIZE	22.22	1.12	22.80	1.30	22.54	1.19	22.47	1.30
RD	5.84	6.43	3.09	2.71	4.00	4.07	4.95	5.96
ETS	0.92	0.07	0.93	0.07	0.93	0.07	0.93	0.07
EF	75.08	6.45	76.22	6.43	75.41	6.75	75.88	6.15
GDP	107.05	0.98	107.23	2.02	107.24	1.72	107.04	1.43
FDI	20.54	0.12	20.53	0.17	20.51	0.16	20.55	0.12
InsD	3.37	1.05	3.56	0.92	3.47	1.07	3.46	0.91

Table A2. Subgroup Descriptive Statistics (high versus low agency-cost firms).

	High AC1		Low AC1		High AC2		Low AC2
	Mean	S.D.	Mean	S.D.	Mean	S.D.	Mean	S.D.
DUR	397.55	342.35	360.88	342.34	389.15	350.70	369.34	334.44
FISR	0.87	0.34	0.83	0.38	0.83	0.37	0.87	0.34
SM	16.32	4.48	16.76	3.72	16.46	4.19	16.63	4.07
SOE	0.15	0.36	0.30	0.46	0.25	0.43	0.20	0.40
POE	0.70	0.46	0.55	0.50	0.60	0.49	0.65	0.48
SIZE	22.22	1.12	22.80	1.30	22.54	1.19	22.47	1.30
RD	5.84	6.43	3.09	2.71	4.00	4.07	4.95	5.96
ETS	0.92	0.07	0.93	0.07	0.93	0.07	0.93	0.07
EF	75.08	6.45	76.22	6.43	75.41	6.75	75.88	6.15
GDP	107.05	0.98	107.23	2.02	107.24	1.72	107.04	1.43
FDI	20.54	0.12	20.53	0.17	20.51	0.16	20.55	0.12
InsD	3.37	1.05	3.56	0.92	3.47	1.07	3.46	0.91

Appendix A.2

To improve the transparency of the robustness analysis, Table A3 reports the distribution of the top ten host countries before and after excluding U.S. transactions. The top ten host countries account for approximately 85% of the total sample. The results indicate that although U.S. observations are removed, the relative composition of major host countries remains broadly stable.

Table A3. Host country distribution.

	Full Sample		Excluding U.S.
	Sample	Percent	Sample	Percent
US	159	31.61	0	0
DE	79	15.71	79	22.97
SG	52	10.34	52	15.12
IT	31	6.16	31	9.01
JP	30	5.96	30	8.72
AU	27	5.37	27	7.85
CA	23	4.57	23	6.69
NL	14	2.78	14	4.07
KR	11	2.19	11	3.2
ES	10	1.99	10	2.91

Table A3. Host country distribution.

	Full Sample		Excluding U.S.
	Sample	Percent	Sample	Percent
US	159	31.61	0	0
DE	79	15.71	79	22.97
SG	52	10.34	52	15.12
IT	31	6.16	31	9.01
JP	30	5.96	30	8.72
AU	27	5.37	27	7.85
CA	23	4.57	23	6.69
NL	14	2.78	14	4.07
KR	11	2.19	11	3.2
ES	10	1.99	10	2.91