Local Procurement and Value Chain Upgrading of Offshore Subsidiaries in China

Abstract

This study integrates the offshore subsidiary (OS) development (OSD) and smile curve paradigms to examine the impact of local procurement on OSs' functional upgrading of value chains (VCs) in China. Based on an offshore subsidiary development framework with headquarters-driven, internal subsidiary-driven, and host economy-driven, we draw on Taiwan-based subsidiaries operating in China from 2008 to 2018 to examine the relationships among value-chain functions, including technological enhancements (TE), marketing operations (MO), procurement mandates, and principal OSD determinants. In sum, the empirical results indicate that Taiwan-based OSs are able to strengthen their competitive advantage in China by engaging in persistent value creation through upgrading both TE and MO. In this study, some insights shed light on subsidiary development processes for multinational enterprises to revisit their strategies for deglobalization of VC configurations.

Full text

Journal of Management and Business Research 2022, Vol . 3 9, No . 3 , 403-429
DOI: 10.6504/JMBR.202203_39(3).0005
Local Procurement and Value Chain Upgrading
of Offshore Subsidiaries in China
Meng-Chun Liu
Chung-Hua Institution for Economic Research
Tung-Lung Chang
Long Island University
National Chung-Hsing University
Paper No.：4001
Received October 22, 2020 → First Revised April 15, 2021 → Second Revised October 14, 2021 → Accepted February
10, 2022
This study integrates the offshore subsidiary (OS) development (OSD) and smile curve paradigms to examine the
impact of local procurement on OSs’ functional upgrading of value chains (VCs) in China. Based on an offshore subsidiary
development framework with headquarters-driven, internal subsidiary–driven, and host economy-driven, we draw on
Taiwan-based subsidiaries operating in China from 2008 to 2018 to examine the relationships among value-chain
functions, including technological enhancements (TE), marketing operations (MO), procurement mandates, and principal
OSD determinants. In sum, the empirical results indicate that Taiwan-based OSs are able to strengthen their competitive
advantage in China by engaging in persistent value creation through upgrading both TE and MO. In this study, some
insights shed light on subsidiary development processes for multinational enterprises to revisit their strategies for
deglobalization of VC configurations.
Keywords: localization, local sourcing, offshore subsidiary development framework, smile-curve logic, global value chain.
Introduction
Globalization motivates multinational enterprises
(hereafter MNEs) to restructure their international
operations to improve their competitiveness in pursuit of
business growth in the global market. In this context, a
viable strategy for enhancing the experience curve effect
(Bartlett & Ghoshal, 1992; Chang, 1996; Fuentelsaz,
Garrido, & Gonzalez, 2022) is to enhance value creation
by making higher-quality products, improving production
efficiency, or incorporating more skilled activities into
Meng-Chun Liu, Research Fellow and director of Mainland
China Division at Chung-Hua Institution for Economic Research
(CIER). 75, Chang-Hsing St., Taipei, Taiwan. liu@cier.edu.tw.
Tung-Lung Steven Chang (corresponding author) Professor of
Marketing and International Business, College of Management,
Long Island University – Post Campus. 720 Northern Blvd.
Brookville, NY 11548, USA. steven.chang@liu.edu. (516) 299-
4195.
global value chains (GVCs) through outsourcing and
offshoring (Humphrey & Schmitz, 2002; Kaplinsky,
2000). To increase their competitiveness, MNEs
commonly employ GVCs to coordinate various functional
activities jointly performed by their offshore subsidiaries
(OSs) and strategic partners through interorganizational
networks distributed throughout the global supply chain
(Buckley & Ghauri, 2004). The effective configuration of
GVCs typically yields favorable outcomes in offshoring
and is thus widely embraced by MNEs, both to reduce
costs and to obtain economies of scale across national
borders (Pla-Barber, Linares, & Ghauri, 2019). MNEs’
OSs can be regarded as a mediator between MNE
headquarters (HQ) and production networks in host
countries (Fuller & Phelps, 2018). They can also serve as a
touchpoint with external actors (Hutzschenreuter &
Harhoff, 2021). While MENs restructures their GVCs,

2 Journal of Management and Business Research September
OSs may transform their mandates especially in playing
more aggressive roles in local sourcing.
Stands out a research issue regarding how OSs can
upgrade themselves by expanding their business functions
within GVCs. Knowledge embedded in an MNE’s internal
and external networks may support OS development
(OSD) by increasing the breadth of the OS’s functional
operations and its competence in performing specific
activities (Pu & Soh, 2018). In OSs, various business
operations typically stem from a single business
function—often marketing, production, or research and
development (R&D)—and branch into a broad spectrum
of business functions within the GVCs in the host country
(Model, 2015). To participate in GVCs, MNEs may be
under pressure to channel value-adding activities into their
VCs. The competitive position of an MNE results from its
OSs’ capacity to upgrade in a GVC and engage in such
activities (Ryan, Buciuni, Giblin, & Andersson, 2020).
Value creation is essential to GVC governance (Kano,
Tsang, & Yeung, 2020). However, relatively few studies
(Kim & Kim, 2020; Lopez-Vega & Tell, 2021; Quesada-
Vazquez & Rodríguez-Cohard, 2019; Winter, 2010) have
empirically examined the effects of OSs’ access to local
advantages on their development in terms of value
creation under the governance of MNEs’ GVCs.
To close this gap in the literature, this study explores
the impact of OS autonomy in procurement mandates and
local sourcing on the VC upgrading of nonproduction
functions in terms of technological enhancement (TE) and
marketing operations (MO). By drawing on Taiwan-based
OSs in China, this study looks at the effect of OSs’
capability building on their VC upgrading in business
functions. The study takes China as the host country of
interest in which most Taiwan-based manufacturing firms’
overseas ventures. With reference to an up-to-date
government data set of Taiwan-based firms’ overseas
business from 2008 to 2018, we investigate the
transformation of OS local embeddedness. Thus, this
study may empirically contribute to bridging GVC
theories and the OSD paradigm. MNEs employ the
concept of “the smile of value creation” as part of their
global VC configuration to capitalize on their offshore
location advantage and on their core competencies
(Mudambi, 2007). As highlighted by Porter (1986), MNEs
can gain a competitive advantage by geographically
concentrating certain global marketing activities or
coordinating dispersed marketing activities to enhance
local responsiveness. Thus, the relevance of support
functions in MNEs’ activities, such as R&D and
marketing, are critical building blocks of GVCs that merit
special scholarly attention (Gal, Nicoletti, von Rüden,
Sorbe, & Renault, 2019).
Recent studies (Chen & Tsou, 2020; Geleilate,
Andrews, & Fainshmidt, 2020; Hsu, Chen, & Caskey,
2017; Liu & Chen, 2007a) have investigated the impact of
capability upgrading for OSD in emerging economies,
centering on the enhancement of subsidiary performance.
Based on Taiwan-based subsidiaries in China, Liao, Yu,
and Seetoo (2006) examined the determinants of OSD,
such as the close relationship between subsidiaries and
headquarters, subsidiaries' resources and the completeness
of VC activities, and external networking relationships.
VC upgrading associated with high value-adding functions
is noted to increase OSs’ overall value creation capability
(Burger, Jindra, Marek, & Rojec, 2018). However,
relatively few studies (Kano, 2018; Kano et al., 2020; Liao
et al., 2006; Narula, 2019) have empirically addressed VC
upgrading conducted by OSs. Herein, by following Model
(2015), this study can define VC upgrading of OSs as the
expansion of business functions from production to higher
value-added functions such as R&D, quality control,
marketing, and after-sales service.
The contributions of this study are summarized as
follows: First, complementing the literature on core
competency–based upgrading in OSs, we investigate the
business function extensions of OSs. Under the smile-
curve logic proposed by Shih (1996), we outline the
patterns of OSs’ value-added upgrading in terms of the
technology and marketing dimensions. Second, by
employing an OSD paradigm, this study reveals that this
value-added upgrading in both technology and marketing
dimensions occurs within a complex system and extends
beyond a linear summation relationship. Third, through
looking at the role played by an OS’s autonomy in local

2022 LIu and Chang 3
procurements, this study highlights OSs with such an
additional mandate not only to extend their value-added
function but also to trigger their overall VC upgrading.
The remainder of this paper is organized as follows.
Section II presents the theoretical background of this study
and the literature review, as well as our OSD-related
hypotheses. Section III introduces the quantitative models
and the quantitative approach for testing the hypotheses.
The empirical results and discussion are presented in
Section IV. In Section V, we summarize our findings and
consider strategic implications.
Literature Review: VC
Upgrading and OSs’ Autonomy
In the GVC governance literature, explorations of the
roles played by OSs, especially in sourcing, conform to
two major schools of thought. The first perspective, based
on economic geography theory, centers on the global
configuration of value creation activities and highlights
the impacts of both economic and noneconomic factors as
well as the strategic choice of location economies
(Buckley & Ghauri, 2004; Cantwell & Mudambi, 2005;
Cantwell & Mudambi, 2011; Nguyen & Cassidy, 2016).
Recent studies have integrated business networks with
internalization theory to explore the relational dynamics of
GVC governance across culturally diverse partners
(Griffith & Myers, 2005; Kano, 2018). Other
investigations reveal that MNEs’ geographical
deployments are consistent with their GVCs (Narula,
2019; Rugman & Verbeke, 2004) in production and are
closely aligned with their strategic and competitive
considerations in innovation (Cantwell & Mudambi, 2011;
Mudambi & Navarra, 2004). In general, GVCs with
favorable governance help MNEs and their partners
reallocate resources quickly and smoothly in response to
increased volatility in the international business
environment (Buckley, 2009; Gereffi, Humphrey, &
Sturgeon, 2005). Thus, OSs that effectively participate in
GVCs in their host countries can facilitate improvements
in their parent MNEs’ competitiveness, thereby enhancing
growth and global performance. Accordingly, an OS
should gain autonomy for localization (Shi, Qiu, & Fan,
2020).
Complementary to the first perspective, the second
perspective has roots in transaction cost theories. It
focuses on buyer–supplier transactions to compare the
effectiveness of arm’s length market relationships with
those of hierarchical governance (Humphrey & Schmitz,
2002; Ryan et al., 2020) and the strategic choice of
internalization, including strategic partnerships and HQ–
subsidiary relationships (Ryan et al., 2020). Tomiura, Ito,
and Wakasugi (2011) empirically compare MNEs’
offshore outsourcing firms with the insourcing firms from
their foreign direct investment (FDI) subsidiaries to
examine the transaction cost interpretation. Geleilate et al.
(2020) investigate OSs’ autonomy and sourcing under
broader transaction cost theories. Increasing subsidiaries’
autonomy in unstable industry environments may
constitute an efficient structural arrangement given that
decentralized decision-making requires less monitoring
and coordination, thus reducing agency costs and
preventing opportunism. In exploring MNEs’ choice of
locations for foreign outsourcing, Pelegrín and Garcia-
Quevedo (2015) suggest that MNEs, which are extremely
dependent on HQ’s services, strive to retain control over
their specific ownership advantage and exploit vertical
integration overseas.
Value creation and value capture have long been
considered pivotal for MNEs to leverage their global
competitiveness. Porter (1986) identifies the global
configuration of MNEs’ VC activities and their
coordination of those activities as two elements to these
firms’ international strategies. For OSs, the extension of
business functions from production to innovation is
generally viewed as the pinnacle for OSs in accomplishing
their upgrading status within GVCs and also achieving
long-term growth (Ryan et al., 2020). In performing the
tasks assigned by its HQ, an OS develops its capabilities
to obtain economies of scale and economies of scope and
to strengthen the experience curve effect. Resources are
necessary to improve the core competencies of an MNE
(Ghoshal & Bartlett, 1988b; Hedlund, 1986; Prahalad &
Doz, 1981; Rugman & Verbeke, 2001). In this regard,

4 Journal of Management and Business Research September
higher-tier OSs play significant roles in value creation and
value capture (Pavlínek &  enka, 2016). Building upon
the configurational approach, Birkinshaw and Morrison
(1995) classified OSs into three essential roles, including
local implementer, specialized contributor, and world
mandate. In general, OSs increase the number of their
value-added business functions as they undergo a local
implementer–world mandate role transition.
Interactions Between Upgrading in TE and MO
The extent of an OS’s autonomy depends on its
position within the GVC. VC upgrading by incorporating
more high value-added functions increases an OS’s overall
value creation capability (Burger et al., 2018). In the
business function portfolio, the autonomy of a subsidiary
shifts over time from operationally to strategically
oriented. The breadth and scope of functional upgrading in
the VC in both technology and marketing induce an
upward shift in MNEs’ capabilities (Chang, 1995, 1996;
Johanson & Vahlne, 1977; Lopez-Vega & Tell, 2021; Shih,
1996). Recent evidence indicates that OSs’ functional
upgrading adheres to a hierarchical mechanism, with firms
specializing in manufacturing and delegating higher return
activities, including design and marketing functions, to
their international partners. As argued by Kaplinsky,
Morris, and Readman (2002) and Pehrsson and Pehrsson
(2020), firms’ functional upgrading changes the
combination of activities within and among links in the
VC. Such upgrading may include moving away from
production and toward design, marketing, and even
branding. Following this logic, upgrading shifts an OS’s
business functions from competing on production
efficiency to competing on innovation in design and
marketing activities in the GVC (Chaminade & Vang,
2008).
In this context, R&D and marketing are two functions
with a distinctive strategic element. In general, they are
more globally integrated and less locally responsive than
other functional activities such as marketing and after-
sales service. Regarding local responsiveness, local
market–oriented OSs typically enjoy greater autonomy in
marketing and sales (Van den Bulcke & Halsberghe, 1984;
Garnier, 1982) and tend to serve a larger portion of the
market (Hood & Taggart, 1999).
Studies have examined the interaction of technology
and marketing units within a firm. Luo, Slotegraaf, and
Pan (2006) investigate competition and conflict among
marketing, R&D, finance, and other departments and
suggest that fostering concurrent cross-functional
cooperation and competition can direct conflicts toward
constructive interactions and further promote a firm’s
overall market learning and performance. Functional
activities (e.g., R&D, design, and marketing) can be more
globally integrated or locally responsive depending on the
location as well as on the type of subsidiary strategy
(Tolentino, Janne, & Chen, 2013).
Cross-functional collaboration between marketing
and R&D sectors increases the probability that new
products are developed in a timely manner and are
successful. In this context, technological and market-
related functions are relevant. A literature review
conducted by Brettel, Heinemann, Engelen, and Neubauer
(2011) confirms that R&D–marketing integration confers
various benefits. In particular, it positively affects the
development of new products. Marketing contributes
information related to clients’ preferences and
competitors’ potential actions. Such data facilitate the
development of successful products, especially with
regard to evaluating the commercial potential of candidate
products and reducing product redesign costs (Lopez-Vega
& Tell, 2021). During the commercialization phase, the
integration of R&D with marketing is particularly
essential in discussions of innovations with leading
clients. At the end of the commercialization phase of new
products, marketing typically depends on information and
technical advice from the technological sector to win the
favor of test clients prior to a broad market launch.
In the context of OS upgrading, expansions in the
scope of technological functions should not be
independent of expansions in the scope of marketing
functions. Instead, such expansions should be interactive
to some extent. Hypothesis 1 is presented as follows:
Hypothesis 1. Offshore subsidiaries’ functional

2022 LIu and Chang 5
upgrading in technology enhancements
(TE) and marketing operations (MO)
interact to a certain extent.
OS Procurement Mandates and Local
Sourcing
The development of local sourcing by OSs has
numerous generic factors that are associated with
expanding local supply chains. Subsidiaries with higher
autonomy over their component and material sources tend
to favor local inputs (Eberhardt, McLaren, Millington, &
Wilkinson, 2004; O'Farrell & O'Loughlin, 1981; Wei, Liu,
Wang, & Wang, 2012). Specifically, OSs with greater local
autonomy may enjoy a wider scope in the nomination and
selection of local vendors and subsequent procurement
processes (Budiyanto, 2016; Hewitt‐Dundas, Andréosso‐
O'callaghan, Crone, Murray, & Roper, 2005). OSs with
greater autonomy in production planning are more likely
to source from local firms than are OSs in which
production planning is largely decided by HQ. Such
subsidiaries with greater autonomy in procurement
leverage the proximity of monitoring and communication
to flexibly adjust products’ technical specifications and
satisfactorily meet clients’ needs.
Local sourcing performed by OSs in China primarily
involves using China as a production and export platform.
Procuring and processing components in a large facility
enable the establishment of the bargaining power that is
crucial to reducing per-unit costs incurred in international
market competition and to serving the expanding Chinese
market through large-scale production. OSs are inherently
inclined to strengthen R&D intensity in their sourcing
activities. Adapting components and parts to the local
market and performing intensively innovative activities
constitute parts of OSs' strategic focus necessary for
supporting these adaptation activities (Kotabe & Zhao,
2002). The relevant viewpoints are summarized as
follows.
First, the use of networks in the host country
enhances an OS’s abilities to collect and process
information and to gain access to desirable assets through
local supply chains. The end goal is to gain a competitive
advantage and thereby promote business growth. Certain
foreign exogenous factors influence opportunities for local
sourcing. Greater participation in local industries and a
high level of managerial autonomy from the parent MNE
facilitate an OS’s access to what is locally available,
enhance the quality of local inputs, and increase the
reliability of local suppliers (Amendolagine, Presbitero,
Rabellotti, & Sanfilippo, 2019).
Second, autonomy helps OSs develop local supply
chains and enables the effective utilization of those supply
chains (Williams, McDonald, Tüselmann, & Turner,
2008). OSs with greater local autonomy may be more
willing (or simply have more scope) to select local
suppliers (Hewitt‐Dundas et al., 2005). Using evidence
from China, Eberhardt et al. (2004) argue that OSs with a
high level of autonomy contribute crucially to local
sourcing decisions.
Third, OSs with the autonomy to grant mandates in
operational decision-making permit OSs to gain both
economies of scale and economies of scope through the
acquisition of know-how and locally available assets. The
consensus reached by most studies on OSs’ pursuit of their
autonomy is that OSs’ relationships with local
stakeholders significantly positively affect HQ’s
willingness to grant them greater autonomy. However,
OSs' building their local networks are beyond improving
their proactive capabilities, eventually aiming to push the
limits of their autonomy (Schmid & Schurig, 2003).
Local Procurement and VC Upgrading from the
Marketing Aspect
Herein, localization refers to the business endeavors
in which OSs engage to integrate into the host economy.
Specifically, this study may measure the extent of OSs’
business localization from two dimensions of their
procurement mandate and their local sourcing autonomy.
The decentralization of authority in procurement enables
an OS to customize its investments, actively in evaluating,
adopting, and establishing its own technologies
(Budiyanto, 2016). With an established supplier base, such
an OS is consequently afforded greater autonomy (Zanfei,
2000). Comparing Japanese manufacturing subsidiaries

6 Journal of Management and Business Research September
with their German counterparts in the British Isles, Hood
and Taggart (1999) indicate that an OS with greater
autonomy usually serves a larger portion of the market.
OSs with well-developed supply chain networks in
the host region may have enhanced abilities to gather and
process information and to gain access to desirable assets.
In the context of marketing and sales, the extent of an
OS’s autonomy may depend on its position within the
group-wide supply chain. Subsidiaries oriented toward
local markets typically enjoy greater autonomy than do
those without such an orientation (Van den Bulcke &
Halsberghe, 1984; Garnier, 1982). OSs develop various
competencies over time (Phelps, 1993) and expand their
autonomy to explore and exploit market opportunities for
both value creation and value capture. However, through
building their local supply chain networks, OSs may
extend the frontier of their autonomy (Schmid & Schurig,
2003), including with regard to MO.
Hypothesis 2a. The extent of an OS’s VC upgrading in
marketing operation (MO) is positively
determined by the status of its autonomy
in sourcing.
Hypothesis 2b. The extent of an OS’s VC upgrading in
marketing operation (MO) is positively
affected on the ratio of its local sourcing.
Local Procurement and VC Upgrading from the
Technological Aspect
OSs in the host region develop their business
networks and are entitled to a certain degree of autonomy
for creating, expanding, and effectively leveraging local
supply chains. Local sourcing drives OSD in innovation
and R&D activities. For OSs, geographical proximity to
their suppliers is an essential facilitator of quality
improvement. In short, to reduce costs and enhance
quality, OSs prefer to collaborate with local suppliers
(Hewitt‐Dundas et al., 2005). This enables OSs to enhance
their abilities of collecting and processing information and
of gaining access to desirable assets such that a
competitive advantage can be obtained (Williams et al.,
2008).
Local sourcing drives OSD in innovation and R&D
activities toward strategic growth. A decentralization of
authority over procurement enables subsidiaries to
customize their investments, including actively evaluating,
adopting, and establishing technologies in accordance with
their own strategies (Budiyanto, 2016). OSs’ R&D
investments in their host countries may access local R&D
resources, develop cross-border R&D networks, or
monitor its competitors’ technological developments
(Casson & Singh, 1993; Dambrine, 1997; Kumar, 2001).
Moreover, OSs usually have an incentive to pursue a
center of excellence. This involves the possession of
mandates for cultivating and reinforcing their core
competencies. In the context of operational decision-
making, an OS’s growing autonomy in sourcing
(including expanding local sourcing) results in the transfer
of new technologies and know-how from other parts of the
MNE. By leveraging this knowledge and locally available
assets, OSs can obtain economies of scale and economies
of scope (Williams et al., 2008).
Studies have empirically demonstrated that an OS
with greater autonomy in sourcing is more likely to
identify local suppliers and establish network relationships
with them (Jarillo & Martíanez, 1990; Jindra, Giroud, &
Scott-Kennel, 2009). Such behavior helps an OS access
location-specific advantages for VC upgrading. The
available capabilities of OSs' local supply chains can be
leveraged to enhance their competitive advantage
(Williams et al., 2008). The following hypotheses are
presented accordingly:
Hypothesis 3a. The extent of an OS’s VC upgrading in
TE is positively dependent on the extent
of its autonomy in sourcing.
Hypothesis 3b. The extent of an OS’s VC upgrading in
TE is positively dependent on
opportunities for local sourcing.
Methodology
3.1 Model and Estimation Approach
Model Setting Under the OSD Paradigm

2022 LIu and Chang 7
In general, manufacturing subsidiaries with higher
autonomy develop extensive local networks to enhance
their local economy embeddedness (Birkinshaw & Hood,
1998; Quesada-Vazquez & Rodríguez-Cohard, 2019), and
some of these OSs acquire nonproduction functions to
upgrade their mandates (Amin, Bradley, Howells,
Tomaney, & Gentle, 1994; Dawley, 2011; Phelps, 1993;
Pike, 1998; Szalavetz, 2018). In sum, the development of
OSs is affected by the influence of HQ, the conditions of
the host region, and the internal development of the OS.
In the OSD paradigm, HQ-driven development may
be manifested as the construction of a new plant, the
transfer of proprietary technologies, or engagement in
investment activities associated with a new product or
market mandate. Such efforts strengthen the resource base
of the OS, enabling it to perform its roles and fulfill its
responsibilities effectively. On the basis of this concept,
we employ the control of subsidiary ownership (OWNER),
technology transferred from HQ (TPARENT), and
mandates of offshore production (LPB) as proxies for HQ-
driven development.
Host region–driven development entails the creation
and maintenance of a local business environment in which
foreign subsidiaries operate, and further enhances OS’s
capabilities and resources (Birkinshaw & Hood, 1997).
OSD is positively affected by the host government’s
ability to foster growth and collaboration in the business
environment through supportive macroeconomic policies
and investments in infrastructure and education. We proxy
host region–driven development by using the regional
consumption power (PER_GDP), the ratio of the regional
R&D ratio to the gross domestic product (GDP;
RD_GDP), and the sales ratio of the OS in the local
market (LMR).
The internal development of a subsidiary is achieved
through both induced and autonomous management and
changes to the stock of resources (Birkinshaw & Hood,
1997). The relationship between the subsidiary’s
managerial actions and the level of its resources is
reciprocal. An OS can influence its own direction by
building up its capacities and controlling its critical
resources (Forsgren & Pahlberg, 1992; Ghoshal &
Bartlett, 1988a). Studies have revealed that OSs typically
earn a high status in the VC through their own
entrepreneurial efforts rather than through HQ’s assistance
(Crookell & Morrison, 1990; Etemad & Dulude, 1986).
We proxy internal subsidiary development by using the
variables FSCALE, RDR, and TLOCAL.
As displayed in Figure 1, procurement status and
local sourcing are two major forces that drive an OS’s
upward status shifts with regard to TE, MO, and the TE–
MO interaction. Furthermore, according to the subsidiary
development paradigm (Hood & Taggart, 1999), the
models take relevant variables into account, including the
tasks assigned by HQ (HQ-driven development), the
subsidiary’s accumulated capabilities (internal subsidiary
development), and the local market conditions of the host
region (host region–driven development).
Local Sourcing
Procurement
Mandate
OSD: subsidiary
development paradigm
HQs-driven development
host country-driven
development
internal subsidiary
development
TE VC
Upgradin
g
MO VC
Upgradin
g
+
+
+
+
+
+

8 Journal of Management and Business Research September
Figure 1. Research Framework of VC Upgrading (TE and MO)
According to an OSD paradigm, we consider two
major determinants and a set of control regressors as
follows:
(1)
(2)
where PROCURE denotes autonomy for procurement,
PLMPR refers to the local sourcing ratio of materials and
components, OWNER indicates whether the OS is wholly
owned, TPARENT refers to the transfer of the OS’s
technology from HQ, LPB denotes the OS’s mandate for
local production in the host country, FSCALE represents
the size of the OS, RDR is the R&D intensity of an OS,
LMR measures the extent of an OS’s local market
orientation, TLOCAL indicates that the OS’s technological
development is mainly local, PER_GDP measures
regional GDP per capita, and RD_GDP is the ratio of the
regional R&D to the GDP. The high-tech sector (HITECH)
and years of operation (YR1–YR10) comprise the control
variables. In equations (1) and (2), respectively, u1 and u2
are the error terms.
Estimation Approach
On the basis of the preceding discussion, a basic
theoretical framework of the subsidiary development
paradigm is presented in which three sets of variables
explain the patterns of subsidiary upgrading in TE and MO
under the smile curve concept. As shown in Figure 2, this
concept is employed to assess the functional upgrading of
multinational subsidiaries for value capture. For example,
Burger et al. (2018) examine the cases of six countries
from Central Europe and Eastern Europe under this
concept. The present model is expressed as
Source: Modified from the framework presented by Shih (1996).
Figure 2. Value Chain Upgrading: The Smile Curve of Value Creation in OSD
Marketing operations (MO)+
Technology enhancements (TE)
MO
(Marketing, after sales
service)
TE
(R&D, Design, Quality
control)
Manufacturing
Up-stream Down-stream

2022 LIu and Chang 9
yj = F (PROCURE, OSD) for j=1 (TE), 2 (MO), (3a)
yj = F (PLMPR, OSD) for j=1 (TE), 2 (MO), (3b)
where yj is the dependent variable, OSD is a set of
variables related to the subsidiary development paradigm,
and PROCURE and PLMPR are the regressors examined
in equations (3a) and (3b), respectively.
The actual levels of subsidiaries’ upgrading are only
observable from their VC functions. Thus, we define and
estimate the model through latent variable methodologies.
Assume that latent variables [y*1i, y*2i] are determined by
𝑦1
∗= 𝑥1𝑖𝛽𝑖+ 𝑢1𝑖 (4a)
and
𝑦2
∗= 𝑥2𝑖𝛽𝑖+ 𝑢2𝑖 (4b)
where x1i and x2i represent the set of all independent
regressors (PROCURE, PLMPR, OSD, and other control
variables) in each equation, β1and β2 are parameter vectors
to be estimated, and u1i and u2i are the error terms.
Because seemingly unrelated regression has no
identification requirements, the model can be freely
constructed with or without the same independent
variables in each equation. We include all the independent
variables in each equation. The error terms u1 and u2
follow a bivariate normal distribution with zero mean, unit
variance, and correlation coefficient ρ:
𝐶𝑜𝑣(𝑢1, 𝑢2|𝑥1, 𝑥2)= 𝜌. (5)
According to Hypothesis 1, the value of ρ is
presumed to be positive and statistically significant. In
accordance with Hypotheses 2(a) and 2(b), both variables
PROCURE and PLMPR have positive effects on MO.
Based on Hypotheses 3(a) and 3(b), both PROCURE and
PLMPR also have positive effects on TE. Table 1 lists the
dependent and explanatory variables as well as their
definitions, corresponding measures, and descriptive
statistics.
Data Sources and Statistical Approaches
To evaluate TE and MO as two ordered endogenous
variables, we estimate equation (4) by using a Seemingly
Unrelated Bivariate Ordered Probit Model. This study
follows the approach pioneered by Birkinshaw, Hood, and
Young (2005) to depict a subsidiary’s upgrading and
mandate by measuring its value-added scope. The value-
added scope of a subsidiary refers to the number of stages
in the value-added chain, conducted in the host economy
(Hogenbirk & Van Kranenburg, 2006).
These high-value functions are typically associated
with the intangible components of a product or service,
including R&D or marketing knowledge. As mentioned,
TE and MO reflect the development patterns of Taiwan-
based subsidiaries in China from technological and
marketing aspects. The figure of TE refers to two when
these subsidiaries have two value-added functions in the
technological aspect: both quality management and R&D
& Design. The figure of MO is two when these
subsidiaries encompass two value-added functions:
marketing and after-sales service. In the same way, when
TE is one, these OSs have a mandate with quality
management or R&D & Design in their value-added
functions in the technological aspect. When MO is one,
these OSs have a mandate with marketing or after-sales in
their value-added functions in the marketing aspect.
The notions behind the measures for both TE and MO
in this study are as follows. Activities within a GVC can
be classified into upstream activities (R&D and product
design), midstream activities (manufacturing and
logistics), and downstream activities (marketing and after-
sales service); (Burger et al., 2018; Mudambi, 2013).
Empirically examining both kaizen, the Japanese style of
endless improvement activities, and the total quality
management, Ueki and Martínez (2020) reveals that firms
transform quality control management into their
innovative capabilities. Other studies (Bourke & Roper,
2017; Martinez-Costa & Martínez-Lorente, 2008) have
highlighted the benefits of various quality management
measures on product and process innovation performance.
These benefits are attributable to the shaping of the
quality–innovation relationship by complementarities and

10 Journal of Management and Business Research September
learning-by-using effects.
Our data are sourced from the Survey on Overseas
Investment by Taiwan-Based MNEs conducted by the
Investment Commission of Taiwan. Table 2 presents the
functional diversification of Taiwan-based subsidiaries in
China for the years 2008 to 2018. During this period,
Taiwan-based manufacturing MNEs retain more than 50%
of their marketing units and almost >30% of their R&D &
Design units in China. By contrast, the ratios
corresponding to quality control units and after-sales
service units in their OSs remain over 45% and 35%,
respectively. Appendix 1 presents the reliability and
validity of the surveys, as determined using Kuder–
Richardson formula 20 (KR-20) and the Kaiser–Meyer–
Olkin (KMO) test.
Table 1 Summary of Descriptive Statistics
Variables
Operational Definition and Citations
Mean
Std. Dev.
Min
Max
Subsidiary
variables
MO
Extent of marketing activities as specified
Amin et al. (1994); Dawley (2011); Phelps (1993);
Pike (1998)
0.93
0.81
0
2
TE
Extent of technological activities as specified
Amin et al. (1994); Dawley (2011); Phelps (1993);
Pike (1998)
0.81
0.81
0
2
FSCALE
OS’s annual sales, in ten thousand USD and
expressed in natural logarithms
Van Egeraat and Breathnach (2012), Chin (2013)
8.83
3.68
-2.30
22.72
RDR
OS’s R&D intensity measured as a ratio of R&D
workers to total employees, in %
Liu and Chen (2007b, 2012)
0.07
0.16
0
2
OWNER
Wholly-owned by HQs in Taiwan, a dummy variable
Roehrig (1994)
0.83
0.37
0
1
TPARENT
Subsidiary’s technologies mainly transferred from its
HQs in Taiwan, dummy variable
0.85
0.35
0
1
TLOCAL
Technology development mainly takes place locally,
a dummy variable
Kumar (2001)
0.39
0.49
0
1
LMR
Local market orientation, i.e., annual local sales as a
ratio of OS sales, in %
Luo (2001)
62.33
39.78
0
100.0
LPB
Local production base, dummy variable
0.87
0.33
0
1
PROCURE
Procurement mandate, a dummy variable
Jia, Lamming, Sartor, Orzes, and Nassimbeni (2014);
Nassimbeni and Sartor (2006); Sartor, Orzes,
Nassimbeni, Jia, and Lamming (2014, 2015)
0.46
0.50
0
1
PLMPR
Cross terms of PROCURE and the ratio of local
materials and components obtained, in %
Casson and Singh (1993); Dambrine (1997); Kumar
(2001); Reddy (2000)
31.52
39.74
0
100.0
Regional
variables
RD_GDP
Regional R&D intensity (%)*
Liu and Chen (2012); Schmid and Schurig (2003)
2.42
0.72
0.68
6.01
PER_GDP
Regional per capita income, in thousand USD**
Liu and Chen (2012); Schmid and Schurig (2003)
10.84
3.62
2.39
23.66
Control
HITECH
High-tech sector, dummy
0.69
0.46
0
1

2022 LIu and Chang 11
variables
Source: Investment Commission, Survey on Overseas Investment by Taiwan-Based MNEs, 2009–2019.
* National Bureau of Statistics of the People’s Republic of China, China Statistical Yearbook on Science and Technology, 2019.
** National Bureau of Statistics of the People’s Republic of China, China Statistical Yearbook, 2019.
Table 2 Value-Added Units of Taiwan-Based Subsidiaries in China
Unit: %
Year
Value-added units
08’
09’
10’
11’
12’
13’
14’
15’
16’
17’
18’
Technological
Aspect
R&D/design
30.98
27.87
33.85
30.72
29.82
28.21
33.05
26.94
30.92
37.89
34.80
Quality control
67.84
50.12
51.36
48.90
48.59
44.44
50.21
42.44
47.39
51.58
50.22
Marketing
Aspect
Marketing unit
54.12
49.65
51.75
53.29
53.21
52.71
56.90
52.03
58.63
54.74
52.42
After sales service
53.73
36.07
39.30
35.11
41.13
41.03
40.17
36.53
39.36
38.42
39.65
#ofobs.
255
427
257
319
389
351
239
271
249
190
227
Note: Numbers in parentheses refer to the number of effective respondents.
Source: Investment Commission, Survey on Overseas Investment by Taiwan-Based MNEs, 2009–2019.
The regional variables are mainly drawn from the
China Statistical Yearbook and China Statistical Yearbook
on Science and Technology for the 2008–2018 period. The
regions considered in this study are Shanghai, Shandong,
Tianjin, Beijing, Sichuan, Jiangsu, Northeast China,
Hebei, Chongqing, Zhejiang, Fujian, and Guangdong. The
sample is formed by pooling the data remaining after the
exclusion of missing values.
Table 3 presents the extent of MO and TE in the
sample subsidiaries in China from 2008 to 2018. The
Pearson χ2 test is performed to test the null hypothesis that
the OSs’ MO and TE (refer to the columns) in a two-way
table are independent. The test yields χ2(4) at the 0.01%
significance level. Thus, the null hypothesis is rejected;
OSs’ mandates in MO and TE interact significantly. This
result indicates that certain factors drive both MO and TE.
On the basis of the OSD paradigm, we further empirically
test the MO–TE relationship.
Table 3 Taiwan-Based Manufacturing Subsidiaries’ MO and TE in China (2008–2018)
TE
# of obs.
0
1
2
MO
0
769
(54.77)
(66.24)
293
(30.05)
(25.24)
99
(12.47)
(8.53)
1,161
(36.59)
(100.0)
1
458
(32.62)
(42.84)
362
(37.13)
(33.86)
249
(31.36)
(23.29)
1,069
(33.69)
(100.0)
2
177
(12.61)
(18.77)
320
(32.82)
(33.93)
446
(56.17)
(47.30)
943
(29.72)
(100.0)
Total
1,404
(100.00)
(44.25)
975
(100.00)
(30.73)
794
(100.00)
(25.02)
3,173
(100.00)
χ2(4) = 598.11***
(0.000)
Source: Investment Commission, Survey on Overseas Investment by Taiwan-Based MNEs, 2009–2019.
Note: *** p < 0.001.
Table 4 presents the correlation coefficients for all the variables used in our empirical models. Most of the

12 Journal of Management and Business Research September
correlation coefficients are <0.5. However, PER_GDP and
RD_GDP are highly correlated; multicollinearity is
present. To avoid this problem, both variables should not
be used in both equations (1) and (2); instead, they should
be employed in the MO and TE equations, respectively
(Table 5).

2022 LIu and Chang 13
Table 5 Determinants of Taiwan-Based MNEs’ Localization in China (Seemingly Unrelated BOPR)
Presumed Impact on
(I)
(II)
TE
MO
TE
MO
TE
MO
OWNER
+
+
-0.029
(0.631)
0.021
(0.714)
-0.052
(0.387)
0.000
(1.000)
TPARENT
+
+
0.064
(0.334)
0.210***
(0.001)
0.147**
(0.027)
0.248***
(0.000)
LPB
+
+
0.674***
(0.000)
-0.340***
(0.000)
0.737***
(0.000)
-0.244***
(0.000)
PROCURE
+
+
1.604***
(0.000)
1.132***
(0.000)
PER_GDP
…
+
0.068***
(0.000)
0.061***
(0.000)
LMR
+
+
0.009***
(0.000)
0.009***
(0.000)
PLMPR
+
+
0.017***
(0.000)
0.011***
(0.000)
RD_GDP
+
…
-0.041
(0.270)
-0.051
(0.168)
FSCALE
+
+
0.060***
(0.000)
0.045***
(0.000)
0.065***
(0.000)
0.050***
(0.000)
RDR
+
+
1.109***
(0.000)
0.221
(0.106)
0.959***
(0.000)
0.165
(0.226)
TLOCAL
+
+
0.811***
(0.000)
0.134***
(0.003)
0.786***
(0.000)
0.161***
(0.000)
HITECH
+
+
0.055
(0.261)
0.112**
(0.014)
0.090*
(0.065)
0.123***
(0.008)
YR1
1.313***
(0.000)
1.360***
(0.000)
YR2
-0.133
(0.169)
1.081***
(0.000)
-0.223**
(0.022)
1.081***
(0.000)
YR3
-0.069
(0.521)
1.057***
(0.000)
-0.200*
(0.065)
1.028***
(0.000)
YR4
-0.210**
(0.043)
0.943***
(0.000)
-0.284***
(0.006)
0.960***
(0.000)
YR5
-0.280***
(0.005)
0.959***
(0.000)
-0.313***
(0.002)
0.985***
(0.000)
YR6
-0.376***
(0.000)
0.866***
(0.000)
-0.425***
(0.000)
0.876***
(0.000)
YR7
-0.226**
(0.049)
0.875***
(0.000)
-0.321***
(0.006)
0.902***
(0.000)
YR8
-0.430***
(0.000)
0.675***
(0.000)
-0.473***
(0.000)
0.688***
(0.000)
YR9
-0.401***
(0.001)
0.600***
(0.000)
-0.479***
(0.000)
0.621***
(0.000)
YR10
-0.657***
(0.000)
0.113
(0.337)
-0.783***
(0.000)
0.110
(0.344)
YR11
-0.835***
(0.000)
-0.954***
(0.000)
Cut1
1.673***
(0.000)
2.991***
(0.000)
1.535***
(0.000)
2.741***
(0.000)
Cut2
2.668***
(0.000)
3.771***
(0.000)
2.582***
(0.000)
3.636***
(0.000)
ATHRHO ρ
0.388***
(0.000)
0.482***
(0.000)
LOG LIK.
-5220.885
-5254.660
Wald χ2 test
χ2 (19) = 1607.92
(0.000)
χ2 (19) = 1362.40
(0.000)
# of obs.
3,173
3,077
Note: * p < 0.10, ** p < 0.05, *** p < 0.01.

14 Journal of Management and Business Research September
Results and Discussion
The estimated results for the explanatory variables
are reported in models I and II of Table 5. The table lists
the Bivariate Ordered Probit Regression (BOPR) estimates
under various specifications corresponding to our
intention to capture the VR upgrading patterns of the OSs.
The figures in parentheses below the estimated
coefficients of the explanatory variables are the p values,
which represent the statistical significance of all of the
explanatory variables. Overall, the results confirm our
expectations. In the model specification, the variables
PROCURE and PLMPR are examined in models I and II,
respectively. The two variables have a correlation
coefficient of 0.85.
Interaction of TE and MO
To determine the potential connection between TE
and MO, we employ BOPR instead of univariate ordered
probit regression (UOPR) to empirically explore VC
upgrading in Taiwan-based OSs in China. As mentioned,
each endogenous discrete variable is associated with
exogenous variables and coefficients. If u1 and u2 are
assumed to be independent and normally distributed, (1)
and (2) can be estimated using UOPR. By contrast, BOPR
is appropriate if u1 and u2 are joint normal and if Cov (u1,
u2) = ρ, where ρ is a scalar correlation parameter. UOPR
can constitute a special case of BOPR where ρ = 0. Both
UOPR and BOPR involve estimations using the maximum
likelihood method. Table 5 presents the results of the TE
and MO equations obtained through BOPR. The
significance of the ρ statistic and the likelihood-ratio test
of the independent equations in Table 5 provide evidence
that the error terms in the two equations are correlated,
justifying the application of BOPR to the TE and MO
equations.
As shown in Table 5, the estimated correlation
coefficients ρ (0.388 and 0.482 in models I and II,
respectively) are significantly positive. Thus, hypothesis 1
is supported. Our findings reveal that an OS engaging in
more extensive upgrading in TE has more technological
capabilities for supporting its marketing campaigns. By
contrast, an OS engaging in more extensive upgrading in
MO uses those marketing capabilities not only to generate
sales from its existing products but also to provide insights
for needed R&D and new product development to explore
the latent demand in the local market. Upgrading in TE
and MO may be dependent on observed OSD-related
variables; we will explore this possibility in a later section.
Both VC-upgrading measures are also reliant on other
unobserved variables.
Our finding of a significant interaction between MO
and TE is in line with the findings of other studies.
Szwejczewski, Goffin, and Anagnostopoulos (2015)
mention that after-sales service is an integral business
element. Service requirements are systematically
evaluated during new product development with the input
of after-sales personnel and the use of field service data to
set design goals. Referring to evidence from telecom
MNEs in Brazil, Yamasaki, MacLennan, and Almeida
(2014) confirm that subsidiaries have the autonomy to
develop new products and services in the local market.
Examining upgrading in manufacturing OSs in Hungary,
Szalavetz (2016) observes that the quality of the
transferred technology is dependent on both the absorption
and marketing capabilities of the recipient. Improvements
in quality control, regarded herein as a technological
element, may contribute to reducing buyer–supplier
transaction costs (Ueki et al., 2020). Thus, TE drives MO
to a certain extent.
Effects of Autonomy in Procurement Mandates
and Local Sourcing on MO
As shown in models I and II of Table 5, the
coefficients of PROCURE and PLMPR are positive and
significant at the 0.1% level in both MO equations (1.132
and 0.011, respectively; p < 0.001 for both), thus
corresponding to both ends of the smile curve and
supporting hypotheses 2a and 3a. Our findings reveal that
the extent of an OS’s autonomy in localization with regard
to procurement and local sourcing explains their strategic
importance in the OS’s VC upgrading of marketing
functions (Brannon, James, & Lucker, 1994; Eberhardt et
al., 2004). Furthermore, OS autonomy, especially in local

2022 LIu and Chang 15
sourcing, can drive the development of networks with
local suppliers in the host country. Specifically, the
development of local networks can directly and
proportionally form backward and forward linkages more
likely, and further enhances an OS’s VC upgrading in
higher value-adding functions with regard to marketing or
R&D (Hewitt‐Dundas et al., 2005).
The extent of the Taiwan-based OSs’ autonomy in
marketing and after-sales service is strongly contingent on
their position within the group-wide supply chain, which
is highly related to their autonomy in procurement
mandates and local sourcing. A firm selects its sourcing
location not only according to the parameters of higher
quality and lower cost; it is striving to secure a higher
position within its production networks. The local
sourcing of materials, components, and capital goods
provides such OSs with a higher likelihood of enhancing
their VC upgrading in MO for value-added marketing
endeavors in China (Kotabe & Zhao, 2002).
Taiwan’s rapidly growing economy and its role as a
bridge to mainland China and Southeast Asia have
propelled its MNEs to respond to the diverse needs of
these markets. In collaboration with their international
clients, Taiwan-based MNEs engage in downstream
functional activities in China to acquire local market
knowledge and exploit local market opportunities
(Nassimbeni & Sartor, 2007). In short, OSs with
autonomy in local sourcing might be at an advantage in
exploring local business opportunities. They may also be
better equipped to exploit their parent MNEs’
competencies in the host economies.
Effects of Autonomy in Procurement Mandates
and Local Sourcing on TE
As shown in models I and II of Table 5, the
coefficients of PROCURE and PLMPR are positive and
significant at the 0.1% level in both TE equations (1.604
and 0.017, respectively; p < 0.001 for both). Thus,
hypotheses 2b and 3b are supported. Our findings are in
line with those of Jarillo and Martíanez (1990) in that
subsidiaries granted stronger procurement mandates by
HQs engage in R&D upgrading and quality control
activities in the host country. In this study, Taiwan-based
MNEs with a mandate for their OSs in China to engage in
local procurement appear to enjoy resource-seeking
offshoring. The present results are also consistent with
those of other studies (Casson & Singh, 1993; Dambrine,
1997; Kumar, 2001) in that they suggest that the
motivation for an MNE’s strategic growth in local
sourcing may include gaining access to host countries’
R&D resources, developing cross-border R&D networks,
and monitoring their competitors’ technological
developments.
Subsidiaries with greater autonomy in sourcing prefer
to choose their local suppliers, develop closer supply
chains, and further promote more extensive local sourcing
activities (Liu, 2011). Numerous forms of interactions that
involve buyer–supplier collaboration are observable from
OSs’ outsourcing in host countries. These include the
transfer of production plans, joint definitions of product
specifications, shared engineering and industrialization
choices, supplier adoption of agreed-upon capacity
planning and manufacturing tools, quality control, and
packaging and transport modalities (Nassimbeni & Sartor,
2007). It has been noted that OSs’ adaptation of local
components and parts to the local market as a strategic
focus requires the support of its intensive innovation
(Kotabe & Zhao, 2002). This suggests that offshore
manufacturing usually involves heavy production
activities with active innovation or higher R&D inputs.
The availability of local material and component supplies
attributable to the emergence of industrial clusters in
China (Puig, Portero, & González-Loureiro, 2017)
motivates OSs’ VC upgrading in R&D design and quality
control.
In general, the empirical results in both the MO and
TE equations for VC upgrading provide evidence to
support a full smile of value creation. Our findings suggest
that OSs’ localization positively affects VC upgrading not
only in R&D, production systems, and product
development but also in marketing, after-sales service, and
customer satisfaction. Drawing on data from a German
survey, Williams et al. (2008) develop a conceptual
framework of the links between autonomy, the use of

16 Journal of Management and Business Research September
networks, and domestic sourcing. They observe that the
autonomy for localization facilitate the merging of OSs
into local markets, enabling them to reap benefits from
locally available resources and networks.
Control Variables: OSD Paradigm
HQ-Driven Development
The coefficients of OWNER in both models are
insignificantly positive in the MO and TE equations.
Taiwan-based MNEs with wholly-owned subsidiaries in
China may intend to keep their core technology assets or
corporate secrets within their business network such that
knowledge spillover can be avoided. However, Roehrig
(1994) indicates that some legal requirements enabled
Chinese partners to retain an effective bargaining position
vis-à-vis their foreign joint venture partners. Specifically,
R&D collaboration drove technology transfer in those
Chinese partner firms. Similarly, Taiwan-based
subsidiaries’ upgrading in the market aspect may rely on
partnerships with local firms to overcome market barriers.
Our findings indicate that Taiwan-based subsidiaries
employ a given mandate to determine the patterns of their
localization to adapt to business environments in the host
country. For these OSs, corporate success is contingent on
effective adaptation to China’s unique business
environment. From the social network perspective, guanxi
culture dominates activities in Chinese society, including
business dealings (Yeung & Tung, 1996). Guanxi
encompasses a system of networks and relationships that
an OS can leverage to attain benefits. Through the
exchange of favors, an OS can obtain valuable information
and resources.
The coefficients of TPARENT are positive and
statistically significant for MO but insignificant for TE,
indicating that the empirical evidence is supported at the
right end of the smile curve (downstream) for MO but not
for TE. We observe that Taiwan-based OSs that rely
strongly on technology transferred from their parent
company tend to focus on building their marketing
capability rather than on technological upgrading. Such
OSs may appear to implement the technology exploitation
strategy of their HQs for business expansion abroad (Niosi
& Godin, 1999).
In both models, the coefficients of LPB are positive
and significant for TE. These findings are in line with the
premise that OSs retaining their production mandates in
China tend to engage in localization concerning TE. The
TE of the present OSs is consistent with the establishment
of local R&D networks embedded in vertical supply
chains in China (Liu & Chen, 2012). Our sample
subsidiaries usually act as original equipment
manufacturing (OEM) or original design manufacturing
(ODM) vendors in the global supply chain by using a
production base in China to serve international clients but
not the local market. The coefficients of LPB are negative
and significant in the MO equations, revealing an OS’s
role options between manufacturing or marketing, but few
taking both.
Host Region–Driven Development
The coefficient of PER_GDP is positive and
significant in the MO equation only; thus, the empirical
outcome is depicted at the right end of the smile curve
(downstream). As mentioned, per capita income
(PER_GDP) is a proxy for measuring the consumption
power of local markets. Our results reveal that China’s
sizable local markets constitute a location advantage that
drives OSD toward a right-inclined smile of value creation
concerning MO localization. This indicates that marketing
functions are essential components of mandates for FDI in
local markets (Majcen, Rojec, Jaklič, & Radosevic, 2005).
Similar to Chiao, Yu, Li, and Chen (2008), we observe
that Taiwan-based subsidiaries in China have relatively
high autonomy in becoming more responsive to the local
environment and are more geared to local market
conditions; hence, they modify their business strategies to
access the local market. This result represents a policy
implication for host countries seeking to drive their
foreign subsidiaries’ upgrading by expanding local market
demands.
We find an insignificantly negative coefficient of
RD_GDP in TE, weakly supported by a left-inclined smile
of value creation. This result is consistent with the premise
advanced by Liu and Chen (2012): Taiwan-based firms’

2022 LIu and Chang 17
R&D internationalization mainly relies on R&D
collaboration with their local supply chains rather than on
the local knowledge of host countries. Most of the
research units established by such subsidiaries in host
countries, especially in developing countries, primarily
function as production support units to support local
production (Birkinshaw & Hood, 1998).
As mentioned, LMR is used as a proxy for measuring
local access to the market. The coefficients of LMR in the
MO equations are significantly positive, suggesting that
the localization of marketing operations tends to be high
for subsidiaries with a strong local market orientation.
This premise is supported by a right-inclined smile of
value creation concerning MO localization. The results
indicate that market-seeking OSs interested in increasing
their local sales tend to upgrade their value-added units in
marketing to respond to various challenges in the
environment (Luo, 2001).
Internal Subsidiary Development
The coefficients for both FSCALE and RDR are
significantly positive in both models for TE and MO,
which means that an OS with high revenue and high
technological intensity engages more successfully in TE
and MO, as reflected by a full smile of value creation. The
coefficients of TLOCAL in model (I) are significantly
positive for TE but not for MO. Thus, TE localization
follows a left-inclined smile of value creation. This result
accords with the recommendation that OSs devote more
effort to R&D to effectively link it with their local
technological resources and thereby gain a global
competitive advantage(Reddy, 2000). OSs with more
R&D investments can further enhance their localization
through value-added technological development.
However, the impact of a subsidiary’s locally developed
technology on MO localization remains unclear.
Other Control Variables
We further examine two control variables, namely
the time trend (YR1–YR10) and HITECH industrial
sectors, over the 2008–2018 period. The coefficient of
HITECH is significantly positive for TE, indicating that
the present OSs’ upgrading of value-added technological
units in high-tech manufacturing sectors is significantly
more extensive than that of their counterparts in service
sectors. According to Liu and Chen (2012), one possible
explanation is that Taiwan-based manufacturing
subsidiaries, both high tech and traditional, leverage
regional industrial clusters to establish more
comprehensive R&D networks and supply chains than do
their service sector counterparts.
As displayed in Table 5, Taiwan-based OSs in China
conform to a “balanced smile” (Figure 2) of value creation
stemming from upgrading TE and MO. The signs of all the
explanatory variables excepting OWNER and RD_GDP in
Table 5 are in line with this premise. Furthermore, our
results provide insight into the driving forces of OSs’
procurement mandates and local sourcing in their
upgrading in TE and MO. We observe that OSs’ upgrading
cannot be regarded as a simple linear model or a
mechanically functional addition but rather entails a
significant interaction between upgrading in TE and
upgrading in MO.
4.5 Robustness Check
We perform a robustness check by employing the
same regression models for two sets of subsample data
covering the 2008–2013 and 2014–2018 periods,
respectively. Most of the core coefficients are comparable
to those in Table 5
1
(data not shown). Hence, our results
are highly robust.
Conclusion
OSs with autonomy can pursue value creation and
value capture through strategic VC upgrading in major
value-adding functions in host countries (Jarillo &
Martíanez, 1990; Pavlínek & enka, 2011). Taking several
forms (Hollenstein, 2005), subsidiary upgrading can
happen not only for efficiency improvement in the
production and management processes but also for VC
upgrading in business functions and relationship building,
1
The robustness BOPR test for the two data subsamples is
available upon request.

18 Journal of Management and Business Research September
taken up and beyond what firms specialize in currently.
On the basis of an OSD paradigm (Birkinshaw, 1996;
Birkinshaw & Hood, 1997, 1998), we highlight the
contribution of OSs’ autonomy in procurement mandates
and local sourcing to upgrading their smile of value
creation—specifically, to the functional upgrading of high
value-adding TE and MO in VCs. Examining Taiwan-
based OSs operating in China from 2008 to 2018, we
explore the major factors that influence OSD through
BOPR. These OSs played a critical, decades-long role in
the global supply chain as OEM/ODM vendors. Herein,
we investigate the effects of regional resources and local
conditions on VC upgrading in TE and MO. We select
China as the host country of interest because of its
geographical and market size and its considerable
differences across regional markets. The extent of these
subsidiaries’ VC activities ranges from manufacturing to
two high value-adding nonproduction functions, TE and
MO. As shown in Table 3, the results are highly robust
and in line with the smile curve logic (Mudambi, 2007;
Shih, 1996; Stöllinger, 2021). The contributions of this
study are summarized as follows.
First, unlike investigations focusing on productivity
or innovation activities (Bresciani & Ferraris, 2014; Liu &
Chen, 2012; Yang & Hayakawa, 2015), this study
investigates Taiwan-based OSs’ both technological
enhancement (TE) and marketing operation (MO) in
China. A subsidiary’s VC upgrading can be measured by
the extended autonomy associated with its portfolio of
business functions. We apply an OSD paradigm with three
kinds of determinants, namely HQs-driven, internal
subsidiary–driven, and host economy-driven factors. The
findings indicate that the patterns of OSD in VC
upgrading should be heterogeneous and multidimensional
across subsidiary types.
Second, the findings demonstrate that more extensive
upgrading in TE corresponds to more extensive upgrading
in MO and vice versa. Attention should be paid to this
positive TE–MO interaction. The smile curve logic of
value creation, based on the concept of VC
disaggregation, is employed to visualize subsidiary
development patterns (Chen, 2004; Mudambi, 2007; Shih,
1996; Stöllinger, 2021). We conclude that OSs’ upgrading
in value-added activities OSs’ upgrading in value-added
activities cannot be considered as a linear relation of
various business functions but as a nonlinear complicated
and interactive system.
Third, the empirical evidence reveals that an OS’s
autonomy in procurement mandates and local sourcing
drives OSD by facilitating upgrading in TE and MO.
Hence, an OS can accomplish a smile of value creation
through OSD by using HQ-assigned procurement
mandates. Such procurement mandates grant OSs the
autonomy to establish and strengthen their network
relationships in local supply chains. Furthermore, the
extension of local sourcing increases OSs’ access to local
resources. In general, an OS can leverage its location
advantage along with its core competencies to gain a
competitive advantage, thereby promoting business
growth.
Finally, for host countries, many of the OSs suffer the
so-called “branch plant syndrome” (Kleibert, 2016; Sonn
& Lee, 2012) since the decline in the location advantages
of a host economy, particularly due to higher production
costs, local rivalry and institutional voids. China’s
domestic market has expanded and become more open in
recent years. Market-seeking MNEs that strive to increase
their sales in China must strengthen their OSs’ VC
network linkages in TE and MO both locally and
internationally. To anchor footloose OSs, host countries
can provide support in the form of business-friendly FDI
policies for market entry. They can also grant foreign
MNEs easy access to local markets and R&D resources.
Amid the rapid advancements in China’s innovation
system and incentive policies, China can motivate foreign
MNEs to partner with domestic R&D organizations
through international joint ventures.
To enhance the generalizability of our findings,
several limitations should be addressed in future studies.
First, we do not examine variables related to business
performance. To better understand the strategic impacts of
VC upgrading in TE and MO in the host country, future
studies should investigate the effect of subsidiary
development configurations on financial performance at

2022 LIu and Chang 19
the firm level and on economic performance at the
industry level. Moreover, future studies can explore the
business performance achieved under different patterns of
VC upgrading at both ends of the smile curve in terms of
value-added functional activities in technology and
marketing.
Second, this study focuses on testing the impact of
major determinants on subsidiary development
determinants at the firm level in the manufacturing sector.
However, it does not explore the same determinants at the
industry level. In the future, a cross-sector comparison can
be performed to understand how local conditions or
country endowments influence the smile curve of
capability building in VC upgrading.
Third, a study on subsidiary-mandated development
(Birkinshaw, 1996) reveals that the product extension
process involves the highly concerted efforts of HQs and
subsidiaries. We do not examine HQ–OS interactions in
upgrading TE and MO. Investigations of such interactions
may enhance our understanding of how HQs can
strengthen their OSs’ competitive advantage by managing
their FDI and VC upgrading endeavors.
Kano et al. (2020) identifies the dynamics of GVC
arrangements, value creation and distribution, and chain-
level performance management as topics of interest for
GVC research. Given the significant TE–MO interaction
in terms of VC upgrading, which this study identifies, a
study of the synergistic effects of TE and MO may
provide insight into how MNEs can improve GVC
governance and thereby promote business growth by
leveraging OSs’ VC upgrading.
References
Amendolagine, V., Presbitero, A. F., Rabellotti, R., &
Sanfilippo, M. 2019. Local sourcing in developing
countries: The role of foreign direct investments and
global value chains. World Development, 113: 73-88.
Amin, A., Bradley, D., Howells, J., Tomaney, J., & Gentle,
C. 1994. Regional incentives and the quality of
mobile investment in the less favoured regions of the
EC. Progress in Planning, 41(1): 1-112.
Bartlett, C. A., & Ghoshal, S. 1992. Transnational
management: Text cases, and readings in cross-
border management homewood. Chicago, IL:
Irwin.
Birkinshaw, J. 1996. How multinational subsidiary
mandates are gained and lost. Journal of
International Business Studies, 27(3): 467-495.
Birkinshaw, J., & Hood, N. 1997. An empirical study of
development processes in foreign-owned subsidiaries
in Canada and Scotland. MIR: Management
International Review: 339-364.
Birkinshaw, J., & Hood, N. 1998. Multinational subsidiary
evolution: Capability and charter change in foreign-
owned subsidiary companies. Academy of
Management Review, 23(4): 773-795.
Birkinshaw, J., Hood, N., & Young, S. 2005. Subsidiary
entrepreneurship, internal and external competitive
forces, and subsidiary performance. International
Business Review, 14(2): 227-248.
Birkinshaw, J. M., & Morrison, A. J. 1995. Configurations
of strategy and structure in subsidiaries of
multinational corporations. Journal of International
Business Studies, 26(4): 729-753.
Bourke, J., & Roper, S. 2017. Innovation, quality
management and learning: Short-term and longer-
term effects. Research Policy, 46(8): 1505-1518.
Brannon, J. T., James, D. D., & Lucker, G. W. 1994.
Generating and sustaining backward linkages
between maquiladoras and local suppliers in
Northern Mexico. World Development, 22(12): 1933-
1945.
Bresciani, S., & Ferraris, A. 2014. The localization choice
of multinational firms’ R&D centers: a survey in the
piedmont area. Journal of Promotion Management,
20(4): 481-499.
Brettel, M., Heinemann, F., Engelen, A., & Neubauer, S.
2011. Cross ‐functional integration of R&D,
marketing, and manufacturing in radical and
incremental product innovations and its effects on
project effectiveness and efficiency. Journal of
Product Innovation Management, 28(2): 251-269.
Buckley, P. J. 2009. Internalisation thinking: From the

20 Journal of Management and Business Research September
multinational enterprise to the global factory.
International Business Review, 18(3): 224-235.
Buckley, P. J., & Ghauri, P. N. 2004. Globalisation,
economic geography and the strategy of
multinational enterprises. Journal of International
Business Studies, 35(2): 81-98.
Budiyanto, C. W. 2016. The reciprocal influence of
parent-subsidiary relationships and enterprise
systems implementation: An exploratory study using
power differential perspective. Unpublished doctoral
dissertation, Swinburne University of Technology,
Melbourne, Australia.
Burger, A., Jindra, B., Marek, P., & Rojec, M. 2018.
Functional upgrading and value capture of
multinational subsidiaries. Journal of International
Management, 24(2): 108-122.
Cantwell, J., & Mudambi, R. 2005. MNE competence‐
creating subsidiary mandates. Strategic Management
Journal, 26(12): 1109-1128.
Cantwell, J. A., & Mudambi, R. 2011. Physical attraction
and the geography of knowledge sourcing in
multinational enterprises. Global Strategy Journal,
1(3‐4): 206-232.
Casson, M., & Singh, S. 1993. Corporate research and
development strategies: The influence of firm,
industry and country factors on the decentralization
of R&D. R&D Management, 23(2): 91-107.
Chaminade, C., & Vang, J. 2008. Globalisation of
knowledge production and regional innovation
policy: Supporting specialized hubs in the Bangalore
software industry. Research Policy, 37(10): 1684-
1696.
Chang, T.-L. 1995. Formulating adaptive marketing
strategies in a global industry. International
Marketing Review, 12(6): 5-18.
Chang, T.-L. 1996. Cultivating global experience curve
advantage on technology and marketing capabilities.
International Marketing Review, 13(6): 22-42.
Chen, L.-J., & Tsou, H.-T. 2020. “Of course, cooperation
first”: The effect of subsidiary capability and
perception gaps for subsidiary development.
Multinational Business Review, 28(3): 401-426.
Chen, S.-H. 2004. Taiwanese IT firms’ offshore R&D in
China and the connection with the global innovation
network. Research Policy, 33(2): 337-349.
Chiao, Y.-C., Yu, C.-M. J., Li, P.-Y., & Chen, Y.-C. 2008.
Subsidiary size, internationalization, product
diversification, and performance in an emerging
market. International Marketing Review, 25(6): 612-
633.
Chin, T. 2013. An exploratory study on upgrading by FDI
OEMs in China. International Business Research,
6(1): 199-210.
Crookell, H., & Morrison, A. 1990. Subsidiary strategy in
a free trade environment. Business Quarterly, 55(2):
33-39.
Daggett, L. M. 2007. All of the Above: Computerized
Exam Scoring of Multiple Choice Items Helps to: (A)
Show How Exam Items Worked Technically, (B)
Maximize Exam Fairness, (C) Justly Assign Letter
Grades, and (D) Provide Feedback on Student
Learning. Journal of Legal Education, 57(3): 391-
422.
Dambrine, C. 1997. Globalization of research and
development: A business viewpoint. Paper presented
at OECD Working Group on Innovation and
Technology Policy conference, Facilitating
International Technology Cooperation. 13–14
August, Seoul, Korea.
Dawley, S. 2011. Transnational corporations and local and
regional development. In A. Pike, A. Rodríguez-Pose,
& J. Tomaney (Eds.), Handbook of local and
regional development: 394-412. London, UK:
Routledge.
Dzikowska, M. 2019. Determinants of functional
responsibilities of subsidiaries: Empirical
verification. In M. Dzikowska (Ed.), The role of the
subsidiary in international business: 107-134.
Cham, Switzerland: Springer.
Eberhardt, M., McLaren, J., Millington, A., & Wilkinson,
B. 2004. Multiple forces in component localisation in
China. European Management Journal, 22(3): 290-
303.
Etemad, H., & Dulude, L. 1986. Managing the

2022 LIu and Chang 21
multinational subsidiary: Response to
environmental changes and to host nation R&D
policies. London: Croom Helm.
Forsgren, M., & Pahlberg, C. 1992. Subsidiary influence
and autonomy in international firms. Scandinavian
International Business Review, 1(3): 41-51.
Fuentelsaz, L., Garrido, E., & Gonzalez, M. 2022. Speed
of institutional change and subsidiary performance:
The moderating impact of home and host country
learning. Global Strategy Journal, 12(1): 163-195.
Fuller, C., & Phelps, N. A. 2018. Revisiting the
multinational enterprise in global production
networks. Journal of Economic Geography, 18(1):
139-161.
Gal, P., Nicoletti, G., von Rüden, C., Sorbe, S., & Renault,
T. 2019. Digitalization and productivity: In search of
the holy grail-firm-level empirical evidence from
European countries. International Productivity
Monitor, (37): 39-71.
Garnier, G. H. 1982. Context and decision making
autonomy in the foreign affiliates of US multinational
corporations. Academy of Management Journal,
25(4): 893-908.
Geleilate, J.-M. G., Andrews, D. S., & Fainshmidt, S.
2020. Subsidiary autonomy and subsidiary
performance: A meta-analysis. Journal of World
Business, 55(4): 101049.
Gereffi, G., Humphrey, J., & Sturgeon, T. 2005. The
governance of global value chains. Review of
International Political Economy, 12(1): 78-104.
Ghoshal, S., & Bartlett, C. A. 1988a. Creation, adoption
and diffusion of innovations by subsidiaries of
multinational corporations. Journal of International
Business Studies, 19(3): 365-388.
Ghoshal, S., & Bartlett, C. A. 1988b. Innovation processes
in multinational corporations. In M. Tushman & W.
Moore (Eds.), Readings in the Management of
innovation: 499-518. New York, NY: Harper
Business.
Griffith, D. A., & Myers, M. B. 2005. The performance
implications of strategic fit of relational norm
governance strategies in global supply chain
relationships. Journal of International Business
Studies, 36(3): 254-269.
Hedlund, G. 1986. The hypermodern MNC—A heterarchy?
Human Resource Management, 25(1): 9-35.
Hewitt‐Dundas, N., Andréosso‐O'callaghan, B., Crone, M.,
Murray, J., & Roper, S. 2005. Selling global, buying
local? What determines the sourcing patterns of
multinational plants in Ireland? Regional Studies,
39(2): 225-239.
Hogenbirk, A. E., & Van Kranenburg, H. L. 2006. Roles of
foreign owned subsidiaries in a small economy.
International Business Review, 15(1): 53-67.
Hollenstein, H. 2005. Determinants of international
activities: are SMEs different? Small Business
Economics, 24(5): 431-450.
Hood, N., & Taggart, J. H. 1999. Subsidiary development
in German and Japanese manufacturing subsidiaries
in the British Isles. Regional Studies, 33(6): 513-528.
Hsu, C.-W., Chen, H., & Caskey, D. A. 2017. Local
conditions, entry timing, and foreign subsidiary
performance. International Business Review, 26(3):
544-554.
Humphrey, J., & Schmitz, H. 2002. How does insertion in
global value chains affect upgrading in industrial
clusters? Regional Studies, 36(9): 1017-1027.
Hutzschenreuter, T., & Harhoff, P.-L. 2021. The
accelerating effect of institutional environment
unfamiliarity on subsidiary portfolio expansion in a
new host country. International Business Review,
30(3): 101793.
Jarillo, J. C., & Martíanez, J. I. 1990. Different roles for
subsidiaries: The case of multinational corporations
in Spain. Strategic Management Journal, 11(7):
501-512.
Jia, F., Lamming, R., Sartor, M., Orzes, G., & Nassimbeni,
G. 2014. International purchasing offices in China: A
dynamic evolution model. International Business
Review, 23(3): 580-593.
Jindra, B., Giroud, A., & Scott-Kennel, J. 2009. Subsidiary
roles, vertical linkages and economic development:
Lessons from transition economies. Journal of World
Business, 44(2): 167-179.

22 Journal of Management and Business Research September
Johanson, J., & Vahlne, J.-E. 1977. The
internationalization process of the firm—a model of
knowledge development and increasing foreign
market commitments. Journal of International
Business Studies, 8(1): 23-32.
Kano, L. 2018. Global value chain governance: A
relational perspective. Journal of International
Business Studies, 49(6): 684-705.
Kano, L., Tsang, E. W. K., & Yeung, H. W.-c. 2020.
Global value chains: A review of the multi-
disciplinary literature. Journal of international
business studies, 51(4): 577-622.
Kaplinsky, R. 2000. Globalisation and unequalisation:
what can be learned from value chain analysis?
Journal of Development Studies, 37(2): 117-146.
Kaplinsky, R., Morris, M., & Readman, J. 2002.
Understanding upgrading using value chain analysis.
Retrieved on April, 3: 2017.
Kim, B.-G., & Kim, G.-B. 2020. An empirical analysis on
the determinants of foreign subsidiary's local
embeddedness: Focusing on Korean MNCs in China.
The Journal of Asian Finance, Economics, and
Business, 7(1): 205-215.
Kleibert, J. M. 2016. Global production networks, offshore
services and the branch-plant syndrome. Regional
Studies, 50(12): 1995-2009.
Kotabe, M., & Zhao, H. 2002. A taxonomy of sourcing
strategic types for MNCs operating in China. Asia
Pacific Journal of Management, 19(1): 11-27.
Kumar, N. 2001. Determinants of location of overseas
R&D activity of multinational enterprises: The case
of US and Japanese corporations. Research Policy,
30(1): 159-174.
Liao, M.-K. J., Yu, C.-M., & Seetoo, D. H. W. 2006. The
determinants of decision-making autonomy of
Taiwanese firm’s subsidiaries in China. Sun Yat-Sen
Management Review, 14(2): 397-427 (in Chinese).
Liu, B. J. 2011. MNEs and local linkages: Evidence from
Taiwanese affiliates. World Development, 39(4): 633-
647.
Liu, M.-C., & Chen, S.-H. 2007a. Cross-Border R&D
Network and R&D Internationalization: A
Perspective of Oversea Subsidiary. NTU
Management Review, 18(1): 211-238 (in Chinese).
Liu, M.-C., & Chen, S.-H. 2007b. International R&D
deployment and locational advantage: A case study of
Taiwan. In T. Ito & A. K. Rose (Eds.), International
Trade in East Asia: 81-108. London, UK: University
of Chicago Press.
Liu, M.-C., & Chen, S.-H. 2012. MNCs’ offshore R&D
networks in host country's regional innovation
system: The case of Taiwan-based firms in China.
Research Policy, 41(6): 1107-1120.
Lopez-Vega, H., & Tell, F. 2021. Technology strategy and
MNE subsidiary upgrading in emerging markets.
Technological forecasting & social change, 167:
120709.
Luo, X., Slotegraaf, R. J., & Pan, X. 2006. Cross-
functional “coopetition”: The simultaneous role of
cooperation and competition within firms. Journal of
Marketing, 70(2): 67-80.
Luo, Y. 2001. Determinants of local responsiveness:
Perspectives from foreign subsidiaries in an emerging
market. Journal of Management, 27(4): 451-477.
Majcen, B., Rojec, M., Jaklič, A., & Radosevic, S. 2005.
Productivity growth and functional upgrading in
foreign subsidiaries in the Slovenian manufacturing
sector. East-West Journal of Economics and
Business, 8(1-2): 73-100.
Martinez-Costa, M., & Martínez-Lorente, A. R. 2008.
Does quality management foster or hinder
innovation? An empirical study of Spanish
companies. Total Quality Management & Business
Excellence, 19(3): 209-221.
Model, U. 2015. Role typologies for foreign subsidiaries.
In D. Morschett, H. Schramm-Klein, & J. Zentes
(Eds.), Strategic International Management: Text
and Cases: 55-77. Wiesbaden, Germany: Springer
Gabler.
Mudambi, R. 2007. Offshoring: Economic geography and
the multinational firm. Journal of International
Business Studies, 38(1): 206-210.
Mudambi, R. 2013. Flatness: The global disaggregation of
value creation. In G. Cook & J. Johns (Eds.), The

2022 LIu and Chang 23
changing geography of international business: 9-16.
Basingstoke, UK: Palgrave Macmillan.
Mudambi, R., & Navarra, P. 2004. Electoral strategies in
mixed systems of representation. European Journal
of Political Economy, 20(1): 227-253.
Narula, R. 2019. Enforcing higher labor standards within
developing country value chains: Consequences for
MNEs and informal actors in a dual economy.
Journal of International Business Studies, 50(9):
1622-1635.
Nassimbeni, G., & Sartor, M. 2006. International
purchasing offices in China. Production Planning &
Control, 17(5): 494-507.
Nassimbeni, G., & Sartor, M. 2007. Sourcing in China: A
typology. International Journal of Production
Economics, 107(2): 333-349.
Niosi, J., & Godin, B. 1999. Canadian R&D abroad
management practices. Research Policy, 28(2-3):
215-230.
Nguyen, T. T. M., & Cassidy, J. F. 2016. MNC subsidiary
embeddedness in the host country an integrated
conceptual framework. VNU Journal of Science:
Economics and Business, 32(2): 77-86.
O'Farrell, P. N., & O'Loughlin, B. 1981. New industry
input linkages in ireland: An econometric analysis.
Environment and Planning A: Economy and Space,
13(3): 285-308.
Pavlínek, P., & Ženka, J. 2011. Upgrading in the
automotive industry: Firm-level evidence
fromCentral Europe. Journal of Economic
Geography: Economy and Space, 11(3): 559-586.
Pavlínek, P., & Ženka, J. 2016. Value creation and
valuecapture in the automotiveindustry: Empirical
evidencefrom Czechia. Environment and Planning
A: Economy and Space, 48(5): 937-959.
Pehrsson, A., & Pehrsson, A. 2020. Instantly upgrading a
foreign subsidiary's value-adding scope: Impact on
growth. Review of International Business and
Strategy, 30(1): 25-41.
Pelegrín, A., & Garcia-Quevedo, J. 2015. Which firms are
involved in foreign vertical integration? Revista de
Economía Aplicada, 23(69): 5-25.
Phelps, C. E. 1993. The methodologic foundations of
studies of the appropriateness of medical care. New
England Journal of Medicine, 329(17): 1241-1245.
Pike, A. 1998. Making performance plants from branch
plants? In situ restructuring in the automobile
industry in the United Kingdom. Environment and
Planning A: Economy and Space, 30(5): 881-900.
Pla-Barber, J., Linares, E., & Ghauri, P. N. 2019. The
choice of offshoring operation mode: A behavioural
perspective. Journal of Business Research, 103:
570-580.
Porter, M. E. 1986. Competition in global industries.
Boston, MA: Harvard Business Press.
Prahalad, C. K., & Doz, Y. L. 1981. An approach to
strategic control in MNCs. Sloan Management
Review, 22(4): 5-13.
Pu, M., & Soh, P.-H. 2018. The role of dual embeddedness
and organizational learning in subsidiary
development. Asia Pacific Journal of Management,
35(2): 373-397.
Puig, F., Portero, B., & González-Loureiro, M. 2017.
Clustering strategy and development of subsidiaries
in China. Economia e Politica Industriale, 44(2):
221-243.
Quesada-Vazquez, J., & Rodríguez-Cohard, J. C. 2019.
Subsidiary upgrading and regional innovation
policies: The case of valeo lighting systems Spain
and the andalusian plastic innovation centre.
Environment and Planning C: Politics and Space,
37(5): 908-928.
Reddy, P. 2000. The globalization of corporate R & D:
Implications for innovation systems in host
countries. London, UK: Routledge.
Roehrig, M. F. 1994. Foreign joint ventures in
contemporary China. New York, NY: St. Martin's
Press.
Rugman, A. M., & Verbeke, A. 2001. Subsidiary‐specific
advantages in multinational enterprises. Strategic
Management Journal, 22(3): 237-250.
Ryan, P., Buciuni, G., Giblin, M., & Andersson, U. 2020.
Subsidiary upgrading and global value chain
governance in the multinational enterprise. Global

24 Journal of Management and Business Research September
Strategy Journal, 10(3): 496-519.
Sartor, M., Orzes, G., Nassimbeni, G., Jia, F., & Lamming,
R. 2014. International purchasing offices: Literature
review and research directions. Journal of
Purchasing and Supply Management, 20(1): 1-17.
Sartor, M., Orzes, G., Nassimbeni, G., Jia, F., & Lamming,
R. 2015. International purchasing offices in China:
roles and resource/capability requirements.
International Journal of Operations & Production
Management, 35(8): 1125-1157.
Sayin, A. 2016. The effect of using relative and absolute
criteria to decide students' passing or failing a course.
Journal of Education and Training Studies, 4(9): 1-
9.
Schmid, S., & Schurig, A. 2003. The development of
critical capabilities in foreign subsidiaries:
Disentangling the role of the subsidiary’s business
network. International Business Review, 12(6): 755-
782.
Shi, Q., Qiu, W., & Fan, Y. 2020. Economic policy
uncertainty and the distribution of business
operations between parent companies and their
subsidiaries. Emerging Markets Finance and Trade,
56(2): 427-456.
Shih, S. 1996. Me - too is not my style : Challenge
difficulties, break through bottlenecks, create
values. Taipei: The Acer Foundation.
Sonn, J. W., & Lee, D. 2012. Revisiting the branch plant
syndrome: Review of literature on foreign direct
investment and regional development in Western
advanced economies. International Journal of
Urban Sciences, 16(3): 243-259.
Srinivasan, M. 2021. Psychometric characteristics of oral
pathology test items in the dental hygiene
curriculum—A longitudinal analysis. Dentistry
Journal, 9(5): 56.
Stöllinger, R. 2021. Testing the smile curve: functional
specialisation and value creation in GVCs. Structural
Change and Economic Dynamics, 56: 93-116.
Szalavetz, A. 2016. Global crisis and upgrading of MNCs'
manufacturing subsidiaries: A case study of Hungary.
Central European Business Review, 5(1): 37-44.
Szalavetz, A. 2018. Impact of greening on the upgrading
of manufacturing subsidiaries’ technological
capabilities–A Hungarian perspective. Journal of
East European Management Studies, 23(3): 426-
446.
Szwejczewski, M., Goffin, K., & Anagnostopoulos, Z.
2015. Product service systems, after-sales service and
new product development. International Journal of
Production Research, 53(17): 5334-5353.
Tan, S. 2009. Misuses of KR-20 and Cronbach's alpha
reliability coefficients. Egitim ve Bilim, 34(152):
101-112.
Tolentino, P. E., Janne, O. E. M., & Chen, P.-C. 2013.
Examining Strategy Diversity and Interdependence in
the MNC’s Subsidiaries and Their Functional
Activities. In G. Cook & J. Johns (Eds.), The
changing geography of international business: 200-
221. Basingstoke, UK: Palgrave Macmillan.
Tomiura, E., Ito, B., & Wakasugi, R. 2011. Offshore
outsourcing decision and capital intensity: Firm‐level
relationships. Economic Inquiry, 49(2): 364-378.
Ueki, Y., & Martínez, J. M. G. 2020. The impact of
engineers’ skills and problem-solving abilities on
process innovation. Economic Research-Ekonomska
Istraživanja, 33(1): 2018-2037.
Van den Bulcke, D., & Halsberghe, E. 1984. Employment
decision-making in multinational enterprises:
Survey results from Belgium. Multinational
Enterprises Program Working Paper No. 32, Geneva:
International Labour Organization.
Van Egeraat, C., & Breathnach, P. 2012. The drivers of
transnational subsidiary evolution: the upgrading of
process R&D in the Irish pharmaceutical industry.
Regional Studies, 46(9): 1153-1167.
Wei, Y., Liu, X., Wang, C., & Wang, J. 2012. Local
sourcing of multinational enterprises in China.
International Journal of Emerging Markets, 7(4):
364-382.
Williams, D., McDonald, F., Tüselmann, H. J., & Turner,
C. 2008. Domestic sourcing by foreign-owned
subsidiaries. Environment and Planning C:
Government and Policy, 26(1): 260-276.

2022 LIu and Chang 25
Winter, J. 2010. Upgrading of TNC subsidiaries: The case
of the Polish automotive industry. International
Journal of Automotive Technology and
Management, 10(2-3): 145-160.
Yamasaki, V., MacLennan, M., & Almeida, L. 2014.
Marketing innovation autonomy for subsidiaries in
new products development: A case study in fixed
telecom. International Journal of Logistics Systems
and Management, 17(3): 275-288.
Yang, C. H., & Hayakawa, K. 2015. Localization and
overseas R&D activity: the case of T aiwanese
multinational enterprises in C hina. R&D
Management, 45(2): 181-195.
Yeung, I. Y., & Tung, R. L. 1996. Achieving business
success in Confucian societies: The importance of
guanxi (connections). Organizational Dynamics,
25(2): 54-65.
Zanfei, A. 2000. Transnational firms and the changing
organisation of innovative activities. Cambridge
Journal of Economics, 24(5): 515-542.
Appendix 1 Estimating Reliability and Validity
Reliability
In this study, KR-20 instead of Cronbach’s alpha is employed to examine the internal consistency of the questionnaire
for nine business functions with the dichotomized response (yes = 1; no = 0;(Tan, 2009). These business functions include
production, quality control, R&D, marketing, after-sale service, procurement, HQ, finance, and others. The KR-20
reliability coefficient is calculated through the item variances and test variances taking place in the test. In this study, the
KR-20 values for 2008–2018 are found to be from 0.703 to 0.778 for the business function items of Taiwan-based
subsidiary samples. The scores meet a critical value of 0.7 (Daggett, 2007; Sayin, 2016; Srinivasan, 2021) to be deemed
acceptable in analysis.
Appendix Table 1 Reliability and Validity Scores Measured by KR-20 and KMO for 2008–2018
year
obs.
KR-20 Values
KMO
2008
261
0.703
0.762
2009
442
0.750
0.798
2010
268
0.732
0.763
2011
337
0.744
0.782
2012
408
0.756
0.808
2013
376
0.734
0.782
2014
253
0.749
0.769
2015
303
0.755
0.792
2016
273
0.733
0.744
2017
204
0.757
0.788
2018
240
0.778
0.819
total
3,174
--
Content Validity

26 Journal of Management and Business Research September
A test for content validity aims to ensure that these measure items include an adequate and representative set of items
to probe the research concept. The more the scale items represent the domain or universe of the concept being measured,
the greater the content validity. In this study, the nine business functions measured in the questionnaire are in line with the
measurements of several studies on subsidiaries’ functional upgrading or functional scopes (Burger et al., 2018;
Dzikowska, 2019).
Validity Test for Factor Analysis
Moreover, the KMO value is employed to test the appropriateness of the data for factorizing. The data are usually
accepted and deemed suitable for a factor analysis according to KMO values; for 2008–2018, all KMO values are above
0.70 (Appendix Table 1).

2022 LIu and Chang 27
在華海外子公司當地採購與價值鏈升級之研究
劉孟俊
中華經濟研究院
張東隆
長島大學
國立中興大學
論文編號：4001
收稿 2020 年10 月22 日 → 第一次修正 2021年4月15 日 → 第二次修正 2021年10 月14 日 → 正式接受 2022
年2月10 日
本研究整合子公司發展架構和微笑曲線邏輯，利用
2008
年至
2018
年在陸台商的營運資料，驗證其享有採
購自主權與於研發與市場行銷的價值鏈提升。本研究結果說明，在陸台商子公司在研發與市場行銷的價值鏈升級
上存在正向互動關係，同時肯定採購授權與當地採購程度有其價值鏈升級的積極效果。本研究根據子公司發展的
分析架構，歸納總部授權、當地資源條件和子公司主動作為等之實證結果，顯示在陸台商透過研發與市場行銷的
價值鏈升級強化其在地主國的競爭優勢。本研究成果提供子公司發展過程的新內涵，重新審視跨國企業的去全球
化價值鏈配置策略。
關鍵字：當地化、當地採購、海外子公司發展架構、微笑曲線邏輯、全球價值鏈。
作者
劉孟俊為中華經濟研究院 (CIER) 研究員兼大陸所所長。通訊作者張東隆為紐約長島大學管理學院市場行銷和國際商務教
授，本計劃進行期間獲邀為訪問學者，在國立中興大學進行研究分析和撰寫，地址：Post Campus. 720 Northern Blvd.
Brookville, NY 11548, USA.，電話：(516) 299-4195，Email: steven.chang@liu. edu.。

28 Journal of Management and Business Research September
劉孟俊為中華經濟研究院 (CIER) 研究員兼大陸所所長。其研究興趣包括中國經濟、科技管理、制度經濟和外商
直 接 投 資 與 貿 易 。 其 重 點 研 究 發 表 在 Research Policy, World Development, Economic Record, Food Policy,
Government Information Quarterly, Agricultural Economics, Scientometrics, Pacific Economic Review 等期刊。
Meng-Chun Liu is a Research Fellow and director of Mainland China Division at Chung-Hua Institution for Economic
Research (CIER), Taipei. His research interests include China's economy, technology management, institutional economy,
and FDI & Trade. His research has been published in Research Policy, World Development, Economic Record, Food
Policy, Government Information Quarterly, Agricultural Economics, Scientometrics, Pacific Economic Review, among
others.
張東隆（通訊作者）為紐約長島大學管理學院市場行銷和國際商務教授。其研究側重於跨國公司於全球行銷和科
技的策略佈局。相關研究發表在 Journal of World Business, International Marketing Review, Decision Support
Systems, Technological Forecasting and Social Change, and International Journal of Advertising 等期刊。
Tung-Lung Steven Chang is Professor of Marketing and International Business at the College of Management, Long
Island University - Post campus. He has centered his research on the strategic marketing and technology aspects of
multinationals' global expansion. His research has appeared in Journal of World Business, International Marketing Review,
Decision Support Systems, Technological Forecasting and Social Change, and International Journal of Advertising, among
others. The author greatly appreciate the support from National Chung Hsing University as a visiting scholar to conduct
this research project in Taiwan in 2019.