The Impact of Global Value Chain Restructuring on the OFDI Transformation of Manufacturing Industry: Evidence from China

Abstract

Global value chain (GVC) restructuring has important implications for the transformation of corporate outward foreign direct investment (OFDI), a process that is closely linked to sustainable economic development. Based on panel data from 2007 to 2021, this paper comprehensively applies the fixed effects model, mediation effects analysis, heterogeneity test, and regression analysis to explore how global value chain restructuring promotes the sustainable transformation of corporate OFDI, and it examines the role mechanisms of factor endowment and market scale expansion in the process. The conclusions are as follows: (1) Global value chain restructuring can promote manufacturing enterprises’ OFDI transformation. (2) Global value chain restructuring promotes the transformation of manufacturing OFDI through two channels: factor endowments and market scale. (3) Against countries’ different backgrounds, there are significant differences in the impacts of global value chain restructuring on enterprises’ OFDI. The research results of this paper can provide important insights for relevant government departments and enterprises in formulating management policies.

1. Introduction

Under the policies of reform and opening up, China has opened its doors and actively and deeply integrates itself into the construction of global value chains. It greatly promotes the development of the country’s manufacturing industry. Currently, Chinese manufacturing enterprises lack core technology, international brands, and advanced management experience. These firms rely on cheap domestic labor resources to embed themselves in GVCs. So it leads them to be stuck in labor-intensive processing and assembly stages, facing the dilemma of falling into value chain “low-end lock” [1]. Furthermore, according to data released by the Chinese Ministry of Commerce, China’s OFDI flow reached USD 177.29 billion in 2023, ranking third globally. Manufacturing is a crucial component of China’s OFDI. Its investment stock in the manufacturing sector shows a continuing upward trend. Particularly, the stock of OFDI in manufacturing reached USD 283.09 billion in 2023, accounting for 48.6% of the secondary industry and ranking at the forefront of investment. However, after expanding its OFDI, China now faces new challenges in “strengthening” its position. China’s OFDI in manufacturing focuses on the textile industry, electronic manufacturing, resource processing industry, and machinery industry. China’s manufacturing outward investment seems to be strong, but it is limited to downstream positions in the GVC. Moreover, it has a lower added value, relatively weak industrial breeding capabilities, slow infrastructure development, and various problems [2]. In addition, since the financial crisis of 2008, GVCs have undergone a period of deep adjustment and have begun to disintegrate and reconfigure. Furthermore, the growing trend of “de-globalization” and “anti-globalization” has led to a gradual contraction of the high rate of expansion of GVCs. The current frequent geopolitical conflicts are deteriorating the external environment of global industrial chains and supply chains. Meanwhile, multiple factors that overlap in time and space contribute to the complexity and diversity of GVCs. In this context, profound changes have emerged in the domestic and international environments and value chain conditions of Chinese manufacturing firms. Therefore, in a complex global environment, some manufacturing enterprises must confront the new choice of transforming their OFDI.

In the existing literature, OFDI studies have provided an in-depth analysis of related topics represented by investment transformation. Since Dunning (1981) [3] and Kojima (2010) [4] put forward the theoretical framework of OFDI, scholars have extensively explored the multifaceted factors affecting OFDI transformation. Specifically, enterprise traits such as international experience [5], tech innovation prowess [6], and ownership [7] strengths are closely tied to OFDI transformation. These factors significantly influence the mode and strategic choices of enterprise foreign investment. Additionally, the external environment is a key driver of the transformation of OFDI. It includes the international economic environment [8], bilateral investment agreements [9], and the market institutions of host countries [10]. With the deepening of global economic integration, domestic research has gradually focused on the transformation path of China’s OFDI. The research focuses on the “Belt and Road” Initiative [11] and foreign exchange management policies [12]. Moreover, it also includes international production capacity cooperation [13], cross-border financing facilitation [14], and the prevention of overseas investment risks. Policies and external environmental factors influence the transformation process of companies’ OFDI. However, existing research often overlooks the deep and ripple effects of GVC restructuring in this transformation process. GVC restructuring is crucial for China’s manufacturing industry to integrate into the global economy and achieve industrial upgrading. It also serves functions like technology transfer, market expansion, and resource allocation optimization [15]. Moreover, it deeply influences China’s manufacturing industry’s foreign investment trends via supply chain shifts, tech innovation, and evolving market needs [16]. Meanwhile, existing research has revealed a series of processes, including knowledge spillovers from TNCs in GVCs [17], technological learning in supply chain partnerships [18], and the construction of firms’ capabilities under the GVC governance model [19]. These factors have a profound impact on China’s manufacturing industry, influencing decisions regarding investment locations, modes of cooperation, and paths for industrial upgrading [20]. Furthermore, in the cross-cutting research field of GVCs and the transformation of manufacturing outward foreign direct investment, the existing literature has developed a multi-level theoretical dialog and empirical findings. From the perspective of GVCs, OFDI is regarded as a key path for the manufacturing industry to break through the “low-end lock-in” and realize the upgrading of the value chain, and its functioning mechanism is mainly reflected in three aspects: firstly, the reverse technology spillover effect, such as high-tech manufacturing industry obtaining core patented technology through direct investment in developed countries, which significantly improves the division of labor position in the GVC; secondly, the factor reconfiguration effect, and the capital and technology-intensive manufacturing industry will be transformed from low-end link to low-end link through OFDI, which is the most important factor in the development of GVC; and thirdly, the market integration effect, such as equipment manufacturing industry through mergers and acquisitions of host country brand channels, from OEM export to the transformation of the operation of the independent brand.

Current research focuses mainly on direct effects analysis. GVC restructuring impacts global market demand and the competitive landscape by altering factor endowments and expanding the market scale. In addition, it indirectly affects the investment decisions of Chinese manufacturing firms. However, this issue lacks thorough exploration. Particularly, it is crucial to investigate how Chinese manufacturing transitions from low-end processing to engaging in high-end design and brand marketing. The driving mechanisms, challenges, and response strategies behind this shift require further exploration [21].

The purposes of this research are as follows: (1) We aim to reveal the structure adjustment mechanism of China’s manufacturing OFDI. We first analyze the trend of GVC restructuring, the current state of development, and the characteristics of China’s manufacturing OFDI. And this then reveals how the restructuring of GVCs affects the transformation of China’s manufacturing OFDI. (2) We aim to examine how factor endowments and market scale mediate the impact of GVC restructuring on the OFDI transition of Chinese manufacturing firms. Both factor endowments and market scale play crucial mediating roles. We also aim to elucidate the specific contributions of factor endowments and market scale in this process. (3) We are committed to exploring the different impacts of GVC restructuring on the OFDI transformation of Chinese manufacturing firms. These explorations are based on different countries. Both developed and developing countries experience varying degrees of reorganization of GVCs. This paper examines how the restructuring of GVCs in countries with different levels of development affects the transformation of China’s manufacturing OFDI.

The marginal contributions of this paper are as follows: (1) We enhance the research on the impact of GVC restructuring on the transformation of manufacturing OFDI. The existing literature mostly examines the influence of policy factors on the transformation of manufacturing OFDI. In addition, this paper analyzes how GVC restructuring affects the transformation of manufacturing OFDI. It focuses on factor endowments and market scale. Moreover, it not only enriches the research content of the existing literature but also presents new research findings. (2) We expand the application scenarios of relevant empirical research methods. We use mediation effects tests and heterogeneity tests to examine how GVC restructuring affects the transformation of manufacturing OFDI. This paper ventures into uncharted territory. Scholars rarely use such methods to study this issue. Therefore, our research not only verifies the impact path of GVC restructuring but also broadens the application scenario of empirical research methods in the field of sustainable economy and investment. (3) This paper inspires new policy directions for sustainable economic development. The research in this paper shows that global value chain restructuring has a facilitating effect on China’s manufacturing OFDI. However, it may be affected by regional heterogeneity, and this paper incorporates these differences into the analytical framework to provide a more accurate reference for enterprises to formulate OFDI strategies. It can also help the government to formulate supportive policies.

The structure of this paper is as follows: (1) Literature review. This section reviews the relevant research literature on GVC restructuring and OFDI. (2) Theoretical analysis and research hypotheses. Based on a theoretical analysis of the relationships between variables, this section also outlines four research hypotheses. (3) Research design. This section outlines the design of econometric models, variable measurements, and data sources. (4) Results. This section includes baseline regressions, mediation effects tests, heterogeneity tests, threshold tests, and robustness tests. (5) Conclusions, implications, and research limitations. This section mainly covers conclusions, implications, research limitations, and future research directions.

2. Literature Review

2.1. Research on Global Value Chain Restructuring

The scholar who first proposed the concept of GVC restructuring is Gereffi (1994). He considers GVC restructuring as, “Changes in comparative advantage factors in pre-existing GVCs lead to contraction and relocation of production over time”.

The related research mainly unfolds from three dimensions. Firstly, at the enterprise level. Enterprises originally at the low end of the GVC improve the efficiency of their marketing management and establish globally influential brands. Subsequently, this action promotes their transformation and upgrading with the aim of upgrading their position in GVCs [22]. Secondly, from the industry perspective. Value chain restructuring optimizes and upgrades industry structures and integrates different industries, and it collectively promotes the overall restructuring of the value chain [23]. Lastly, from the national perspective. Emerging economies strive to build value chain systems led by their own country by fully utilizing their resource advantages or participating in regional cooperation mechanisms. This aims to reshape the layout of the GVCs so as to increase their voice and position [24].

GVC restructuring is the result of a combination of internal and external factors. In terms of external drivers, GVCs are facing significant impacts from environmental change, technological innovation, and institutional adjustments. Furthermore, international trade protectionism, geopolitical conflicts, and friction among major powers have led to its stagnation [25]. Simultaneously, new technologies and industrial revolutions are changing the ways of production being organized. In addition, the limitations of traditional multilateral systems and global governance models are driving significant adjustments in international economic and trade rules. These factors are key external forces influencing the reorganization of GVCs. However, the core direction is more influenced by internal factors on the supply and demand sides [26]. On the demand side, the fragility of the global division of labor reduces market risk preferences. Simultaneously, consumer demand is shifting towards green and personalized products. This shift alters the manufacturing industry’s global dominance, contributes to supply chain shifts, and affects GVC restructuring [27]. On the supply side, national endowments determine comparative advantages in the international division of labor [28]. Moreover, a specialized division of labor promotes global resource integration [29]. The input–output structure influences value’s addition [30]. These factors play a crucial role in GVC restructuring.

2.2. Impact of Global Value Chain Restructuring on the Transformation of OFDI in the Manufacturing Industry

GVC restructuring is profoundly affecting the transformation path of OFDI by manufacturing enterprises. This impact is mainly reflected in three dimensions: Firstly, although the OFDI of manufacturing enterprises presents the characteristics of geographical dispersion, its essence is based on a functional integration layout under the reconstruction of GVCs and the realization of a specialized division of labor through transnational production networks. Secondly, in the process of GVCs’ reconstruction, the developed multinational enterprises that master the core link extend the value chain link through the advantage of governance, dominating the pattern of the Interests Payment Chain, which compels the manufacturing enterprises to enhance their own position in the value chain through strategic OFDI [31]. Finally, GVCs’ reconfiguration shifts the logic of the division of labor from comparative advantage in final products to comparative advantage in the level of processes. This prompts manufacturing firms to make reverse investments based on the global distribution of costs, technology, and other factors, especially through intermediate goods trading networks, to realize the cross-border allocation of production links. As a result, the division of labor pattern of GVCs makes it possible for developing countries with comparative advantages in terms of process to make a reverse investment in developed countries. The mechanisms of the impact of GVC restructuring on the OFDI transformation of manufacturing firms are reflected in the following aspects:

2.2.1. The Strategic Need to Build a National Global Value Chain System

GVCs’ reconstruction pushes manufacturing enterprises to adjust their OFDI mode from passive embedding to the active construction of their own value chain system. Developed countries have long dominated the division of labor in the GVC, controlling high value-added segments such as research and development and marketing. This has led to the passive integration of developing countries into the international production network, mainly through the undertaking of FDI, and to the plight of “low-end lock-in” [32]. In order to break through this situation, developing countries’ manufacturing enterprises are changing their OFDI strategies. They are choosing to shift from decentralized investment to cluster investment, extending the industrial chain through transnational production collaboration and gradually building autonomous and controllable regional and GVC networks.

2.2.2. Seeking Optimal Economies of Scale

GVC restructuring pushes manufacturing firms to achieve optimal scale effects through differentiated OFDI strategies. On the one hand, manufacturing enterprises are adopting buyer-driven investment in developed country markets, especially in traditional labor-intensive areas such as garments, toys, and footwear. They obtain market scale effects by embedding themselves in local sales networks and ultimately achieve brand upgrading and value chain upgrading by leveraging the global marketing channels of dominant firms. On the other hand, manufacturing firms focus on producer-driven investment in developing countries and obtain cost scale effects by optimizing the global production layout. This differentiated OFDI strategy enables enterprises to allocate different production segments to the most comparative advantageous locations according to the characteristics of the host countries. They not only expand the scale of demand through market-oriented investment in developed countries but also optimize the scale of production with the help of cost-oriented investment in developing countries, and they ultimately form a synergistic and optimal transnational production network in the context of GVC restructuring [33], thus realizing the transformation and upgrading from a single cost-oriented to a composite strategy.

2.2.3. Forcing Mechanisms of Developed Countries

The forcing mechanism of developed countries under GVC restructuring is profoundly affecting the strategic transformation of OFDI by manufacturing enterprises. Developed countries enhance their control over the higher end of GVCs through “re-industrialization” strategies and high-standard trade agreements [34]. On the one hand, the wave of advanced manufacturing revival led by the United States has prompted the United Kingdom, France, and Japan to step up their control over the R&D and innovation links on the left side of the ”smile curve”; on the other hand, they are locking the right-hand side of service and trade advantages through high-standard trade frameworks, such as the TPP and the TTIP. This “both ends of the blockade” strategy forced China, as the representative of developing countries’ manufacturing enterprises, to change the traditional development path, from passive acceptance of technological spillover to strategic OFDI through an active breakthrough: (1) reverse investment in developed countries to obtain innovative elements and break the technology blockade; (2) mergers and acquisitions of high-tech enterprises to realize the value chain leap; and (3) the establishment of overseas R&D centers to absorb cutting-edge technology.

2.3. Literature Gaps

In conclusion, scholars study the GVC restructuring and OFDI in manufacturing. They lay an important foundation for our learning. However, their research has the following shortcomings: (1) Limited research perspectives. Few scholars have included factor endowments and market scale in the study of the impact of GVC restructuring on the OFDI transformation of enterprises. (2) Lack of empirical research. Few scholars have used regression methods to study the impact of GVC restructuring on firms’ OFDI transformation through factor endowments and market scale. (3) Insufficient specificity in research. Few scholars have examined how the reorganization of GVCs affects the OFDI transition of different countries’ categories of firms. To address these shortcomings in scholarly research, this paper delves into the aforementioned deficiencies. Consequently, this paper will enrich the relevant theoretical research content. Our goal is to provide valuable insights into the transformation of China’s manufacturing OFDI in the context of GVC restructuring.

3. Theoretical Analysis and Hypotheses

3.1. Effects of Global Value Chain Restructuring on the Transformation of OFDI in the Manufacturing Industry

The transformation of manufacturing OFDI involves a systemic change encompassing investment fields, investment locations, and other aspects. Meanwhile, at the center of this transformation is the expansion of overseas investment by manufacturing firms. They shift focus from traditional labor-intensive sectors to technology-intensive and brand-driven industries [35]. Simultaneously, the location selection for manufacturing OFDI is shifting from developing countries to developed countries. The rising labor costs and resource constraints in developing countries contribute to this trend [36]. Developed countries have increasing advantages in technology innovation, market systems, and legal environments [37]. Additionally, investing in developed countries benefits companies by directly absorbing advanced technology, management knowledge, and market insights. Moreover, it accelerates the transformation and upgrading process of companies and helps them avoid trade barriers while exploring high-end markets.

On the one hand, technology-seeking OFDI becomes crucial, with enterprises acquiring advanced technologies and upgrading their position in GVCs through cross-border mergers and acquisitions or setting up R&D facilities in developed countries [38].

On the other hand, the “decoupling” of the U.S. supply chain from China has prompted Chinese firms to maintain global supply chain connectivity through “corridor country” strategies, such as setting up production bases in Southeast Asia and Europe to avoid trade barriers. In addition, cost-containing OFDI has shifted to low-cost developing countries to maintain the integrity of the industrial chain and optimize the global resource allocation. Overall, the restructuring of GVCs is driving a shift in manufacturing OFDI towards a technology orientation, risk diversification, and regional synergies [39].

Hypothesis 1 (H1).

GVC restructuring positively promotes the transformation of China’s manufacturing industry in OFDI.

3.2. The Mediating Effect of Global Value Chain Restructuring Affecting the Transformation of Manufacturing OFDI

Firstly, factor endowment can promote the transformation of manufacturing enterprises’ OFDI. By fully considering and utilizing the factor endowments of different countries, manufacturing firms are able to choose their investment targets more effectively and achieve transformation and upgrading [40]. In addition, in GVC restructuring, firms can adjust their investment strategies according to their factor advantages, such as labor, capital, and technology. Specifically, labor-intensive firms may shift production to regions with lower labor costs to reduce production costs. In addition, capital-intensive firms may invest in high-end equipment and technology to improve production efficiency and quality. Furthermore, technology-intensive firms may seek technological innovation and cooperate with local R&D institutions. Meanwhile, the optimal allocation of factor endowments can also promote the relative position of manufacturing firms in GVCs and accelerate the pace of their OFDI transformation [41].

Secondly, the expansion of the market scale can promote the transformation of manufacturing OFDI. Specifically, enterprises can gain more sales opportunities and profit margins by entering countries with large markets through OFDI. Meanwhile, the expansion of market scale can bring about economies of scale and enhance competitiveness [42]. In addition, the economy of scale effect can also prompt enterprises to innovate in technology, management, and other aspects. Furthermore, countries with large markets are usually accompanied by better infrastructure and resource advantages, which provide better production conditions and a development environment for manufacturing enterprises [43]. By utilizing local resources and infrastructure, enterprises can reduce production costs and improve efficiency, thus realizing the transformation and upgrading of OFDI. Based on the above analysis, the following hypothesis is proposed:

Hypothesis 2a (H2a).

GVC restructuring positively impacts manufacturing enterprises’ OFDI by optimizing factor endowments.

Hypothesis 2b (H2b).

GVC restructuring positively impacts manufacturing enterprises’ OFDI by expanding the market scale.

3.3. Heterogeneous Effects of Global Value Chain Restructuring on the Transformation of Manufacturing OFDI

The restructuring of GVCs in developed countries actively contributes to the transformation of OFDI by manufacturing firms. As core nodes, developed countries possess advanced technologies, mature market mechanisms, and rich experience. Therefore, manufacturing enterprises investing directly in these countries can directly access high-end resources, accelerate technological innovation, improve management levels, and expand markets, thus deepening the transformation of investment [44]. Specifically, in terms of technology, the leading innovation and R&D capabilities of developed countries allow manufacturing firms to learn by investing in proximity. In addition, in terms of management, it can improve the efficiency of enterprise management by learning from the mature market mechanism and management experience of developed countries. Furthermore, in terms of market, the strong consumption capacity and diversified demands of developed countries provide enterprises with a broad space for development [45]. Based on the above analysis, the following hypothesis is proposed:

Hypothesis 3 (H3).

The restructuring of GVCs in developed countries has contributed more significantly to the OFDI transformation of manufacturing firms.

Based on the analysis above, we have constructed the theoretical model of this paper. There exists an inextricable relationship between the restructuring of GVCs and the transformation of manufacturing OFDI. GVC restructuring can drive the transformation of manufacturing OFDI. Moreover, factor endowments and market scale can facilitate the transformation of manufacturing OFDI through GVC restructuring.

As depicted in Figure 1, in order to further validate the theoretical model, we will integrate relevant research data and employ empirical methods. We will utilize baseline regression models, mediating effects, threshold effect analysis, and regional heterogeneity analysis. Furthermore, we will conduct robustness tests. These empirical methods are used to test the relationships proposed in the theoretical model, thus making the conclusions more scientific and accurate.

4. Research Designs

4.1. Variables

4.1.1. Explanatory Variable

The quantification of global value chain restructuring can be measured in various ways, broadly categorized into macro and micro levels. At the macro level, global input–output tables are mainly used to calculate the length of GVCs, trade value added, and related data to explore the overall trend and position of countries/sectors in the GVCs. The micro level focuses on firm-level data. By studying the participation of firms in GVCs, we can measure the efficiency of participants in cross-border production activities. In order to further optimize the index algorithm to better fit the characteristics of restructuring in the era of the digital economy, we adopt Liu and Liu’s methodology (2023) [46]; we use the Global Value Chain Relative Position Index (Gvc-dbc) and the Global Value Chain Bilateral Cooperation Index (Gvc-rpi) to depict GVCs’ restructuring, with specific measurement methods as shown in Equations (1) and (2):

$$\mathrm{GVC}\text{-}{\mathrm{rpi}}_{\mathrm{c} \mathrm{r}}=\left[\ln \right(1+{\displaystyle \frac{IVcr}{Ec+Er}})-\ln (1+{\displaystyle \frac{FVcr}{Ec+Er}}\left)\right]\times 100$$

(1)

$$\mathrm{GVC}\text{-}{\mathrm{rpi}}_{\mathrm{c} \mathrm{r}}=\left({\displaystyle \frac{IVcr}{Ecr}}+{\displaystyle \frac{FVcr}{Ecr}}\right)\times 100$$

(2)

In Equation (1), IV_cr represents the value added by country r exported to country c and re-exported by country c. FV_cr represents the value added from country c in the exports of country r. Ec denotes the total exports of the partner country, while E_r represents the total exports of China. A higher value of this index indicates that country r has a higher GVC position relative to country c. In Equation (2), IV_cr represents the value added by country r exported to country c and re-exported by country c. The ratio of this value to the sum of the exports of the two countries can to some extent represent the dependency of country c on country r. FV_cr represents the value added from country c in the exports of country r. Similarly, the ratio of this value to the sum of the exports of the two countries can to some extent represent the dependency of country c on country r. The greater the mutual dependency between two countries, the closer their economic and trade relations, indicating a higher level of bilateral cooperation in GVCs.

4.1.2. Explained Variable

The explanatory variable of this study is manufacturing OFDI transformation. We use the scale of China’s OFDI in high-tech manufacturing to indicate the transformation of China’s manufacturing OFDI, and this indicator captures both the expansion of investment scale and the transformation characteristics of upgrading the investment structure to technology-intensive areas [47]. Because OFDI stock data is more stable than flow data, it is less likely to be subject to fluctuations. Therefore, we choose China’s high-tech manufacturing OFDI stock as an explanatory variable. However, it is currently impossible to directly obtain the data of China’s high-tech manufacturing OFDI. Therefore, the product of the value added of the high-tech manufacturing industry of the partner country as a proportion of the country’s GDP and the stock of China’s high-tech OFDI in the country is used to express China’s direct investment in high-tech manufacturing industry in the country [48], that is

$${\mathrm{OFDI}}_{\mathrm{rt}}={\mathrm{OFDI}}_{\mathrm{rt}}\times {\displaystyle \frac{{MVA}_{rt}}{{GDP}_{rt}}}$$

(3)

where OFDI_rt denotes the amount of OFDI stock from China’s high-tech manufacturing sector to partner country r in period t. MVA_rt denotes the value added of the high-tech manufacturing sector in partner country r in period t. GDP_rt denotes the GDP of member country r in period t. This approach is used to estimate the scale of OFDI from China’s high-tech manufacturing sector to other countries. This approach is used to estimate the scale of China’s OFDI in high-tech manufacturing to other countries.

4.1.3. Mediating Variable

The mediating variables in this paper are factor endowment and market scale. Theoretical analysis suggests that factor endowments and market scale may mediate the role of GVC restructuring in the transformation of Chinese manufacturing firms’ OFDI. Therefore, we will empirically investigate the relationships among them.

(1)

Factor endowment. The level of factor endowment serves as a composite indicator, encompassing technological factor endowment (X1), labor factor endowment (X2), and capital factor endowment (X3). This paper uses the ratio of R&D expenditure to a country’s GDP to reflect the level of technological factor endowment. The labor factor endowment level is represented by the ratio of employment in public units and private entities to the total year-end population in each country. The capital factor endowment level is indicated by the GDP per capita at the end of the year for each country [49]. These sub-indicators are then input into the following calculation formula:

$$Di{g}_{it}=\ln \left[\left(X_{1it}+X_{2it}+X_{3it}\right)/3\right]$$

(4)

In the equation, i and t represent the country and year, respectively. Dig represents the level of factor endowment. X1, X2, and X3 represent technological, labor, and capital factor endowment, respectively.

(2)

Market scale. Generally, the larger the market scale of the host country, the greater the competitive pressure. Therefore, Chinese manufacturing firms increase their efforts in innovation and industrial upgrading when investing. This enhances China’s position in the value chain and promotes the transformation of China’s manufacturing OFDI. This paper uses the growth rate of a country’s Gross Domestic Product as an indicator of market scale [42].

4.1.4. Control Variables

In addition to bilateral cooperation and GVC positioning, we identify multiple factors influencing the transformation of manufacturing firms’ OFDI. To address endogeneity, we control for the following variables: (1) Population Density (Popden). Measured as population per square kilometer, a higher density indicates greater market potential and economies of scale, reducing unit production costs [50]. (2) Infrastructure (Mcs). Represented by mobile cellular subscriptions per 100 people, robust infrastructure enhances a market’s attractiveness, lowers logistics costs, and improves operational efficiency [50]. (3) Government Intervention (Gov). Measured by the ratio of government final consumption expenditure to GDP, it reflects the host country’s investment environment stability and influences firms’ investment strategies [50]. (4) Gross Fixed Capital Formation (Gfc). Total fixed capital formation indicates the host country’s capacity for industrial upgrading, promoting manufacturing sector transformation [51]. (5) Internet Coverage (Int). Measured by internet users per ten thousand people, a higher coverage facilitates access to market information, policy insights, and industry trends, aiding firms in evaluating investment opportunities and strategies [52].

These controls ensure a comprehensive analysis of factors shaping manufacturing OFDI transformation.

4.2. Model Construction

In order to study the impact of GVC restructuring on the transformation of China’s manufacturing OFDI, we constructed the following econometric model:

$$L{\mathrm{nOFDI}}_{\mathrm{it}}={\beta }_0+{\beta }_{1}G{\mathrm{vcr}}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{l}}^{\mathrm{x}}{\beta }_{\mathrm{x}}{X}_{\mathrm{it}}+{\lambda }_{\mathrm{i}}+{\mu }_{\mathrm{t}}}+{\epsilon }_{\mathrm{it}}$$

(5)

In model (5), OFDI_it represents the scale of China’s manufacturing direct investment in country i at time t. Gvcr_it denotes the level of GVC restructuring in country i at time t. X_it represents a set of control variables. β₀ represents the intercept term in the model. β₁ is the coefficient indicating the impact of GVC restructuring on the transformation of China’s manufacturing industry. β_n represents the coefficients of the control variables. λ_i and μ_t denote individual and time fixed effects. ε_it is the random error term.

In order to explore the mediating effects of factor endowment and market scale on the transformation of China’s manufacturing OFDI under the restructuring of GVCs, we adopt the mediation test proposed by Jiang [53]. And we establish a model of the mediating effects of factor endowment and market scale on the transformation of OFDI in Chinese manufacturing industry under GVC restructuring:

$$K_{\mathrm{it}}={\alpha }_0+{\alpha }_{1}G{\mathrm{vcr}}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{l}}^{\mathrm{x}}{\alpha }_{\mathrm{x}}{X}_{\mathrm{it}}+{\lambda }_{\mathrm{i}}+{\mu }_{\mathrm{t}}}+{\epsilon }_{\mathrm{it}}$$

(6)

where K_it represents factor endowments and market scale, α₁ indicates the direct impact of GVC restructuring on the mediating effect, and the other variables are the same as in model (5).

4.3. Description of the Data Source

As the OECD input–output tables are only updated until 2021, this paper primarily focuses on data from 2007 to 2021. During this period, China has undergone significant changes in its relative position in GVCs, and bilateral cooperation in GVCs and OFDI underwent significant changes.

The relative position in GVCs and bilateral cooperation indicators are sourced from the OECD input–output tables and calculated by the authors. Indicators of the scale of China’s OFDI in high-technology manufacturing industries are derived from the Statistical Bulletin of China’s Outward Investment. Furthermore, factor endowments, market scale, trade openness, population density, infrastructure, government intervention, total fixed capital, and internet coverage are from the WDI database. The sample covers 41 significantly different economies across the globe, including developed industrial countries, emerging markets, transition economies, and small open economies, and it is geographically balanced across Europe, Asia, the Americas, and Oceania and is representative of the diverse patterns of GVC participation at different stages of development, industrial structures, and regional characteristics.

Table 1. Descriptive statistics of the primary variables.

Categories of Variables	Variables	Observations	Max.	Min.	Mean
Explained variable	LnOFDI	615	18.77	5.770	13.69
Explanatory variable	GVC-rpi	615	3.890	0.190	2.640
	GVC-dbc	615	4.240	0.820	3.440
Control variable	LnPopden	615	7150	1.670	320.4
	LnMcs	615	5.770	2.980	4.740
	Gov	615	0.380	0.02	0.230
	LnGfc	615	29.23	20.66	25.36
	Int	615	17.29	0.01	3.640

Note: The authors generated this method by using Stata 17.

provides a descriptive statistical analysis of the variables, showcasing the data characteristics of all variables in this paper.

In order to minimize the interference of data scales, the authors have applied logarithmic transformations to variables with large magnitudes. Specifically, LnOFDI, Lnpopden, Int, and Lngfc exhibit high standard deviations of 2.560, 1480, 2.420, and 1.590, respectively. These variables demonstrate significant standard deviations, indicating substantial disparities and fluctuations among China’s OFDI in high-technology manufacturing, population densities in various countries, internet coverage rates, and total fixed capital amounts.

The standard deviations of GVC-rpi, GVC-dbc, Gov, and LnMcs are relatively low, with values of 0.700, 0.560, 0.070, and 0.270, respectively. These variables exhibit small standard deviations, indicating minimal fluctuations in China’s bilateral cooperation in GVCs, GVC relative position index, government intervention levels in various countries, and infrastructure.

5. Results

5.1. Baseline Regression Analysis

The results of the impact of GVC restructuring on the OFDI transformation of China’s high-tech manufacturing sector are presented in

Table 2. Baseline regression results.

Variables	LnOFDI	LnOFDI	LnOFDI	LnOFDI
	(1)	(2)	(3)	(4)
GVC-dbc			0.338 ***	1.428 ***
			(0.130)	(0.175)
GVC-rpi	0.402 ***	0.848 ***
	(0.108)	(0.157)
Lngfc	0.181	0.162	0.240	0.397 **
	(0.145)	(0.201)	(0.148)	(0.198)
Lnpopden	−0.002 *	0.000	−0.002	0.000
	(0.001)	(0.001)	(0.001)	(0.001)
Lnmcs	−0.861 ***	0.577 *	−0.812 ***	0.806 **
	(0.242)	(0.321)	(0.248)	(0.313)
Gov	−3.672 *	1.935	−3.774 *	1.864
	(2.065)	(2.762)	(2.080)	(2.678)
Int	−0.016 ***	0.056 ***	−0.016 ***	0.046 ***
	(0.004)	(0.004)	(0.004)	(0.004)
Constant	14.632 ***	0.450	12.739 ***	−8.544 *
	(3.279)	(4.562)	(3.499)	(4.621)
N	615	615	615	615
R2	0.939	0.860	0.938	0.868
Area fixed effect	YES	YES	YES	YES
Year fixed effect	YES	NO	YES	NO

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

. It is shown in models (2) and (4) that when we do not control for year fixed effects and only control for region fixed effects, the impacts of GVC-rpi and GVC-dbc on the OFDI transformation of China’s high-tech manufacturing sector are both significantly positive at the 1% level. After we control for year and region fixed effects, the coefficients of the impacts of GVC reconstruction and OFDI transformation in China’s high-tech manufacturing industry are shown in models (1) and (3). The data shows that the regression results decrease compared with models (2) and (4) but are still significantly positive at the 1% statistical level. This indicates that GVC-rpi and GVC-dbc can effectively promote the transformation of OFDI in China’s high-tech manufacturing industry. In other words, GVC restructuring may affect the transformation of China’s manufacturing OFDI.

The coefficient on infrastructure is negative and significant at the 1% level. This may be due to the fact that better infrastructure in host countries tends to signal an optimized investment environment and attract more OFDI. However, China’s manufacturing industry has a relative disadvantage in innovation and productivity, so its investment is mostly concentrated in low-end industries. In the long run, this may hinder the transformation and upgrading of its OFDI. In addition, the coefficient of degree of government intervention is negative and significant at the 10% level. The presumed reason is that countries with high government intervention tend to lack the market environment and policy support for innovation incentives. It restricts Chinese manufacturing firms’ access to technological innovation resources and then constrains the transformation of their OFDI. Meanwhile, the coefficient of internet coverage is negative and significant at the 1% level. The reason for this may be that the high internet coverage leads enterprises to favor short-term gains and market share. Thus, enterprises ignore long-term structural adjustment and the enhancement of innovation capacity, which hinders their transformation into high value-added areas.

Investment in fixed assets will not have an impact on the transformation of Chinese manufacturing OFDI. This may be due to the fact that the host country’s fixed asset investment is mainly concentrated in areas not directly related to manufacturing, such as infrastructure construction and services. Even if the host countries have fixed asset investments in the manufacturing sector, if the direction of their investments is not consistent with the transformation direction of Chinese manufacturing firms, the impact of these investments on the transformation of Chinese firms will be limited. In addition, the population density of the host country does not affect the transformation of Chinese manufacturing OFDI. It is speculated that the direction of transformation of Chinese manufacturing OFDI mainly focuses on technology-intensive or knowledge-intensive industries. These industries are relatively less dependent on the amount of labor, so the population density of the host country may not be a major factor affecting the transformation.

5.2. Mediation Analysis

In model (6), intermediating variables are characterized. If this coefficient is significant, it suggests that factor endowment and market scale play a mediating role in the transformation of China’s manufacturing OFDI under the reconstruction of GVCs. However, while the above mediation model has been theoretically confirmed, it still lacks sufficient empirical evidence. Based on this, the following model is constructed, incorporating factor endowments and market scale into the mediation effect test model for regression, with results reported in

Table 3. Examining the mediating effect of factor endowments.

Variables	Dig		LnOFDI
	(1)	(2)	(3)	(4)
GVC-dbc	0.378 **		0.267 **
	(0.192)		(0.126)
GVC-rpi		0.321 **		0.342 ***
		(0.159)		(0.104)
Dig			0.188 ***	0.185 ***
			(0.028)	(0.028)
Constant	9.291 *	11.662 **	10.470 ***	11.736 ***
	(5.423)	(5.140)	(3.557)	(3.359)
N	615	615	615	615
R2	0.376	0.377	0.943	0.944
Control variable	YES	YES	YES	YES
Area fixed effect	YES	YES	YES	YES
Year fixed effect	YES	YES	YES	YES

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

and

Table 4. Examining the mediating effect of market scale.

Variables	Scale		LnOFDI
	(1)	(2)	(3)	(4)
GVC-dbc	0.523 **		0.293 **
	(0.264)		(0.129)
GVC-rpi		0.470 **		0.363 ***
		(0.219)		(0.107)
Scale			0.085 ***	0.083 ***
			(0.021)	(0.021)
Constant	−0.320	2.866	12.244 ***	13.658 ***
	(7.466)	(7.074)	(3.636)	(3.428)
N	615	615	615	615
R2	0.340	0.341	0.940	0.941
Control variable	YES	YES	YES	YES
Area fixed effect	YES	YES	YES	YES
Year fixed effect	YES	YES	YES	YES

Note: **, and *** denote significance at the 5% and 1% levels, respectively.

, respectively [54].

$$L{\mathrm{nOFDI}}_{\mathrm{it}}={\delta }_0+{\delta }_1{K}_{\mathrm{it}}+{\delta }_x{\displaystyle \sum _l^{\mathrm{x}}{X}_{\mathrm{it}}+{\lambda }_{\mathrm{i}}+{\mu }_{\mathrm{t}}}+{\epsilon }_{\mathrm{it}}$$

(7)

The data in Table 3 show that the effect of factor endowment on the transformation of OFDI in Chinese manufacturing industries is significant at the 1% level. In addition, the coefficient of model (3) decreases compared with model (1). This implies that factor endowments mediate the transformation of manufacturing OFDI in the reorganization of GVCs. Hypothesis 2a is confirmed.

Differences in factor endowments form the basis of the international division of labor and guide the flow of Chinese manufacturing investment [54]. By investing in countries with abundant resources, low-cost labor, or high-end technology, enterprises can reduce costs and improve their production efficiency. Meanwhile, it helps to promote industrial upgrading. For example, China’s manufacturing investment in Southeast Asia is precisely focused on the region’s abundant labor resources and good manufacturing base. Moreover, factor endowments enhance the position of China’s manufacturing sector in GVCs. Enterprises’ investing in countries with superior high-end technology and innovation environments can bring them closer to the international technological frontier and enhance their technological level and innovation capacity.

Observe that Table 4 reveals that the impact of market scale on the transformation of China’s manufacturing industry through OFDI is statistically significant at the 1% level. Furthermore, the coefficient of model (3) decreases compared to model (1), suggesting that the market scale of the host country mediates the reorganization of GVCs and the transformation of China’s manufacturing industry through OFDI. Hypothesis 2b is supported.

In the restructuring of the global value chain, China’s manufacturing OFDI not only directly guides the flow of investment through the vast market space and abundant consumer demand created by the host country’s huge market scale, but it also creates knowledge spillover and technical cooperation effects by attracting high-end technology, talents, and management expertise, thus motivating enterprises to climb from low value-added processing to high value-added design, research and development, and brand marketing segments, and finally this will stimulate enterprises to move from low value-added processing to high value-added design, R&D, and brand marketing and ultimately promote production growth.

5.3. Regional Heterogeneity Analysis

This paper further analyzes regional heterogeneity. Specifically, it provides insights into whether the restructuring of GVCs in countries with different levels of economic development will have different transformative effects on OFDI by manufacturing firms. Following the IMF classification method, we categorize sample countries into advanced economies and developing economies [55]. The specific results are shown in

Table 5. Examining the mediating effect of market scale estimation.

Variables	Developed Country		Developing Country
	LnOFDI	LnOFDI	LnOFDI	LnOFDI
	(1)	(2)	(3)	(4)
GVC-dbc	0.457 **		0.400 **
	(0.190)		(0.187)
GVC-rpi		0.439 ***		0.392 **
		(0.137)		(0.170)
Lngfc	0.080	−0.033	0.206	0.196
	(0.228)	(0.223)	(0.167)	(0.166)
Lnpopden	0.037 ***	0.034 ***	−0.002 **	−0.002 **
	(0.011)	(0.011)	(0.001)	(0.001)
Lnmcs	−1.167 **	−1.366 ***	−0.211	−0.282
	(0.483)	(0.451)	(0.248)	(0.247)
Gov	−0.294	−1.248	−5.375 ***	−4.751 ***
	(4.364)	(4.305)	(1.626)	(1.659)
Int	−0.019 **	−0.018 **	−0.003	−0.003
	(0.008)	(0.008)	(0.005)	(0.005)
Constant	17.082 ***	21.394 ***	7.528 *	8.129 **
	(6.204)	(5.622)	(4.212)	(4.114)
N	405	405	210	210
R2	0.924	0.925	0.935	0.935
Area fixed effect	YES	YES	YES	YES
Year fixed effect	YES	YES	YES	YES

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

, models (1)–(4). Models (1) and (2) report on the impact of China’s manufacturing industry investing in advanced countries under GVC restructuring. The coefficient for the impact of China’s manufacturing industry investing in advanced countries under GVC restructuring is positive and significant at the 1% level. This indicates that the restructuring of GVCs in advanced countries can promote the transformation of China’s manufacturing industry’s OFDI. Observing the data from models (3) and (4), it is evident that the coefficient for the impact of China’s manufacturing industry investing in developing countries under GVC restructuring is positive and significant at the 1% level. This suggests that investing in developed countries can facilitate the transformation of China’s manufacturing OFDI under the restructuring of GVCs.

The role of investment in developed countries is playing a more important role in driving the transformation of OFDI in the Chinese manufacturing industry. This might be explained by the fact that developed countries have a clear advantage in high-end manufacturing and technological innovation, providing Chinese firms with access to cutting-edge technology and management expertise. In addition, developed countries dominate the high-end links of the global industrial chain and possess a strong brand influence and market control. Chinese manufacturing enterprises’ investing in these countries can help them deeply integrate into the global industrial chain and enhance their international competitiveness. In addition, developed countries have a huge consumer market and mature consumer groups, with a strong demand for high-quality, high-tech products, which is conducive to the transformation of OFDI by manufacturing enterprises. Hypothesis H3 is assumed to hold true.

5.4. Analysis of Nonlinear Threshold Effects

In order to further examine the nonlinear relationship between GVC restructuring and the transformation of China’s manufacturing OFDI, a panel threshold model is used for the test, and trade openness is selected as the threshold variable. Before conducting the threshold effect test, the corresponding F statistic is obtained by repeated sampling through the bootstrap method, and its corresponding p-value is used to verify the existence of the threshold variable first, and the specific results are shown in

Table 6. The analysis results concerning the existence of thresholds.

Threshold Variable	Quantity of Thresholds	F	P	BS Frequency	Threshold Value	1%	5%	10%
GVC-dbc	Single threshold	112.77 ***	0.0000	500	87.4200	80.2908	63.8166	55.6444
	double threshold	51.72 **	0.0200	500	58.0000	40.1326	46.0869	57.2555
GVC-rpi	Single threshold	118.26 ***	0.0040	500	87.4200	73.1162	80.3698	103.8764
	double threshold	56.96	0.0580	500	56.8175	50.0444	59.0461	40.0431

Note: **, and *** denote significance at the 5% and 1% levels, respectively.

. When the independent variable is the relative position of the GVC, the double-threshold p-value of trade openness is 0.0200, and the threshold value is 58.0000, which proves that the relative position of the GVC on the transformation of China’s OFDI in the manufacturing industry is affected by the double threshold of trade openness; when the independent variable is the degree of bilateral cooperation in the GVC, the single-threshold p-value of trade openness is 0.0040, and the threshold value is 87.4200, proving that the degree of bilateral cooperation of GVCs has a single-threshold effect on the transformation of China’s manufacturing OFDI by trade openness.

Further, the regression of the model with the specific number of thresholds set is conducted, and the specific results are shown in

Table 7. Results of threshold regression analysis.

Variables	GVC-dbc	GVC-rpi
Open (Open < 58.0000)	0.750 ***
	(0.174)
Open (58.0000 < Open ≤ 87.4200)	1.077 ***
	(0.163)
Open (Open > 87.4200)	1.491 ***
	(0.160)
Open (Open < 87.4200)		0.124
		(0.134)
Open (Open ≥ 87.4200)		0.724 ***
		(0.138)
Control variables	Control	Control
N	615	615
R2	0.473	0.366

Note: *** denote significance at the 1% levels.

. The relative position of GVCs and the degree of bilateral cooperation of GVCs have a significant nonlinear relationship with the OFDI transformation of China’s manufacturing industry. From the nonlinear relationship between the relative position of GVCs and the OFDI transformation of China’s manufacturing industry, under the influence of dual thresholds, the promotion effect of the relative position of GVCs on the OFDI transformation of China’s manufacturing industry shows a process from weak to strong.

From the nonlinear characteristics of the degree of bilateral cooperation of GVCs on the OFDI transformation of China’s manufacturing industry, the coefficient of the degree of bilateral cooperation of GVCs on the OFDI transformation of China’s manufacturing industry is positive in all intervals and passes the test of significance of 1%, which indicates that there is a threshold effect in the influence of the degree of bilateral cooperation of GVCs on the OFDI transformation of China’s manufacturing industry. It can be found from observing the data that the promotion effect of the degree of bilateral cooperation of GVCs on the transformation of China’s manufacturing OFDI shows a nonlinear change process from weak to strong, which means that after the degree of trade openness exceeds a certain interval, the promotion effect of the degree of bilateral cooperation of GVCs on the transformation of China’s manufacturing OFDI is gradually enhanced.

5.5. Robustness Testing

5.5.1. Shrinkage Treatment

To mitigate the impact of outliers on the regression results, the research data undergoes Winsorization by trimming the top and bottom 1%. Values below the 1st percentile and above the 99th percentile are removed, and the model is refitted for analysis. The results are shown in

Table 8. Shrinkage treatment results.

Variables	LnOFDI	LnOFDI
	(1)	(2)
GVC-dbc	0.344 **
	(0.165)
GVC-rpi		0.399 ***
		(0.131)
Lngfc	0.313 **	0.272 *
	(0.159)	(0.156)
Lnpopden	−0.003 ***	−0.004 ***
	(0.001)	(0.001)
Lnmcs	−0.985 ***	−1.046 ***
	(0.271)	(0.258)
Gov	−5.167 **	−5.088 **
	(2.153)	(2.136)
Int	−0.016 ***	−0.016 ***
	(0.004)	(0.005)
Constant	11.348 ***	12.649 ***
	(3.884)	(3.671)
N	615	615
R2	0.932	0.933
Area fixed effect	YES	YES
Year fixed effect	YES	YES

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

. The data shows that the former conclusions remain reliable after data processing.

5.5.2. Shortening the Sample Period

We take into account the potential impact of the 2008 financial crisis and the 2020 COVID-19 pandemic on investment behavior. The sample period is shortened by excluding 2008 and 2020, and the model is refitted for analysis. See

Table 9. Testing for sample period truncation.

Variables	LnOFDI	LnOFDI
	(1)	(2)
GVC-dbc	0.561 ***
	(0.194)
GVC-rpi		0.563 ***
		(0.142)
Lngfc	0.176	0.093
	(0.178)	(0.182)
Lnpopden	−0.002 ***	−0.002 ***
	(0.001)	(0.001)
Lnmcs	−0.627 **	−0.726 **
	(0.302)	(0.286)
Gov	−4.256 *	−3.951 *
	(2.208)	(2.176)
Int	−0.017 ***	−0.017 ***
	(0.005)	(0.005)
Constant	12.473 ***	15.208 ***
	(4.324)	(4.379)
N	533	533
R2	0.932	0.933
Area fixed effect	YES	YES
Year fixed effect	YES	YES

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

for the results. The data indicates that the previous conclusions remain robust after excluding these outlier years.

5.5.3. Endogeneity Test

In the baseline regression, we control for country-specific fixed effects and time effects using a two-way fixed effects model. It helps to address the endogeneity problem posed by omitted variables but may not fully account for the potential endogeneity posed by two-way causality. To further mitigate potential endogeneity between the key variables and the dependent variable, we re-estimate the regression with a one-period lag on the explanatory variables. As shown in

Table 10. Lagged regression of explanatory variables.

Variables	LnOFDI	LnOFDI
	(1)	(2)
L.GVC-dbc	0.302 ***
	(0.116)
L.GVC-rpi		0.320 ***
		(0.113)
Constant	14.790 ***	16.267 ***
	(4.039)	(3.657)
N	574	574
R2	0.941	0.941
Control variable	YES	YES
Area fixed effect	YES	YES
Year fixed effect	YES	YES

Note: *** denote significance at the 1% levels.

, when using a lagged period for the key explanatory variables, the coefficients remain positive and significant at the 1% level, indicating the robustness of our findings.

6. Conclusions, Implications, and Research Limitations

6.1. Conclusions

(1)

GVC restructuring has a positive impact on the OFDI transformation of manufacturing firms. Furthermore, this impact process is not the result of a single factor. Based on robustness tests, the research findings are reliable. Therefore, in promoting the transformation and upgrading of OFDI, countries should actively integrate into and take advantage of the strategic opportunities of GVC restructuring. They should also use it as a key driving force for transformation and upgrading. Meanwhile, the country should adopt a more comprehensive and in-depth analytical approach to ensure the scientific and sustainable nature of the transition path.

(2)

In terms of mediating effects, GVC restructuring can indirectly promote the OFDI transformation of manufacturing firms through the mediating variables of factor endowment and market scale. In other words, factor endowments and market scale play a crucial role in the process of GVC restructuring that affects the transformation of manufacturing OFDI. Especially in the context of improving factor endowments and expanding market scale, the transformation of OFDI by manufacturing firms will be accelerated.

(3)

In countries with different levels of economic development, there are differences in the relationship between value chain restructuring and the OFDI transformation of manufacturing enterprises. Among them, the reorganization of GVCs in developed countries has the most significant impact on the OFDI transformation of Chinese manufacturing enterprises. Therefore, it is crucial to emphasize increased investment in developed countries when formulating OFDI strategic planning for manufacturing firms.

(4)

The positive driving effect of GVC restructuring on the transformation of China’s manufacturing OFDI is affected by the degree of trade openness and presents a nonlinear characteristic. Overall, the promotion effect of GVC restructuring on the transformation of China’s manufacturing OFDI shows a nonlinear process of change from weak to strong.

6.2. Implications

In order to visualize the findings and their practical application value, we constructed Figure 2, which clearly illustrates how the findings can be translated into practical recommendations.

(1)

The government should set up a mechanism to assess and assist the GVC-OFDI Transformation Index. Relying on the mechanism of interaction between value chain restructuring and OFDI transformation, the government can construct a comprehensive assessment framework that includes technical complexity, resource integration capacity, and institutional compatibility, among other dimensions. By integrating data from multiple sources of industries and enterprises, the government can generate dynamic rankings of provincial transformation indices on a timely basis and at the same time set up an open and transparent release platform.

For regions that consistently rank low, the government can implement a “customized assistance mechanism”, set up a team of sustainability experts, and formulate targeted OFDI transformation strategies based on the industrial characteristics of each region. In addition, with regard to the traditional manufacturing industry, the government needs to focus on building a platform for technological cooperation and promoting the breaking of the “low-end lock-in”.

(2)

The government should strengthen the supply system of innovation factors. The government can set up a special support fund as a way to promote the optimization of the incentive model of enterprise R&D innovation. Meanwhile, the government should focus on introducing high-end talents in the field of environmental technology and training programs in sustainable technology and strengthen cooperation with internationally renowned green technology research institutions. This is conducive to the promotion of a “talent, technology, industry” synergistic development of a virtuous cycle.

(3)

Relevant enterprises need to strengthen in-depth market research in target markets. By understanding market scale changes and consumer trends, they provide a precise positioning and market orientation for manufacturing enterprises’ foreign investments. The government should encourage enterprises to actively participate in international green market competition and expand their share of the global market for sustainable products. Also, the government should promote the development of cross-border e-commerce to achieve effective market expansion and provide new opportunities for the transformation of manufacturing OFDI.

(4)

Relevant enterprises should focus on high-end markets. Manufacturing enterprises need to keep pace with the restructuring of GVCs in developed countries and focus on investing in technology-intensive industries. In addition, relevant enterprises can set up overseas R&D centers in technologically developed countries such as the US and Germany and also focus on green technology mergers and acquisitions and sustainable cooperation in research and development in key areas such as clean energy and the circular economy. At the same time, relevant enterprises can participate in the host country’s innovation network through equity participation in local leading enterprises and the joint establishment of cross-border laboratories.

(5)

The government should optimize the trade openness policy. It should adjust the degree of trade openness in phases, focusing on lowering institutional barriers at the initial stage and strengthening high-end openness such as digital trade and service trade in the middle and late stages, so as to match the accelerating effect of GVCs’ reconstruction and promote the transformation of China’s manufacturing OFDI. On the other hand, the government should dynamically adjust the OFDI support policy: at the stage of low openness, it should provide information consultation and risk protection; at the stage of higher openness, it should strengthen financial and tax incentives and promote the extension of OFDI to high value-added segments.

6.3. Limitations and Future Research Directions

Due to constraints on this paper’s length and the authors’ time and resources, this paper, while achieving some results, still has several limitations, pointing towards directions for future research: (1) The research sample in this paper needs further expansion. The analysis only covers the overall situation of GVC restructuring in 41 countries, potentially overlooking others. Future research will refine the sample to go deeper into the micro level of firms and analyze the OFDI transformation of manufacturing firms of different scales and industries in the context of GVC restructuring. (2) The research theme requires strengthening. This paper solely focuses on the manufacturing industry. However, the impact of GVC restructuring is cross-sectoral. Different industries have different positions, roles, and challenges in GVCs. Therefore, future research can broaden to include more sectors such as services and agriculture. (3) Insufficient industry segmentation. Although this study focuses on the manufacturing industry as a whole, it fails to adequately differentiate between the differentiated characteristics of labor-intensive, capital-intensive, and technology-intensive manufacturing industries and their different performance in the reconstruction of global value chains. Future research will focus on the segmentation of labor-intensive, capital-intensive, and technology-intensive manufacturing industries and analyze their differentiated performance in the restructuring of global value chains. (4) There is room for optimizing the choice of proxy variables for external factors. Future research will construct comprehensive external shock indicators to more accurately measure the impact of multiple factors such as technological innovation and geopolitics on the transformation of manufacturing OFDI.