The theory of institutional pluralism emphasizes that enterprises must proactively adjust their marketing strategies in heterogeneous institutional environments to enhance institutional adaptability, thereby gaining legitimacy in the host country market and reducing institutional friction costs. Parietti (2017) confirm that under conditions of significant institutional distance, multinational corporations need to adopt localization strategies to reduce external environmental uncertainty and mitigate operational risks. Li et al (2022) argue that the dual pressure mechanism generated by institutional distance includes compliance pressure stemming from regulatory differences and legitimacy pressure resulting from cultural cognitive differences, forcing enterprises to implement deep localization transformations in dimensions such as product customization, channel reconstruction, and promotion strategy localization. Shin et al (2020) define institutional distance as encompassing both formal and informal dimensions. Formal institutional distance reflects differences between the home country and the host country in written institutions such as legal norms and policy systems, while informal institutional distance characterizes divergences between the two countries in non-written norms such as cultural traditions and moral values, involving socially recognized behavioral norms like values and customs, and exhibiting spontaneous characteristics. Qian and Zhou (2024) analyze the dynamic game relationship between institutional distance and corporate strategic choices from the perspective of resource dependence theory, arguing that a higher level of institutional differences increases the risk of information asymmetry and deepens the dependence of enterprises on the resources of stakeholders in the host country, thereby prompting enterprises to adopt more comprehensive localization strategies to form a strategic buffer mechanism. Based on the above research, institutional distance encompasses both formal and informal dimensions. The former refers to differences in written systems such as legal norms and policy systems, while the latter refers to differences in non-written norms such as cultural traditions and moral values. A higher level of institutional differences increases the risk of information asymmetry, strengthens the dependence of enterprises on the resources of stakeholders in the host country, and further promotes enterprises to adopt more thorough localization strategies to build a strategic buffer mechanism. Based on this, this study proposes the following hypotheses:

H1: Institutional distance between home and host countries is positively related to localization decisions.

H1-a: Formal institutional distance between home and host countries is positively related to localization decisions.

H1-b: Informal institutional distance between home and host countries is positively related to localization decisions.

Based on information processing patterns and thinking characteristics, managerial cognition can be divided into two basic dimensions: centrality and complexity. Centrality is reflected in managers who limit their information processing scope to specific professional fields or fixed thinking frameworks, similar to shooting with a fixed-focus lens, which can only capture limited image content. Complexity refers to managers’ ability to process diverse information and integrate different thinking modes, operating like panoramic photography, which can accommodate multiple image elements. Scholars such as Kubera (2023) point out that the deeper a manager’s understanding of the host country’s institutional environment, the more they can identify the challenges and opportunities brought by institutional distance and thus are more inclined to adopt localization strategies to reduce operational risks. Research by Farías (2021) shows that managers with high cultural cognitive ability can better understand and adapt to cultural differences in the host country and thus make more effective localization adjustments in product and marketing strategies. Lane (1998) proposes that managerial cognitive complexity lies in the ability to analyze heterogeneous institutional signals and integrate contradictory information, thereby enhancing firms’ sensitivity to host country market specificities and facilitating precise adaptation of marketing strategies in dimensions such as product mix, distribution channels, and brand communication. If managers’ thinking is overly focused on a specific field, it is often difficult to fully grasp the diverse needs of the target market. However, the cognitive complexity of managers enhances decision-making comprehensiveness by integrating diverse information to meet local needs, anticipating and responding to risks through multi-angle thinking, innovating localization strategies with cross-disciplinary thinking, coordinating resources from all parties to ensure decision implementation, taking into account long-term development and sustainable factors, and promoting more scientific and effective localization decisions from multiple dimensions. Based on this, this study proposes the following hypotheses:

H2: Managerial cognition is positively related to localization decisions.

H2-a: Managerial cognitive complexity is positively related to localization decisions.

H2-b: Managerial cognitive centralization is positively related to localization decisions.

Institutional distance constitutes an external contextual factor for multinational corporations’ operations, profoundly influencing the formation of managers’ cognition and prompting them to adjust their information focus and processing approaches. Formal institutional distance manifests as differences in laws and regulations, driving managers to prioritize information such as compliance requirements and policy utilization and subsequently adjust their operational strategies to adapt to the institutional environment of the host country. Informal institutional distance primarily manifests in differences in cultural values, compelling managers to emphasize cultural adaptation and consumer psychology. Wu et al (2018) point out that the differences between the institutional systems of the host country and the home country affect strategic decision-making through a dual mechanism: the compliance pressure brought by institutional distance directly drives the localization adjustment of marketing strategies; and institutional differences also trigger managers’ cognitive reconstruction of market characteristics, indirectly shaping the implementation intensity of localization strategies by expanding the depth and breadth of environmental scanning. Nguyen et al (2022) define the boundary conditions of mediation effects from the perspective of dynamic capabilities, arguing that managers with high cognitive complexity can identify strategic opportunities within institutional distance and transform institutional constraints into competitive advantages; whereas cognitive focus affects the efficiency of transforming institutional pressure into strategic action by optimizing attention allocation. The mediation mechanism of managerial cognition between institutional distance and localized decision-making spans three stages: cognitive evaluation, decision-making formation, and strategy adjustment. High cognitive complexity facilitates a comprehensive identification of compliance costs and policy opportunities brought about by institutional differences, while those with low cognitive complexity tend to be limited to local information. In the decision-making stage, managers with high cognitive ability are able to integrate multidimensional information such as legal, cultural, and market factors to form comprehensive strategies, while those with low cognitive capacity tend to adopt single strategies. In strategy adjustment, managers with high cognitive complexity can promptly revise actions based on market feedback, while those with low cognitive ability often lag in response or make insufficient adjustments. Based on this, this study proposes the following hypotheses:

H3: Managerial cognition mediates institutional distance and localization decisions between home and host countries.

H3-a: Managerial cognitive complexity mediates formal institutional distance and localization decisions between home and host countries.

H3-b: Managerial cognitive complexity mediates informal institutional distance and localization decisions between home and host countries.

H3-c: Managerial cognitive centralization mediates the role of Formal institutional distance between home and host countries mediates localization decisions.

H3-d: Managerial cognitive centralization mediates informal institutional distance between home and host countries and localization decisions.

The institutional foundation view proposes that the institutional environment plays a decisive role in corporate strategy, and political affiliation serves as an important link for enterprises to embed themselves in the institutional structure, helping to regulate the relationship between institutional distance and localization strategy. With the expansion of institutional distance, the demand for localization among enterprises increases, while political connections alter the extent and execution of this demand by reducing institutional uncertainty. The resource dependence theory suggests that the degree of dependence of multinational corporations on institutional resources in the host country determines the resource-bridging function of political connections. The larger the institutional distance and the more difficult it is to obtain resources, the more significant the moderating effect of political connections. The cognitive theory of managers argues that political connections affect managers’ risk perceptions and opportunity identification regarding institutional distance, thereby indirectly regulating the influence of cognition on localized decision-making. Low-level political connections mainly provide passive adaptive regulation, while high-level political connections exhibit active shaping regulation. Sun (2012) notes that that resource orchestration theory defines the boundary conditions for such regulation: once the correlation strength surpasses a specific threshold, the efficiency of deploying strategic flexible resources within enterprises improves significantly, and the promotion effect of institutional distance on localization strategy exhibits a marginally increasing trend. The political market integration model proposed by Chen (2023) also indicates that enterprises with high-level political connections can leverage policy lobbying or engage in institutional design to transform institutional differences into competitive advantages. The fundamental difference between the two types of association is that low-level associations, as responders to institutional pressure, alleviate the negative impact of institutional distance through a linear mechanism, whereas high-level associations serve as the builder of the institutional environment, strengthening its positive impact through non-linear mechanisms, reflecting the transformation of political correlation from instrumental resources to strategic resources. The regulatory mechanism has evolved from reducing information asymmetry and providing institutional buffering to actively shaping institutions, and the regulatory effect shifts from linear increase to marginal increase. Based on this, this study proposes the following hypotheses:

H4: Firms’ political affiliation moderates the relationship between institutional distance and localization decisions in home–host country contexts.

H4-a: High levels of political affiliation strengthen the positive impact of formal institutional distance on localization decisions in home–host country contexts.

H4-b: Low levels of political affiliation weaken the positive impact of formal institutional distance on localization decisions in home–host country contexts.

H4-c: High levels of political affiliation strengthen the positive impact of institutional distance on informal institutional distance on localization decisions in home–host country contexts.

H4-d: Low levels political affiliation weaken the positive impact of informal institutional distance on localization decisions in home–host country contexts.

According to the aforementioned theoretical analysis, combined with the research hypotheses proposed, the theoretical model of this study was constructed, as shown in Fig. 1.

Fig. 1.

Theoretical model.

This study selects enterprises carrying out cross-border investment in China’s A-share market between 2015 and 2024 as the object of analysis, verifies the enterprise directory based on the China Stock Market & Accounting Research Database (CSMAR) Overseas Investment Database, and performs a strict sample screening procedure: it excludes Special Treatmen (ST)-type listed companies with abnormal financial conditions; filters out observations where investment targets are located in tax-avoidance economies (e.g., Hong Kong, China, and the Cayman Islands); prioritizes manufacturing firms with significant embeddedness within the host country’s institutional network (covering 12 sub-industries such as high-end equipment, electronics manufacturing, chemicals, etc.), while controlling for sub-type differences through industry dummy variables in the model. relies on the entry mode classification of CSMAR’s “Overseas Direct Investment” database, retaining only firms implementing overseas expansion through cross-border mergers and acquisitions or greenfield investments, while eliminating cases of mixed strategies. For multiple investments in the same fiscal year, retains the records with the largest registered capital or the earliest transaction chronology to strengthen the identification of systemic distance directionality, and eliminates samples with missing key variables. The screened sample shows that Chinese companies invested 32.7% in Eastern European economies (covering 11 post-socialist countries such as Poland, Hungary, and the Czech Republic), while the rest were mainly in Association of Southeast Asian Nations (ASEAN) (28.1%), Organization for Economic Cooperation and Development (OECD)-developed countries (23.5%), and other regions along the “Belt and Road” route (15.7%). Positive institutional distance in Eastern European markets, due to the uniqueness of institutional transitions (e.g., regulatory disruptions, cultural reconfigurations), places more complex legitimacy pressures on managerial cognition.

The survey data collection focuses on the top decision-making level of the enterprise, and for organizations that do not have a Chief Executive Officer (CEO) position, the chairman or general manager who actually exercises the strategic decision-making function is included as the research target. Through a multi-stage screening process, a research sample of 126 valid observations was finally formed. The measurement tool was designed by integrating established scale frameworks from international mainstream literature and adapting them to the local research context. For the four key constructs—institutional distance, localized decision-making, managerial cognition, and corporate political affiliation—quantitative measures were implemented using a seven-point Likert scale, where 1 represents strong disagreement, 7 represents strong agreement, and the intermediate values capture varying degrees of agreement. The pre-test of the questionnaire showed that the internal consistency coefficients of each scale exceeded 0.85, and the reliability and validity of the measurement tools were rigorously tested, which effectively guaranteed the scientific rigor of data collection and the verifiability of the empirical findings.

The data were entered, processed and analyzed using SPSS 22.0 software (Version: 22.0, IBM, Armonk, NY, USA), and the statistical tools adopted included descriptive statistics to outline the basic characteristics of the data, correlation analysis to explore the intrinsic association between variables, and t-tests to verify the significance of differences in means.

To ensure the accuracy of subsequent test results, we first conducted a reliability and validity analysis, and the results are shown in Table 1.

Reliability analysis: The Cronbach’s alpha coefficients for each variable ranged from 0.732 to 0.798, with the highest reliability coefficient for “high level of political affiliation” (0.798) and “managerial cognitive complexity” (0.732). All coefficients exceed the critical value of 0.7, indicating that the internal consistency of the measurement items of the variables—including formal institutional distance, informal institutional distance, managerial cognitive centralization, managerial cognitive complexity, high-level political association, low-level political association, and localization decision-making—is good and that the reliability of the scale meets the requirements of the study.

Analysis: The Kaiser-Meyer-Olkin (KMO) values range from 0.747 to 0.813, with the highest KMO values for “localization decision-making” (0.813) and “managerial cognitive focus” (0.747), both greater than the standard value of 0.7, indicating a strong correlation between the variables, suitable for factor analysis. The Significance (SIG) value of Bartlett’s sphericity test is 0.000 (p 0.05), indicating that the correlation coefficient matrices of the variables are significantly different, and the validity of the factor analysis is confirmed, showing that the measurement tool effectively reflects the theoretical concepts of the variables.

In this study, the issue of common method bias was systematically assessed using Harman’s one-factor test, with exploratory factor analyses of all measurement items conducted through principal component analysis without rotation. The testing process followed a strict procedure: first, a complete set of indicators containing the key variables of institutional distance, managerial cognition, and localization decisions was created, the factors were extracted, and the variance explained by the largest single factor was calculated. As seen in Table 2, the variance explained by the first factor extracted (the largest single factor) without rotation is 29.480%, which does not exceed the critical value of 40%, indicating that there is no serious common method bias in this study.

As seen in Table 3, the results of the descriptive statistics of the variables show that the distribution of the values of the core research variables is characterised by significant differences. The institutional distance indicator has a mean value of 1.122 units, a standard deviation of 0.534, a minimum value of 0.123, and a maximum value of 3.876, showing significant heterogeneity in the institutional environments faced by the sample firms in different national markets. The intensity of localisation decisions has a mean value of 2.263, a standard deviation of 0.892, and a range of values from 0.345 to 4.123, reflecting a significant divergence in the strategic adaptability of firms in host country markets. The managerial cognitive score has a mean value of 0.287, a standard deviation of 0.145, and a range of values covering –0.456 to 0.987, suggesting directional differences in decision-makers’ ability to interpret institutional environments. Firms’ political affiliation intensity has a mean value of 0.123, a standard deviation of 0.067, and a range of values from 0.012 to 0.345, indicating a hierarchical distribution of political embeddedness among the sample firms.

As seen in Table 4, the results of the correlation coefficient analyses show that different degrees of statistical correlation exist among the study variables. The localisation decision is significantly and positively correlated with firm size (0.273), firm R&D investment (0.367), and home government subsidies (0.243); weakly correlated with institutional distance (0.026); significantly and positively correlated with managerial cognition (0.032) and with political affiliation of the firm (0.065) but with a smaller coefficient; and significantly and negatively correlated with the age of the firm (–0.056). Institutional distance is highly positively correlated with the degree of economic development of the host country (0.875) and significantly negatively correlated with the size of the host market (–0.217). Managerial cognition is significantly and positively related to corporate political affiliation (0.042). Firm political affiliation is significantly and positively related to host country market size (0.352) and host country economic development (0.478). Firm size is significantly and positively related to firm R&D investment (0.773) and host country market size (0.378). Firm R&D investment is highly positively related to firm size (0.773). Home country government subsidies are significantly and positively related to firm size (0.486). Host country market size is highly positively correlated with host country economic development (0.784). Home country gross domestic product (GDP) is significantly and positively related to firm political affiliation (0.436***).

As seen in Table 5, according to the analysis of the Variance Inflation Factor (VIF) values in the table, the VIF values of institutional distance (12.581), firm size (10.362), firm R&D investment (9.728), and home country GDP (13.284) in the first test exceeded or were close to 10, which indicates that there was a serious problem of multicollinearity. Subsequently, these variables were processed (excluding home country GDP and taking logarithmic values for firm size), and the second VIF was calculated. The VIFs of the relevant variables decreased significantly to a safe range (5) after processing, indicating that the multicollinearity was effectively resolved through variable processing, with VIFs of the retained variables below the threshold, making the results of the subsequent analyses based on the model more reliable and able to accurately explore localisation decision-making, institutional distance and other variables.

As seen in Table 6, the results show that the direct paths of influence between the variables meet the requirement of statistical significance. The standardized path coefficient of institutional distance on localization decision is 0.298 (Standard Error (S.E.) = 0.097, Critical Ratio (C.R.) = 3.072, p = 0.002), which indicates that for every one standard deviation increase in institutional distance between the host country and the home country, the intensity of the enterprise’s localization decision increases by 0.298 standard deviations, confirming that there is a significant positive driving effect of institutional distance on localization decisions. The path coefficient of managerial cognition on localization decisions is 0.364 (S.E. = 0.139, C.R. = 2.619, p = 0.009), revealing that managerial cognition facilitates such decisions, and the strength of its influence exceeds the direct effect of institutional distance. The significance tests of the path coefficients were confirmed using both the critical ratio (C.R. $>$1.96) and probability value (p 0.05), and the robustness of the estimated parameters of the model was statistically supported, providing empirical evidence for the theoretical framework of the dual roles of institutional distance and managerial cognition.

As seen in Table 7, the results of the hypothesis testing are as follows: the standardised coefficient of formal institutional distance on localisation decisions is 0.178 (t = 2.753, p = 0.043 0.05), indicating a significant positive influence of formal institutional distance on localisation decisions and supporting hypothesis H1-a. The standardised coefficient of informal institutional distance on localisation decisions is 0.295 (t = 3.937, p = 0.001 0.001), indicating a significant positive effect of informal institutional distance on localisation decisions, and supporting hypothesis H1-b. the standardised coefficient of managerial cognitive complexity on localisation decisions is 0.276 (t = 2.344, p = 0.002 0.01), confirming a significant positive effect on localization decisions and supporting hypothesis H2-a. Finally, managerial cognitive centralization records a standardized coefficient of 0.301 (t = 3.082, p = 0.003 0.01), also indicating a significant positive impact and supporting hypothesis H2-b.

As seen in Table 8, the bootstrap method was used to test the mediating effect of managerial cognition between institutional distance and localisation decision, and the results showed that: the mediating effect value of institutional distance $\to$ managerial cognition $\to$ localisation decision corresponding to H3 was 1.423, with a standard error of 0.078, a bias-corrected 95% confidence interval of [0.923, 1.673], and a percentile 95% confidence interval of [0.875, 1.712], and as none of the confidence intervals contain 0, this indicates that managerial cognition plays a significant mediating role between institutional distance and localisation decision, and H3 is established; the mediating effect value of formal institutional distance $\to$ managerial cognitive complexity $\to$ localisation decision in H3-a is 0.798, with a standard error of 0.073, a bias-corrected 95% confidence interval of [0.573, 0.934], and a percentile 95% confidence interval of [0.612, 1.064], which does not contain 0, thereby confirming the significant mediating effect of formal institutional distance on localisation decisions through managerial cognitive complexity, thus supporting H3-a; the effect value of informal institutional distance $\to$ managerial cognitive complexity $\to$ localization decision in H3-b is 0.627, with a standard error of 0.068, bias-corrected 95% confidence interval [0.491, 0.823], and a percentile 95% confidence interval [0.556, 0.977], which does not contain 0, indicating a significant mediating effect of informal institutional distance on localization decision through managerial cognitive complexity, and supporting H3-b. The effect value of formal institutional distance $\to$ managerial cognitive centralization $\to$ localization decision in H3-c is 0.680, with a standard error of 0.069, bias-corrected 95% confidence interval [0.342, 0.735], and percentile 95% confidence interval [0.492, 0.813]. The interval does not cover 0, suggesting that formal institutional distance plays a significant role in localization decisions through managerial cognitive centralisation, thus supported by H3-c; the effect value of informal institutional distance $\to$ managerial cognitive concentration $\to$ localization decision in H3-d is 0.598, with a standard error of 0.067, bias-corrected 95% confidence interval [0.387, 0.837], and percentile 95% confidence interval [0.497, 0.901], and the interval does not encompass 0, confirming that informal institutional distance has a significant mediating effect on localization decisions through managerial cognitive centralization, and H3-d holds.

As seen in Table 9, the standardised coefficient of “firm’s political affiliation and institutional distance $\to$ localisation decision” in H4 is 0.276, and the C.R. value is 2.673, p = 0.014 0.05, indicating that the interaction of political affiliation and institutional distance has a significant positive effect on the localisation decision, i.e., the combination of the degree of political affiliation and institutional distance can enhance the tendency of enterprises to make localization decisions. In the sub-hypothesis, the standardised coefficient of H4-a is 0.036, p = 0.034 0.05, which shows that under high levels of political affiliation, formal institutional distance has a weak positive effect on the localisation decision, probably because high levels of political affiliation help firms to alleviate the pressure of differences in the formal system and promote localisation adjustment; the standardised coefficient of H4-b is –0.021, p = 0.039 0.05, which indicates that formal institutional distance has a negative impact on localisation decisions at low levels of political affiliation, and the lack of political resources makes it difficult for firms to cope with the differences in the formal system, thus inhibiting the localisation decisions; H4-c has a standardised coefficient of 0.027, p = 0.031 0.05, which indicates that informal institutional distance has a weak positive impact on localisation decisions at high levels of political affiliation. H4-c has a standardised coefficient of 0.027, p = 0.031 0.05, indicating that at high levels of political affiliation, informal institutional distance has a weak positive effect on the localisation decision, which may be due to the fact that the resources brought by high levels of political affiliation indirectly help firms partially adapt to the differences in the informal system; the standardised coefficient of H4-d “low levels of political affiliation $\times$ informal institutional distance $\to$ localisation decision” is –0.019, p = 0.029 0.05, indicating that under low levels of political affiliation, informal institutional distance has a negative effect on localisation decision, and due to the lack of political affiliation and limited resources, it is difficult for enterprises to adapt to the differences in the informal system, thus hindering the localisation decision.

This study makes a threefold theoretical breakthrough in the intersection of institutional distance theory, strategic management theory and cognitive theory, and provides an innovative analytical framework for the study of MNCs’ localization decision-making. The specific theoretical contributions are as follows:

This study is based on the real needs of multinational corporations’ global marketing strategies, combining the heterogeneity of institutional environments with the dynamic interaction of managers’ cognition to provide multidimensional practical guidelines for multinational corporations to optimize their localization decisions:

Although this study has made some progress at the level of theoretical construction and empirical analysis, there are still some limitations that must be addressed in subsequent studies: the research sample focuses on multinational investment enterprises in China’s A-share market and is dominated by the manufacturing industry, and the industry concentration and geographic specificity of the sample may limit the generalizability of the conclusions. There are significant differences in the sensitivity of different industries to institutional distance, and the localization decision-making mechanisms of service-oriented or high-tech enterprises may present different characteristics.

Future research can deepen the theoretical and practical exploration of MNCs’ localization decisions in the following directions: expanding the sample coverage to include MNCs from different industries and geographic regions and comparing the differentiated impact mechanisms of institutional distance on localization decisions. For example, future studies could explore the strategic response patterns of service-oriented firms under high cultural cognitive institutional distance or analyze the localization paths of high-tech firms in markets with significant differences in technological standards. Researchers may also adopt a longitudinal research design to track the long-term localization practices of MNEs in host markets and examine the synergistic evolution of institutional distance and managerial cognition. The focus should be on analyzing the dynamic adjustment process of managers’ cognitive schema and its impact on strategic decision-making in the context of rapid changes in the institutional environment (e.g., drastic policy changes, technological revolutions).