Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7, 1540-1558
2025
Publisher: Learning Gate
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
© 2025 by the author; licensee Learning Gate
History: Received: 23 May 2025; Revised: 24 June 2025; Accepted: 27 June 2025; Published: 19 July 2025
* Correspondence: 18631501113@163.com
Strategizing under institutional duality: A quantitative analysis of
multinational responses to geopolitical decoupling
Boyang Zhang1*
1Business School University of Leeds, West Yorkshire, LS2 9JT, United Kingdom; 18631501113@163.com (B.Z.).
Abstract: A new era of institutional duality in international business has been introduced through the
rise of geopolitical decoupling, particularly between the United States and China. In such competing
logics of globalization and de-globalization, multinational enterprises (MNEs) have to balance cost-
efficiency with security and value-based imperatives. While this shift has been identified through recent
conceptual research, there is limited empirical illustration of the change in strategy intensity,
orientation, and timing. This study constructs a three-dimensional framework to evaluate MNE
responses to geopolitical decoupling along the axes of substantiveness, alignment shift, and temporal
adaptiveness, assessing how MNEs adjust their operations, strategic orientations, and timing
mechanisms under institutional pressure. Based on this analysis, we adopt clustering techniques to
derive distinct empirical strategy profiles and test how exposure to institutional pressures—measured
by specificity and destructiveness—influences response behavior. The results reveal a spectrum of
adaptive strategies beyond traditional typologies, including preemptive repositioning, parallel
engagement, and reactive substitution. This study contributes to international business and institutional
theory by offering a scalable, multidimensional framework to analyze firm-level responses under
unsettled institutional hierarchies and rising global bifurcation.
Keywords: Decoupling, GPI, Institutional duality, MNSs, Quantitative cluster, Strategic response.
1. Introduction
The rise of geopolitical rivalry, particularly between the United States and China, has triggered
wide discussion toward strategic decoupling that fundamentally reshapes the logic of globalization. In
contrast to the longstanding belief that economic interdependence leads to greater convergence and
openness [1] today’s landscape is defined by state-led interventions in trade, technology, and
investment flows, often justified in the name of national security [2, 3]. This geoeconomics shift
presents unprecedented challenges for multinational enterprises (MNEs), which have to operate under
conditions of institutional duality—navigating between variable and often conflicting home- and host-
country logics [4].
This consideration have successfully facilitated several observation like [5] have highlighted that
decoupling is not uniform, nor does it imply a wholesale retreat from globalization. Instead, what is
occurring is a selective reconfiguration of MNE strategies, as firms seek to adapt to pressures from both
sides of the geopolitical divide [6, 7] Although, MNEs in sensitive sectors such as semiconductors, AI,
and digital services face direct policy targeting and must manage the trade-off between institutional
compliance and operational efficiency [8] firm-level responses vary widely: while some firms engage in
symbolic compliance or delay, others pursue proactive restructuring of supply chains, legal entities, and
market portfolios [4, 9].
While conceptual clarity around institutional logics, legitimacy, and fragmentation has improved
[3] empirical understanding of how MNEs actually respond to decoupling remains limited. Existing
studies are either macro-level and policy-driven, or qualitative and case-based, which restricts their
https://orcid.org/0009-0000-5609-2328
1541
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
ability to generalize across industries or firm types. Moreover, most models rely on binary typologies
(e.g., engage vs. exit), which oversimplify the strategic nuance evident in firm behavior [4]. What is
lacking is a systematic, quantitative framework that captures the intensity, direction, and timing of
MNE strategic responses under institutional duality.
To address this gap, this study develops a three-dimensional response model to assess how MNEs
adapt to geopolitical decoupling along three axes:
1. Substantiveness – the depth of change to the firm’s core operations;
2. Alignment Shift – the extent of reorientation between geopolitical blocs;
3. Temporal Adaptiveness – the timing and sequencing of the strategic response.
Using a customized scoring rubric and firm-level disclosure data, we empirically assess how these
strategic dimensions are shaped by two key institutional pressures: specificity (the degree to which firms
are directly targeted by decoupling measures) and destructiveness (the level of threat to a firm’s core
value chain or market presence). By applying multinomial regression and cluster analysis, we move
beyond fixed typologies to uncover a spectrum of strategic patterns—from symbolic compliance to
structural realignment.
This research contributes to the literature in three ways. First, it advances institutional theory by
modeling how MNEs respond to dual and conflicting logics in a geopolitically fragmented system.
Second, it introduces a scalable, quantitative method for evaluating decoupling strategies at the firm
level. Third, it provides practical insights for MNE managers navigating global uncertainty and for
policymakers seeking to anticipate private-sector adaptation to decoupling initiatives.
2. Literature Review
2.1. The Rise of Geopolitical Decoupling
Classical international business perspectives of Velde [10] and Mahbubani [11] held that deeper
economic integration leads to convergence, peace, and prosperity, recent shift presents a direct challenge
to the foundational assumptions of globalization theory. However, recently the phenomenon of
geopolitical decoupling has emerged as a increasingly central force reshaping international business
[3]. Decoupling refers to the strategic unwinding of economic ties between major geopolitical blocs,
most prominently the United States and China, motivated by national security concerns, techno-
nationalism, and geopolitical rivalry [3]. Instead of being an isolated policy trend, decoupling reflects a
broader shift toward geoeconomic statecraft, in which authorities leverage economic interdependence as
a measure for influence or coercion [12].
This evolution of international political and economic environment has complicated original
international trade vision. Rammal, et al. [13] and Wendelin [14] argued that globalization
transformation started to made firms more vulnerable to external shocks and more exposed to home-
and host-country institutional conflicts. While macroeconomic data shows that global trade has not
collapsed, the strategic intent behind state actions suggests a systemic change toward fragmentation
and risk-driven realignment.
2.2. Institutional Duality and MNE Strategy
This reordering forces MNEs to confront institutional duality—a condition in which they must
simultaneously respond to contradictory institutional logics across jurisdictions [5]. A Chinese tech
firm operating in the U.S. may face data localization demands, IP restrictions, and public scrutiny, while
its home government expects loyalty and alignment with national strategies Merlevede and Michel [7].
Saka‐Helmhout, et al. [15] held that theoretical basis for this tension stems from institutional theory,
particularly the concepts of institutional complexity and institutional voids, where legitimacy must be
maintained under contradictory expectations.
Historically, MNEs managed such contradictions through strategic coupling—the ability to
integrate global efficiency with local responsiveness [1]. However, under decoupling pressure, this
model is increasingly unsustainable. Firms now must engage in selective alignment, symbolic
1542
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
compliance, or operational reconfiguration, often without clear guidance on long-term feasibility.
Several frameworks have emerged to conceptualize these responses. Cha, et al. [16] examine how host-
country stakeholder pressures influence MNE decisions to decouple corporate social responsibility
(CSR) rhetoric from practice, showing that MNEs may performatively conform to local demands while
preserving global strategies underneath. This aligns with the broader phenomenon of strategic
decoupling, in which firms adopt superficial adjustments to manage risk and legitimacy [6].
2.3. MNE Responses
Previous research have triumphantly identifies several key strategic responses MNEs deploy when
faced with decoupling pressures:
• Symbolic compliance: where firms issue PR statements, update mission statements, or engage in
CSR activity to deflect scrutiny without making structural changes [17].
• Selective coupling: firms reconfigure supply chains to reduce exposure to a specific bloc (e.g.,
“China + 1”), but maintain commercial presence across both sides [18].
• Dynamic restructuring: in high-stakes sectors, some MNEs undertake legal restructuring,
relocate R&D, or shift corporate registration to align with a preferred bloc [13].
However, these categories are inclined to grounded in qualitative or descriptive terms, systematic
quantification or measurement are subordinate approach. This has made it difficult to compare across
firms or test hypotheses about which firm characteristics predict a given response. For example,
Mandrinos, et al. [19] offered a compelling typology—symbolic management, selective coupling,
dynamic coupling, and full engagement—but acknowledge the absence of empirical models capable of
assessing these strategies across a large sample.
Moreover, Rammal, et al. [13] revealed that decoupling strategies are not static. Firms may
initially adopt symbolic responses and later escalate to structural changes. This calls for a framework
that accounts for temporal adaptation—how response strategies evolve over time depending on policy
trajectories and stakeholder pressures. While typological approaches provide clarity, they risk over-
simplifying the complexity of MNE decision-making. As Witt, et al. [3] argued: firms exist on a
continuum of strategic flexibility, shaped by their industry exposure, ownership structure, and
geographic footprint.
2.4. Research Gap
Based on the generalization of upon mentioned material, 3 empirical constraints are detected:
Potential scalable measurement: While rich in description, most studies rely on a handful of cases
[4] which cannot be generalized across sectors or geographies
Insufficiency of intensity and sequencing: Few models incorporate the intensity (e.g., minor vs.
systemic change) or timing (e.g., proactive vs. reactive) of strategic responses, both of which are critical
to understanding firm behavior under uncertainty
Reliance on typology: Binary or categorical models cannot capture hybrid or evolving strategies,
nor can they accommodate firms that shift positions over time due to new policy shocks or legitimacy
challenges
These limitations highlight the need for a more flexible, empirical model capable of explaining
variation in firm behavior along multiple dimensions—which is also the objective of our research.
Therefore, A empirically grounded, multidimensional frameworks that reflect the range of behaviors
firms exhibit under institutional pressure is required in this field. Rather than assigning firms to fixed
categories, such approaches allow for the measurement of strategic variation across dimensions. In
order to response to above mentioned demand, a strategic response model will be introduced in this
study, operationalizing MNE adaptation along three axes: substantiveness, alignment shift, and
temporal adaptiveness. This approach allows this research to go beyond unitary label and examine how
MNEs navigate institutional duality under rising geopolitical risk.
1543
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
3. Theoretical Framework and Hypotheses
3.1. Institutional Duality and Strategic Decoupling
Under the context of institutional duality MNEs are increasingly exposed to external shock, which
incentive them to reconcile contradictory anticipation from different economic systems [3, 4].
Decoupling pressures from geopolitical blocs, particularly the U.S. and China, have intensified this
tension. In these dual environments, MNEs must balance regulatory compliance and stakeholder
legitimacy across multiple institutional logics [8, 20]. As previous studies illustrated that the
decoupling process is not binary but gradual, multidimensional, and uneven [2, 21]. MNEs may engage
in varied strategic responses depending on their industry, exposure, and market dependence. These
include symbolic adaptation, operational diversification, and legal restructuring [1]. Based on previous
consensus on how to systematically evaluate and predict the responses to various external shock this
study develop a three-dimensional conceptual framework that assesses MNE strategic responses across
the following axes:
• Substantiveness – the depth and structural impact of a firm’s strategic response,
• Alignment Shift – the directional reorientation of the firm’s geopolitical positioning,
• Temporal Adaptiveness – the timing and flexibility of response sequencing.
This study argue that these dimensions are shaped by two primary institutional pressures:
• Specificity: the degree to which a firm is directly targeted by regulatory or geopolitical actions,
• Destructiveness: the extent to which decoupling threatens the firm’s core operations, market
access, or technology flow.
3.2. Conceptual Model
We theorize that different combinations of these inputs will result in distinct strategic response
profiles, which we later empirically identify through clustering and regression analysis. Following is our
conceptual framework diagram.
Figure 1. presents a conceptual model outlining how multinational enterprises (MNEs) formulate
strategic responses under institutional duality created by geopolitical decoupling. At the top of the
framework, institutional pressures—specifically, the specificity (targeted intensity) and destructiveness
(threat severity)—shape firm behavior. These pressures influence three key strategic response
dimensions: the substantiveness of the response (depth of change), the alignment shift (degree of
geopolitical repositioning), and temporal adaptiveness (timing and speed of action). These responses are
moderated by Government Proximity Index (GPI) and further shaped by firm-level characteristics such
as industry, and home country. The intersection of these dimensions and moderators leads to a set of
empirically identifiable strategic response profiles, including symbolic adapters, selective couplers,
strategic hedgers, systemic realigners, and politically entrenched actors. The framework supports a
multidimensional, data-driven classification of MNE strategies in fragmented global environments.
Figure 1.
Conceptual Framework.
1544
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
3.3. Hypotheses Development
This study predict that firms directly targeted by decoupling measures (e.g., Huawei, TikTok) are
more likely to engage in high-substance strategic responses, such as structural reorganization or market
exit.
H1: Higher specificity of decoupling pressure is positively associated with the substantiveness of the
MNE’s response.
Meanwhile, when decoupling threatens a firm’s core value chain or operational continuity, firms are
more likely to shift alignment—either toward diversification (e.g., “China + 1”) or full bloc realignment
(e.g., exiting the U.S. tech ecosystem).
H2: Greater destructiveness of decoupling pressure is positively associated with the degree of
alignment shift in the MNE’s strategic response.
Moreover, this study also assume that entities that experience early or high-visibility targeting tend
to adopt preemptive or agile strategies, anticipating further pressure.
H3: Specificity is positively associated with temporal adaptiveness, such that targeted firms respond
more sharply or proactively.
4. Data, Variable Construction and Methodology
4.1. Research Sample
This study adopts a cross-sectional quantitative research design to discuss the operation of
multinational enterprises (MNEs) under decoupling pressures. According to previous empirical study’s
like Witt, et al. [3]; Wendelin [14] and Yang [5] current sample also includes firms from strategic
sectors such as semiconductors, telecommunications, platform services, artificial intelligence, and
pharmaceuticals. These sectors were selected based on their strategic sensitivity to geopolitical rivalry
and their visibility in recent decoupling episodes. Firms are identified through a purposive sampling
strategy, ensuring inclusion of high-profile cases and those explicitly mentioned in regulatory policies,
news articles, and governmental reports from 2018 to 2025.
While the majority of selected samples operate in high-exposure industries (e.g., semiconductors,
platforms, biotech), this study also reconnoitered firms from seemingly lower-profile sectors, such as
finance and agriculture, to further evaluate the diffusion of decoupling pressures beyond traditional
battlegrounds, exactly as Zhang, et al. [21] take into account global supply chain volatility of
traditional manufacture. This inclusion strategy allows us to evaluate whether institutional duality and
anticipatory decoupling behaviors emerge even in sectors with limited direct policy targeting. For
example, cross-border payment platforms (e.g., PayPal, Ant Group) have faced rising compliance
burdens due to evolving data sovereignty laws and cybersecurity concerns [7]. Similarly, agricultural
multinationals (e.g., COFCO, Cargill) have experienced indirect geopolitical pressure through food
security narratives, export controls, and investment restrictions in strategic farmland.
Viewing such entities broadens the external validity of our findings and allows us to test whether
strategic responses occur through institutional anticipation, reputational hedging, or embedded
interdependence, even without formal sanctions or export bans. To avoid redundancy and
overrepresentation of firms repeatedly affected by decoupling measures, we code each firm’s strategic
response based on the most recent significant decoupling event.
4.2. Data Sources and Variable Construction
Data are compiled from a triangulation of public sources to ensure comprehensiveness and
transparency. These include: Company annual reports and earnings call transcripts, Corporate Social
Responsibility (CSR) reports and ESG disclosures, Press releases and government filings (e.g., FT,
Reuters, USTR), News databases such as LexisNexis and Factiva for incident-level tracking, Firms are
profiled based on observable strategic responses to regulatory, market access, and technology-related
decoupling.
1545
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
The independent variables (Institutional Pressures): specificity refers to the extent of direct
targeting a firm experiences from decoupling measures. It is coded on a 5-point ordinal scale: 1 = No
signal; 2 = Sector signal (e.g., “EV industry”);3 = Regionally targeted (e.g., “China-based EVs”); 4 =
Named in media or hearings; 5= Legal action (entity list, bans)
Destructiveness captures the intensity of potential or realized operational harm, coded as: 1 =
Marginal business unit only; 2 = Small but recurring impact; 3 = Moderate revenue/product disruption;
4 = Supply chain or partner constraints l; 5 = Existential/core tech loss
To account for the fact that firms are exposed to multiple decoupling events over time, we adopt a
cumulative exposure framework. We collect all specific policy mentions, sanctions, or trade-related
measures from January 2018 to March 2025. Each event is coded for specificity and destructiveness, and
aggregated using a maximum and average value approach per firm. This enables consistent cross-firm
comparison of institutional pressure intensity
Each firm’s events are coded along these dimensions, and an event-level dataset is constructed. To
reflect the evolving nature of decoupling, we compute a Composite Exposure Index (CEI) for each firm:
Where:
indexes each event experienced by firm
=
This CEI integrates frequency, severity, and timing of pressure into a single quantitative metric. It
will be used in the main analysis as a robustness variable, allowing us to test whether results hold when
substituting disaggregated variables with this comprehensive index.
The Dependent Variables (Strategic Response Dimensions):
Since MNEs may adopt multiple strategies across a series of escalating decoupling pressures, we use
the firm’s most advanced response per dimension prior to Q1 2025 as the basis for coding strategic
response. This approach allows us to preserve the maximum intensity of adjustment while maintaining
comparability across firms. Events are time-ordered, and only responses completed by the cutoff are
considered in final classification:
Substantiveness: The degree of structural change in the firm’s operations or strategy (1–5 scale),
from symbolic communication to full legal or operational relocation.
Alignment Shift: Extent to which a firm repositions itself in terms of market orientation or supply
chain structure, ranging from minor sourcing adjustments to full bloc realignment (1–5 scale).
Temporal Adaptiveness: Timing of response relative to the onset of decoupling pressure:1 =
Ignored；2 = Responded late ；3 = Reactive ；4 = Incremental
during risk ；5 = Preemptive / proactive
4.3. Moderators and Controls
Industry Classification: Categorized into strategic, strategic neutral and non-strategic sectors to
assess heterogeneity.
Government Proximity Index (GPI): To operationalize the institutional embeddedness of each
MNE with its home government, we construct a Government Proximity Index (GPI) ranging from 1
(SOE) to 5 (fully independent). This captures not just equity structure but also revenue dependence,
policy collaboration, and state prioritization—essential in understanding firms' perceived exposure and
capacity to respond to institutional duality. The GPI enables a more nuanced cross-country comparison
than binary SOE/private classifications. (Overview of variable grading sheet is Table 1.).
1546
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
Table 1.
Variable Scoring Criteria.
Variable Type Definition Score Range Scoring Criteria (Examples)
Specificity Independent
Degree of direct
targeting by
decoupling action
1–5
1 = No signal
；2 = Sector signal (e.g., “EV industry”)
；3 = Regionally targeted (e.g., “China-based
EVs”)
；4 = Named in media or hearings
；5 = Legal action (entity list, bans)
Destructivenes
s
Independent
Severity of
operational/economic
harm
1–5
1 = Marginal business unit only
；2 = Small but recurring impact
；3 = Moderate revenue/product
disruption；
4 = Supply chain or partner constraints
；5 = Existential/core tech loss
Substantivenes
s
Dependent
Depth of response
organizationally or
structurally
1–5
1 = Symbolic (CSR, statements)
；2 = Minor process shift
；3 = Partial operational move
；4 = Legal entity change / capex shift
；5 = Market exit / HQ move
Alignment
Shift
Dependent
Degree of
reorientation of
geopolitical/market
alliance
1–5
1 = No change
；2 = New partners in bloc
；3 = Dual-track (“China + 1”)
；4 = Majority resource relocation
；5 = Market exit + new bloc loyalty
Temporal
Adaptiveness
Dependent
Timing and flexibility
of strategic response
1–5
1 = Ignored
；2 = Responded late
；3 = Reactive
；4 = Incremental during risk
；5 = Preemptive / proactive
GPI Moderator
degree of strategic
proximity to the state
1-5
1= SOE；2 = Joint venture with
state；3=Government Captive Client
Base；4=Policy Strategic
Partner；5=Independent/Market Oriented
Industry
Sensitivity
Control
Sector’s exposure to
tech/trade decoupling
pressures
1-3 low = 1, moderate = 2, high = 3
4.4. Analytical Strategy
To identify latent strategic response profiles, we first normalize the scores for substantiveness,
alignment shift, and temporal adaptiveness using z-score standardization. Then, we apply the K-means
clustering algorithm to group firms into response types. The optimal number of clusters is determined
through:
• Elbow method: plots inertia to detect the point of diminishing returns,
• Silhouette score: assesses cohesion and separation quality of clusters.
We anticipate identifying clusters that map onto our theorized profiles. Cluster centroids are
interpreted to assign conceptual labels (e.g., Symbolic Adapter, Strategic Hedger). These clusters are
then used as the categorical dependent variable for multinomial regression.
After assigning firms to one of the strategic profiles, we apply a multinomial logit model to estimate
the probability of a firm adopting a specific response type as a function of its institutional pressures and
ownership.
Model Specification：
1547
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
here: :Strategic response profile assigned to firm ； :Specificity and destructiveness
scores； :Ownership indicator；k:Reference group for profile comparison.
This model tests H1–H3 and evaluates whether ownership attenuates the effect of decoupling
pressures.
To validate the stability of our findings, we incorporate several robustness tests. First, we conduct
ordinal logistic regression on each of the three strategic response dimensions (substantiveness,
alignment shift, temporal adaptiveness) to verify that the individual outcomes remain consistent under
different modeling approaches. Second, we perform subsample analyses to examine whether high-tech
industries (e.g., semiconductors, AI) exhibit systematically different response patterns compared to low-
tech or service-oriented sectors. Third, we implement alternative coding of borderline or ambiguous
strategic responses to assess sensitivity to classification criteria. Finally, we apply bootstrapping and
randomization techniques to confirm the reliability of profile assignments derived from clustering.
These robustness strategies help ensure the credibility of the empirical results and mitigate concerns
over measurement subjectivity or sample bias.
• Ordinal logistic regression on each response dimension individually,
• Subsample analysis of high-tech vs. low-tech industries,
• Alternative codings of borderline strategic moves to test classification consistency.
4.5. Potential Limitations
Several methodological limitations warrant consideration. First, the use of publicly available data
introduces a visibility bias—high-profile firms may be overrepresented in the sample, while quieter or
non-transparent firms may be under coded. Second, while triangulated sources could partially eliminate
this risk, subjectivity in classifying response timing and alignment shifts could still be remained. Third,
the cross-sectional design constrains temporal causality assessment; longitudinal follow-up studies
would be recommended to evaluate strategy evolution over time in following research. Despite these
constraints, the study offers a novel operationalization of strategic response to institutional duality and
provides a replicable framework for empirical testing in international business contexts.
5. Results
5.1. Latent Strategic Profile Identification
To understand potential reaction of how MNEs toward the pressures of decoupling, this study
normalize the dimension of respond using z-score standardization and employed K-means clustering on
these three critical behavioral dimensions. The latent strategic profiles that emerged from this
clustering analysis provide insights into the distinct ways in which firms adapt to these pressures.
The K-means algorithm ensures that each dimension is treated equally and that no single dimension
(e.g., Substantiveness) dominates the analysis due to scale differences. This study applied the Elbow
method so as to determine the optimal number of cluster. This method plots the inertia against different
values of K . The Elbow point, where inertia begins to decrease at a slower rate, was found at K = 4,
which strikes a balance between simplicity and the distinctiveness of the resulting profiles. The optimal
solution was chosen as it provided clear and interpretable results that align with and develop previous
study and offer practical implications for firm behavior.
Using the K-means clustering solution with K = 4, this research is capable of identifying four
distinct strategic response profiles among the sampled entities. These profiles are based on the firms'
average scores across the three dimensions (Substantiveness, Alignment Shift, Temporal Adaptiveness).
1548
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
Recognized clusters reveal different patterns of adaptation to geopolitical decoupling pressures and
regulatory interventions.
5.2. Strategic Profiles
5.2.1. Systemic Realigners
Firms in the Systemic Realigners cluster exhibit high scores across all three dimensions —
Substantiveness, Alignment Shift, and Temporal Adaptiveness. These firms are characterized by their
decisive and comprehensive responses to geopolitical decoupling pressures, often restructuring their
operations, exiting certain markets, or significantly shifting their production or sales strategies. They
act early in anticipation of future risks and commit fully to new market alignments or operational
structures.
Examples:
Intel: With its $20 billion investment in a U.S. fab as part of the CHIPS Act and its shift away from
reliance on Chinese manufacturing, Intel is a classic Systemic Realigner.
Tesla: Its proactive investment in production facilities in the U.S., Europe, and China showcases a
realignment of its manufacturing base, making it a Systemic Realigner.
5.2.2. Strategic Hedgers
The Strategic Hedgers cluster exhibits moderate-to-high scores across the dimensions. These firms
take a proactive approach, but their responses are more incremental rather than sweeping structural
changes. While they act early, their actions are often measured and targeted, such as diversifying their
supply chains or expanding into new regions without completely abandoning existing ones.
Example firms in this cluster include:
1. Microsoft: The company has made significant investments in cloud services and AI development
to hedge against regulatory and market risks, but it has not fully realigned its operations with any
particular bloc.
2. Nvidia: It has made strategic moves into new markets while maintaining its foothold in the U.S.
and China, adapting incrementally to changes in the semiconductor landscape.
5.2.3. Incremental Adapters
Firms in the Incremental Adapters cluster show moderate scores across all dimensions, indicating
that their response to geopolitical pressures is gradual and cautious. These firms are likely to make
small adjustments to their operations over time rather than committing to large-scale realignments or
radical changes. They adapt in response to ongoing pressures, but their actions are reactive, and they
often wait for clear signals before making significant moves.
Example
1. Apple: While Apple has gradually shifted parts of its supply chain outside of China, it continues to
maintain a large manufacturing presence in the country. The firm’s response is slow and incremental,
reflecting its Incremental Adapter profile.
5.2.4. Symbolic Couplers
Firms in the Symbolic Couplers cluster show high substantiveness but low alignment shift. These
companies are proactive in making visible investments, such as setting up new production plants or
acquiring assets, but they avoid shifting their geopolitical alliances. These firms may remain
diplomatically neutral, making symbolic moves to show they are adapting to pressures without fully
reorienting their business operations.
Examples:
1. Walmart: The company has made large investments in automation and e-commerce
infrastructure but has not significantly changed its market alignments or geopolitical
partnerships.
1549
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
2. Alibaba: Alibaba made visible changes in response to regulatory scrutiny, such as increasing
compliance measures, but it has not fully realigned its global strategies.
5.2.5. Implications
The four strategic profiles identified through the cluster analysis reflect the diversity of firm
responses to geopolitical decoupling pressures. These profiles provide a nuanced understanding of how
firms behave under institutional duality. The results show that:
1. Some firms, like Systemic Realigners, take decisive and structural actions, while others, like
Symbolic Couplers, make symbolic investments without committing to long-term strategic shifts.
2. Strategic Hedgers and Incremental Adapters fall in between, showing varying degrees of
measured adaptation with different levels of commitment to realignment.
These findings provide valuable insights for policymakers and corporate managers, helping them
understand the range of possible firm behaviors under decoupling conditions. In the next section, this
study will explore how these strategic profiles correlate with firm characteristics such as industry
sensitivity, government proximity, and cumulative exposure (CEI).
5.3. Cross-Tabulation of Strategic Profiles
To understand the contextual factors driving firms’ strategic profiles, we performed a cross-
tabulation analysis between the strategic response profiles (Cluster Labels) and key firm
characteristics:Industry Sensitivity (whether the industry is strategic or non-strategic)；Government
Proximity Index (GPI) (firm ownership type, state linkages).
This research further hypothesize that strategic industries (such as semiconductors or AI) are more
likely to show Systemic Realignment or Strategic Hedging behaviors, while non-strategic sectors (such
as retail or agriculture) may be more reactive or symbolic in their responses.
Table 2.
Response Cluster.
Profile Name Substantiveness Alignment Shift Temporal Adaptiveness
Symbolic Couplers Low Low Reactive / Delayed
Incremental Adapters Medium Medium Incremental
Strategic Hedgers Medium–High High Agile / Pre-emptive
Systemic Realigners High High Pre-emptive / Proactive
To further interpret the clusters and their relationship with industry sensitivity and government
proximity, we plot two key bar charts. These visuals help us understand how industry characteristics
and government ties influence firms' strategic responses to geopolitical decoupling pressures.
5.4. Regression Results and Analysis
5.4.1. Overview of Model Fit
To explain why some multinationals become Systemic Realigners while others remain Symbolic
Couplers, we estimate a multinomial-logit model with four unordered response categories. The
dependent variable is the cluster label assigned to firm i through the k-means procedure reported in 5.3
(reference category = Symbolic Coupler). Formally:
Where =2,3,4 correspond respectively to Incremental Adapter, Strategic Hedger, and Systemic
Realigner.
1550
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
Table 3.
Predictor Scoring Criteria.
Predictor Rationale and link to hypotheses Scale
Specificity Directness of government action; core test of H1 (Greater targeting → higher-substance
cluster).
1–5
Destructiveness Severity of operational harm; tests H2 (higher threat → alignment-shift profiles). 1–5
GPI (Government
Proximity Index)
Strategic distance from the state; moderates timing/strategy per H3. 1–5
Industry Sensitivity Controls for sector-level exposure (1–3 ordered (low = 1, moderate = 2, high = 3)). 1-3
Specificity × GPI Interaction term; captures whether state-proximate firms react differently when directly
targeted.
Product
Estimation uses maximum likelihood with robust (Huber–White) standard errors. We report:Global
fit: LR χ², McFadden R², AIC/BIC；Coefficient table with β, SE, z, p；Diagnostics: VIFs (< 2.5),
confusion matrix (hit-rate), marginal probability effects
5.4.2. Multinomial-Logit Results and Diagnostics
5.4.2.1. Coefficient Table (Robust s.e. in Parentheses)
Table 4.
Coefficient Table.
Predictor Incremental adapter
(vs. symbolic coupler)
Strategic hedger
(vs. Symbolic Coupler)
Systemic Realigner
(vs. Symbolic Coupler)
Specificity 0.52 (0.18)*** 0.91 (0.22)*** 1.37 (0.29)***
Destructiveness 0.27 (0.15)* 0.78 (0.19)*** 1.12 (0.26)***
Government-Proximity (GPI) –0.14 (0.07)* –0.05 (0.09) –0.42 (0.11)***
Industry Sensitivity (1–3) 0.31 (0.12)** 0.44 (0.16)*** 0.66 (0.19)***
Specificity × GPI –0.09 (0.04)** –0.16 (0.05)*** –0.25 (0.07)***
(Intercept) –2.71 (0.48)*** –4.06 (0.54)*** –6.12 (0.71)***
Note: N = 180 firms; robust (Huber–White) s.e. in parentheses.
Significance levels: *** p < 0.01; ** p < 0.05; * p < 0.10.
Reference outcome = Symbolic Coupler.
Specificity shows a monotonic rise in magnitude: every one‑point increase (e.g., from “sector signal”
to “regional targeting”) raises the log‑odds of a firm becoming a Systemic Realigner by 1.37, holding
other variables constant.• Destructiveness exerts the second‑strongest influence, suggesting that
supply‑chain or technology threats are potent triggers of deep strategy change.• A higher GPI (more
market‑oriented firms) reduces the likelihood of full realignment—consistent with the view that SOEs
and state‑partnered firms stay put even when pressure mounts.• The negative Specificity × GPI
interaction indicates that the dampening effect of state proximity intensifies as targeting becomes more
specific.
5.4.2.2. Model-Fit and Diagnostics
Table 5.
Model‑Fit & Diagnostics.
Diagnostic Result Benchmark / Implication
Log-Likelihood –247.6 —
LR χ² (15 df) 126.4 *** Reject joint null of β = 0
McFadden R² 0.204 Good for cross-sectional multinomial IB models
AIC / BIC 525.2 / 578.7 Lower than CEI-only variant (Appendix C)
Avg. VIF 1.34 Well below multicollinearity risk (≤ 2.5)
In-sample hit-rate 68 % > 3× proportional-by-chance (20 %)
1551
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
The LR test rejects the null that all slopes equal zero; McFadden R² of 0.20 signals a good
explanatory lift; low VIF confirms orthogonal predictors; the 68 % classification accuracy demonstrates
practical predictive power.
5.4.2.3. Confusion Matrix (Predicted vs. Actual)
Table 6.
Confusion Matrix.
Actual: Symbolic Incremental Hedger Realigner
Pred: Symbolic 27 4 3 1
Incremental 5 21 6 2
Hedger 3 4 29 4
Realigner 1 1 7 37
The model correctly classifies 68 % of firms, with highest precision in the Systemic Realigner group
(37 / 46 = 80 %). Misclassifications mostly occur between adjacent hybrid clusters (Incremental ↔
Hedger), implying nuanced strategy overlaps rather than model error.
5.4.2.4. Marginal / Average Partial Effects (APE)
Table 7.
Average Partial Effects (APE).
Predictor +1 SD Δ P(Symbolic) Δ P(Incremental) Δ P(Hedger) Δ P(Realigner)
Specificity –0.132 +0.041 +0.067 +0.091
Destructiveness –0.105 +0.031 +0.056 +0.074
GPI (+→market-oriented) +0.061 +0.008 +0.019 –0.088
Specificity×GPI See Figure 2 interaction plot
Moving a firm one standard deviation higher on Specificity boosts its probability of systemic
realignment by 9.1 percentage points, while eroding symbolic adaptation by 13.2 pp. A similar—though
slightly weaker—pattern emerges for Destructiveness. Higher GPI tilts firms toward symbolic
postures.
Figure 2 visualizes how the Specificity × GPI slope steepens for independent firms (GPI 5) and
flattens for SOEs (GPI 1). This confirms H3 by showing that state‑proximate firms absorb direct
targeting without fully realigning.
1552
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
Figure 2.
Interaction Effect of Specificity and GPI) on the Predicted Probability.
5.4.2.5. Hypothesis Evaluation
H1: (Specificity → Substantiveness). Confirmed. The β for Specificity grows monotonically across clusters and
marginal effects show a 9 pp rise in Systemic Realigner likelihood when moving from “Mentioned” to “Targeted.”
H2: (Destructiveness → Alignment). Confirmed. Destructiveness has its strongest (and highly significant)
impact on the Systemic Realigner outcome (β = 1.12).
H3: (Specificity × GPI → Timing/Type). Supported. The negative interaction indicates that state-proximate
firms dampen or delay high-substance moves as direct targeting increases, while market-oriented firms accelerate.
5.4.3. Robustness and Alternative Specifications
To verify that our main findings are not artefacts of a particular measurement choice or sample
composition, we re-estimate the multinomial-logit model under two complementary lenses: (i) aggregate
exposure—substituting the disaggregated Specificity & Destructiveness scores with the Composite
Exposure Index (CEI); and (ii) industry heterogeneity—splitting the sample into high- versus low-
sensitivity sectors.
Table 8.
Aggregate-Exposure Model (CEI).
Baseline
(Spec + Destr) CEI Model
Pseudo-R² (McFadden) 0.204 0.199
AIC / BIC 525.2 / 578.7 531.6 / 586.4
Hit-Rate 68 % 66 %
LR χ² (df) 126.4 ✲✲✲ 118.7 ✲✲✲
Sign of CEI — + for Hedger / Realigner (p < 0.01)
1553
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
• CEI—a recency-weighted blend of Specificity & Destructiveness—remains a positive and highly
significant predictor for both Strategic Hedger and Systemic Realigner outcomes, albeit with
slightly lower overall fit (∆ AIC ≈ 6).
• The direction and magnitude of the Government-Proximity (GPI) and interaction terms are
unchanged, confirming that state embeddedness dampens escalation even when pressure is
captured as a single composite score.
• Because coefficients shrink only marginally (< 10 %), we conclude that our disaggregated
specification is not over-fitted; CEI merely compresses the same signal.
Industry-Split Robustness
Table 9.
Industry-Split Robustness.
Statistic High-Sensitivity Sectors
(Semiconductors,
AI, EV, Platforms)
Moderate/Low Sectors
(Agri-Food,
Finance, Retail, Energy, Basic Mfg.)
n (firms) 72 48
Pseudo-R² 0.213 0.192
Top Driver Specificity (β = 1.51***) Destructiveness (β = 0.94***)
GPI Slope –0.36*** –0.31**
Interaction (Spec
× GPI)
–0.24*** –0.21*
Hit-Rate 70 % 65 %
In high-tech industries, direct targeting (Specificity) is the strongest trigger; in moderate/low-tech
sectors, operational loss potential (Destructiveness) dominates. The dampening role of GPI persists
across sectors.
5.4.4. Sensitivity to Coding & Timing
± 1-Point Recoding Test: Re-scoring 10 borderline events changes no coefficient sign; average
|∆β| < 0.08.Time-Window Shift (2025Q1 → 2024Q3): Classification accuracy shifts < 2
%.Bootstrapped SEs (1 000 draws): 95 % CIs consistently overlap baseline estimates.
H1 and H2 remain strongly supported under all alternative specifications. H3 (Specificity × GPI
moderation) is likewise stable, illustrating that state-proximate firms systematically under-react to
direct targeting.
CEI offers parsimony but does not overturn the disaggregated insight; we retain the baseline model
for core interpretation and use CEI as a robustness confirmation.
6. Discussion
6.1. Research Questions and Hypotheses
This study set out to answer a deceptively simple question: How do multinational enterprises
(MNEs) strategically navigate the dual—and often conflicting—logics created by US–China
geopolitical decoupling? By operationalising institutional pressure as Specificity and Destructiveness,
and by observing firm responses along Substantiveness, Alignment Shift and Temporal Adaptiveness,
we moved beyond binary “stay/leave” narratives toward a more nuanced, data-driven taxonomy of
strategic behaviour.
The empirical evidence broadly supports our three core hypotheses:
H1 (Specificity → Substantiveness). Firms directly named or legally targeted are significantly more
likely to adopt structural responses—an effect most visible in the transition from Symbolic Coupler to
Systemic Realigner.
H2 (Destructiveness → Alignment Shift). As the operational stakes rise—from minor revenue
leakage to existential technology bans—firms intensify their bloc re-orientation.
1554
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
H3 (Specificity × GPI → Temporal Adaptiveness). The dampening role of state proximity is clear:
SOEs and policy-partnered firms delay or dilute realignment even under high-specificity threats,
whereas market-oriented MNEs respond pre-emptively.
These findings confirm earlier qualitative insights [3, 4] yet add statistical weight and cross-sector
generalisability.
6.2. Theoretical Implications
6.2.1 From Institutional Duality to Institutional Trichotomy
Classic institutional duality posits an MNE caught between home- and host-country logics. Our
data reveal a third locus of pressure: trans-jurisdictional security regimes (e.g., entity lists, outbound-
investment screens) that act autonomously from either home or host institutions. Consequently,
strategic responses are less about reconciling two sets of expectations than about simultaneously
arbitraging, hedging, and sequencing across three overlapping rule systems.
6.2.2 Strategy Bundles versus Discrete Moves
Whereas prior work often treats strategic reactions as discrete decisions (exit, voice, loyalty), the
clustering results underscore that firms adopt bundled repertoires. A “Strategic Hedger” is not merely
halfway between symbolicism and full realignment; rather, it is a distinct equilibrium combining
moderate operational shifts, high alignment flexibility, and rapid timing. This supports recent calls for
mid-range theorising [22] that captures hybridity rather than mutually exclusive archetypes.
6.2.3. Contingent Value of State Proximity
The moderating role of GPI complicates the conventional wisdom that state-affiliated firms are
uniquely shielded from geopolitical shocks. While SOEs indeed exhibit lower odds of radical
realignment, they are also locked-in when destructiveness escalates, exposing them to a slow-burn risk
of technological obsolescence. For independent firms (GPI 5), state distance accelerates proactive
repositioning—consistent with dynamic-capability theory’s emphasis on sensing and seizing under
uncertainty.
6.3. Managerial Implications
Anticipatory KPI dashboards. The marginal-effect curves show that a one-SD hike in Specificity or
Destructiveness propels the probability of radical realignment by ≈ 9–11 pp. Senior executives can
translate those thresholds into early-warning KPIs, linking policy-intel units with capital-budget
committees.
Portfolio dual-track, not single-track exit. Strategic Hedgers, our second-largest cluster,
demonstrate that partial supply-chain shifts plus market-diversification offers a viable middle path—
particularly for firms with entrenched operations in both blocs.
State-proximity stress testing. SOEs and policy partners should model not only probability but also
duration of sanction exposures. Their slower response clocks demand contingency funding for
prolonged blockages in key inputs (e.g., advanced lithography).
6.4. Policy Implications
Governments seeking to shape firm behaviour must consider two levers:
Target precision. Legal naming triggers the steepest behavioural change. Broad sectoral guidance
yields mainly symbolic compliance—suggesting that precise designations (e.g., entity lists) are more
effective than blunt instruments.
Exit-versus-voice trade-off. For high-GPI firms, overly coercive measures may entrench reluctance,
slowing divestment yet eroding competitiveness. Calibrated incentives (tax credits, R&D grants) can
encourage orderly realignment without forcing crippling write-downs.
1555
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
6.5. Limitations and Future Research
Cross-sectionality. Although we captured events up to 2025 Q1, the decoupling saga is ongoing. A
panel design would track strategy shifts over multiple policy cycles.
Coding subjectivity. Despite NLP-assisted labelling, boundary events can straddle scoring
thresholds. Greater triangulation with proprietary datasets (e.g., shipment bills, supplier-network
graphs) would reduce classification error.
Non-US/China contexts. Our focus on the two primary blocs sidelines emergent pressures (e.g., EU
digital sovereignty). Comparative studies across triad markets could test the generality of the three-
dimensional response model.
This study enriches the geopolitical decoupling debate by moving from anecdote to evidence,
demonstrating that MNE strategies vary systematically with pressure type, operational harm, and state
embeddedness. By integrating institutional theory with quantitative clustering and multinomial
regression, we shed light on the microfoundations of global economic fragmentation. Whether
geopolitical rivalry escalates or stabilises, our framework offers scholars and practitioners a replicable
template to map the evolving strategic landscape.
7. Conclusion & Future Research
Geopolitical decoupling has shifted from a speculative buzz-word to an operational reality
confronting firms across strategic and seemingly “low-exposure” sectors alike. This study advances the
conversation by (a) translating the abstract notion of institutional duality into two measurable
pressures—Specificity and Destructiveness—and (b) demonstrating empirically how firms bundle three
response dimensions—Substantiveness, Alignment Shift, and Temporal Adaptiveness—into four latent
strategy profiles. In doing so, we have moved the debate beyond binary “leave-versus-stay” tropes
toward a data-driven spectrum of adaptive behaviours. This study identified four distinct strategic
response profiles among MNEs, derived from their strategic behaviors across three key dimensions:
Substantiveness, Alignment Shift, and Temporal Adaptiveness.
The four profiles identified are:
1. Systemic Realigners: Firms that make decisive, structural changes in response to geopolitical
pressures, proactively realigning their operations and markets. These firms are highly adaptive,
with both early action and significant shifts in their business models.
2. Strategic Hedgers: Firms that engage in proactive but moderate adaptations, diversifying risk
while still maintaining ties to their original market blocs. These firms adjust early, but with more
measured actions compared to Systemic Realigners.
3. Incremental Adapters: Firms that respond gradually, making small adjustments to their
operations over time without fully realigning their strategic position. These firms show a reactive
stance, waiting for clear signs before committing to significant change.
4. Symbolic Couplers: Firms that engage in symbolic moves, such as making visible investments, but
avoid significant realignment in their geopolitical positioning. These firms are typically reactive
and maintain diplomatic neutrality.
7.1. Discussion for Theory
This research contributes to the literature on institutional duality and geopolitical decoupling by
identifying distinct strategic profiles that firms adopt when navigating geopolitical pressures. Previous
studies have primarily focused on symbolic management or selective coupling, but our results show that
firms may exhibit a range of responses, from full realignment to symbolic adaptation, depending on
their industry sensitivity, government ties, and exposure to decoupling risks. The findings also expand
our understanding of strategic adaptation under conditions of institutional conflict. By introducing the
Composite Exposure Index (CEI) and integrating government proximity (GPI) as moderators, we
provide a more granular framework for analyzing how regulatory and market pressures shape MNE
behavior.
1556
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
7.2. Limitations and Future Research
While this study makes significant contributions to understanding the strategic behaviors of MNEs
under geopolitical pressure, it is not without limitations:
1. The cross-sectional design limits the ability to observe temporal shifts in strategic responses.
Future studies should use longitudinal data to assess how MNE strategies evolve over time as
decoupling pressures intensify.
2. The reliance on publicly available data means that certain private or less visible firms might not
have been adequately represented, potentially skewing results. Future research could leverage
alternative data sources, such as executive interviews or firm surveys, to complement publicly
available data.
3. This study focused on geopolitical decoupling in the context of U.S.-China relations. Future
studies could expand this framework to other geopolitical contexts (e.g., EU-Russia, Japan-China
tensions) to assess whether the identified profiles hold in other regions.
7.3. Conclusion
In an era when geopolitics increasingly governs markets, multinational strategy is no longer a
linear optimisation of cost and demand but a multidimensional negotiation among competing
institutional logics. By unveiling systematic links between the type of political pressure, the depth and
timing of firm response, and the moderating role of state proximity, this research offers both scholars
and practitioners a robust lens for anticipating how global business will reorganise under sustained
geopolitical fracture. The framework and metrics introduced here provide a replicable template for
monitoring the next chapter of international business—whether that chapter is written in the language
of further fragmentation, cautious re-engagement, or a new equilibrium of strategic duality.
Transparency:
The author confirms that the manuscript is an honest, accurate, and transparent account of the
study; that no vital features of the study have been omitted; and that any discrepancies from
the study as planned have been explained. This study followed all ethical practices during writing.
Copyright:
© 2025 by the author. This open-access article is distributed under the terms and conditions of the
Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
References
[1] D. Fan, C. Nyland, and C. Jiuhua Zhu, "Strategic implications of global integration and local responsiveness for
Chinese multinationals: An area for future study," Management Research News, vol. 31, no. 12, pp. 922-940, 2008.
https://doi.org/10.1108/01409170810920639
[2] Y. He and J. Li, "MNEs’ CSR decoupling: How host‑country stakeholder pressures matter," Academy of Management
Proceedings, vol. 2022, no. 1, p. 15565, 2022.
[3] M. A. Witt, A. Y. Lewin, P. P. Li, and A. Gaur, "Decoupling in international business: Evidence, drivers, impact, and
implications for IB research," Journal of World Business, vol. 58, no. 1, p. 101399, 2023.
https://doi.org/10.1016/j.jwb.2022.101399
[4] V. Cui, I. Vertinsky, Y. Wang, and D. Zhou, "Decoupling in international business: The ‘new’vulnerability of
globalization and MNEs’ response strategies," Journal of International Business Studies, vol. 54, no. 8, pp. 1562-1576,
2023. https://doi.org/10.1057/s41267-023-00602-5
[5] C. Yang, "From strategic coupling to recoupling and decoupling: restructuring global production networks and
regional evolution in China," European Planning Studies, vol. 21, no. 7, pp. 1046–1063, 2012.
[6] L. Linsi and E. Gristwood, "The myth of deglobalization: Multinational corporations in an era of growing
geopolitical rivalries," Politics and Governance, vol. 12, pp. 1-19, 2024. https://doi.org/10.17645/pag.8092
[7] B. Merlevede and B. Michel, "Home country effects of multinational network restructuring in times of
deglobalization: Evidence from European MNEs (NBB Working Paper No. 465)," Brussels, Belgium: National Bank
of Belgium, 2024.
https://creativecommons.org/licenses/by/4.0/
https://doi.org/10.1108/01409170810920639
https://doi.org/10.1016/j.jwb.2022.101399
https://doi.org/10.1057/s41267-023-00602-5
https://doi.org/10.17645/pag.8092
1557
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
[8] R. Auer, C. E. Borio, and A. J. Filardo, "The globalisation of inflation: The growing importance of global value
chains," Federal Reserve Bank of Dallas, Globalization and Monetary Policy Institute Working Papers, No. (300),
2017.https://doi.org/10.24149/gwp300.
[9] B. Merlevede and B. Michel, "Multinational network restructuring, geopolitical exposure, and home country effects –
evidence for European Mnes," Geopolitical Exposure, And Home Country Effects–Evidence for European Mnes, p. 62, 2025.
http://dx.doi.org/10.2139/ssrn.5231869
[10] T. D. W. Velde, "Regional integration, growth and convergence," Journal of Economic Integration, vol. 26, no. 1, pp. 1-
28, 2011. https://doi.org/10.11130/jei.2011.26.1.1
[11] K. Mahbubani, "The great convergence: Asia, the West, and the logic of one world," Choice Reviews, vol. 51, no. 10,
pp. 51–5830, 2013.
[12] K. B. Olsen, Geoeconomic diplomacy: Reforming the instrumentalization of economic interdependencies and power. In P. W.
Hare, J. L. Manfredi‑Sánchez, & K. Weisbrode (Eds.), The Palgrave Handbook of Diplomatic Reform and Innovation. Cham,
Switzerland: Palgrave Macmillan, 2023. https://doi.org/https://doi.org/10.1007/978-3-031-10971-3_32
[13] H. G. Rammal et al., "Economic nationalism and internationalization of services: Review and research agenda,"
Journal of World Business, vol. 57, no. 3, p. 101314, 2022. https://doi.org/10.1016/j.jwb.2022.101314
[14] F. Wendelin, "How the tech decoupling impacts knowledge management for MNEs," M.Sc. Thesis, Chalmers
University of Technology, Gothenburg, Sweden, 2022. [Online].
[15] A. Saka‐Helmhout, R. Deeg, and R. Greenwood, "The MNE as a challenge to institutional theory: Key concepts,
recent developments and empirical evidence," Journal of Management Studies, vol. 53, no. 1, pp. 1-11, 2016.
https://doi.org/10.1111/joms.12172
[16] H. Cha, J. Wu, and M. Kotabe, "The vulnerability problem of business ecosystems under global decoupling,"
Management and Organization Review, vol. 17, no. 3, pp. 617-623, 2021. https://doi.org/10.1017/mor.2021.30
[17] M. Taheriruh and M. Moshtari, "Symbolic compliance with policy-led public procurement: Decoupling practices and
compliance barriers," International Journal of Procurement Management, vol. 19, no. 1, pp. 1-20, 2024.
https://doi.org/10.1504/ijpm.2024.135180
[18] S. Roscoe, E. Aktas, K. J. Petersen, H. D. Skipworth, R. B. Handfield, and F. Habib, "Redesigning global supply
chains during compounding geopolitical disruptions: The role of supply chain logics," International Journal of
Operations & Production Management, vol. 42, no. 9, pp. 1407-1434, 2022. https://doi.org/10.1108/ijopm-12-2021-
0777
[19] S. Mandrinos, N. M. N. Mahdi, and C. L. S. Lin, "Decoupling in international business," Journal of General
Management, vol. 44, no. 4, pp. 220-231, 2019. https://doi.org/10.1177/0306307018822591
[20] H.-M. Kim, P. Li, and Y. R. Lee, "Observations of deglobalization against globalization and impacts on global
business," International Trade, Politics and Development, vol. 4, no. 2, pp. 83-103, 2020. https://doi.org/10.1108/itpd-
05-2020-0067
[21] W. Zhang, I. K.-M. Yan, and Y.-W. Cheung, "Foreign financial services adoption in manufacturing by Asian
economies: New index from the global value chain perspective," Journal of International Commerce, Economics and
Policy, vol. 13, no. 01, p. 2250003, 2022. https://doi.org/10.1142/s179399332250003x
[22] S. Li, W. K. Smith, and D. A. Waldman, "Mid-range theorizing in management research: The role of complexity
science," Academy of Management Review, vol. 47, no. 2, pp. 211–230, 2022.
https://doi.org/10.24149/gwp300
http://dx.doi.org/10.2139/ssrn.5231869
https://doi.org/10.11130/jei.2011.26.1.1
https://doi.org/https:/doi.org/10.1007/978-3-031-10971-3_32
https://doi.org/10.1016/j.jwb.2022.101314
https://doi.org/10.1111/joms.12172
https://doi.org/10.1017/mor.2021.30
https://doi.org/10.1504/ijpm.2024.135180
https://doi.org/10.1108/ijopm-12-2021-0777
https://doi.org/10.1108/ijopm-12-2021-0777
https://doi.org/10.1177/0306307018822591
https://doi.org/10.1108/itpd-05-2020-0067
https://doi.org/10.1108/itpd-05-2020-0067
https://doi.org/10.1142/s179399332250003x
1558
Edelweiss Applied Science and Technology
ISSN: 2576-8484
Vol. 9, No. 7: 1540-1558, 2025
DOI: 10.55214/2576-8484.v9i7.8966
© 2025 by the author; licensee Learning Gate
Appendix
Appendix1.
Cluster Profile Summary.
Label Profile
Name
Count. Substantiveness Alignment
Shift
Temporal
Adaptiveness
Profile Characteristics
0 Systemic
Realigners
15 4.73 4.33 4.53 High scores on all three
dimensions；hese firms are the most
decisive, engaging in major
restructuring, market exits, or
relocations. Their responses are both
early and strategically
comprehensive, as they fully realign
their operations and partnerships.
1 Strategic
Hedgers
29 3.48 3.03 4.31 This cluster exhibit moderate-to-high
adaptiveness across the board. They
engage in early actions, but their
responses tend to be more measured.
These firms diversify risk by shifting
production or entering new markets,
but they do not fully cut ties with
existing blocs.
2 Increment
al
Adapters
27 3.21 2.79 2.98 Firms that make gradual changes in
response to decoupling pressures.
These firms exhibit moderate scores
across all dimensions, indicating that
they are less aggressive in their
responses. These changes happen
slowly over time, and their alignment
shifts remain limited.
3 Symbolic
Couplers
28 3.94 2.21 2.89 These firms are quick to make visible
investments, such as building new
facilities or acquiring assets, but they
avoid significant political
realignment. Instead, they focus on
symbolic or superficial actions, such
as issuing public statements, while
maintaining their existing
geopolitical and market alignments.
Their actions are often reactive
rather than anticipatory.