The Impact of Global Value Chain Reconstruction on the Innovative Latitude High-Quality Development of Reverse OFDI in China—From the Perspective of Jiangsu Province

Abstract

Based on the current unpredictable patterns of globalization and the impact of the COVID-19 pandemic, global value chains (GVCs) are undergoing restructuring. The resolution of the 20th Communist Party of China National Congress emphasizes high-quality development and the “going out” strategy. It highlights the crucial role of global value chain reconstruction in driving the high-quality development of outward foreign direct investment (OFDI). Innovation is crucial in reaching this high-quality development objective. This study uses Jiangsu Province in China as a case study to estimate the relationship between the innovative direction of the high-quality development indicator score system for Jiangsu Province’s reverse OFDI and global value chain restructuring from 2007 to 2021. The findings indicate that global value chain restructuring has a suppressive effect on the innovative direction of high-quality development in Jiangsu’s reverse OFDI. Additionally, further heterogeneity analysis reveals that urbanization levels mitigate the negative impact of global value chain restructuring on the innovative direction of high-quality development in Jiangsu’s reverse OFDI.

1. Introduction

The spring breeze of reform and opening up, along with the ongoing implementation of the “going out” strategy, has steadily boosted the country’s OFDI. According to data from the Ministry of Commerce and the Foreign Exchange Bureau, China’s OFDI in various sectors amounted to CNY 522.18 billion during the first half of 2023. However, the proportion of investments directed towards offshore financial centers in China’s current outward foreign investments is excessively high. This situation has led to an overestimation of the actual level of outward foreign investments. This kind of overestimation hinders the achievement of high-quality development in China’s outward foreign direct investments. The going out strategy remains a pivotal approach for Chinese firms to engage in high-quality international investments. At present, China is vigorously pursuing the advancement of high-quality OFDI.

Recently, trade protectionism and geopolitical tensions have increasingly disrupted the global economic landscape. The shift towards regionalization within global value chains has become more pronounced, intensifying the phenomenon of de-globalization. Historically, developed nations have controlled the upper tiers of value chains. This dominance has created an imbalance, hindering economic cooperation and sustainable growth among countries [1]. The rise of developing nations has reconfigured these production chains, while advancements in technology have streamlined global production and supply networks. These technological developments and the shift away from initial production models have led to the reorganization of value chains. Developing nations, often reliant on low-cost labor, have found themselves entrenched at the lower end of these chains. It is essential for these countries to leverage ongoing global value chain reorganization to improve their positions. They are actively striving to ascend within this restructuring trend. As the leading developing economy, China has taken on significant roles in this process. Navigating the high-quality advancement of OFDI in China amidst the current global value chain reorganization presents a formidable challenge.

Currently, academics hold varying perspectives on whether OFDI can promote the development of global value chains. Numerous studies suggest that OFDI can potentially facilitate the evolution of global value chains [2,3]. Conversely, some academics argue that OFDI may inhibit the status of global value chains. For example, Ren (2023) argued that OFDI negatively affects the position of the manufacturing processing trade within the value chain [4]. Today, China is striving toward the goal of high-quality advancement. The high-quality advancement of OFDI is also a crucial approach for China to become a prominent country in outward foreign investment [5]. Innovation is a key element that China urgently seeks in the high-quality development stage of OFDI [6]. Numerous researchers have suggested that the reverse technology spillover of OFDI serves as a mechanism for acquiring technological innovation [7,8]. Building on reverse spillovers in OFDI activities can enhance the country’s value chain position [9,10]. Most studies have comprehensively explored the consequences of global value chain positions from the perspective of OFDI. There is relatively less research conducted in the opposite direction. Additionally, a detailed plan for the high-quality development model of various facets of OFDI has yet to be established.

Jiangsu Province stands out as one of the leading regions in economic development within China. According to the 2022 report from the Jiangsu Provincial Bureau of Statistics, the province’s gross regional product reached CNY 12,287.56 billion. This figure marks an ascent to a new threshold of CNY 12 trillion, showing a 2.8% increase from the previous year. Unlike other less developed provinces, Jiangsu Province has certain advantages in resilience during economic impacts. Consequently, this study uses Jiangsu Province as a case study. In this article, sub-indexes are derived from the number of overseas subsidiaries established by the high-tech industry in Jiangsu Province and the amount of outward foreign research and development (R&D) capital stock acquired through OFDI channels in 16 developed nations between 2007 and 2021. The score for the innovation dimension of the high-quality development index is determined using the entropy weight method. Therefore, based on the aforementioned background and the literature, this study focuses on discussing how the restructuring of global value chains has impacted the high-quality development of innovative dimensions in reverse OFDI in Jiangsu Province. How can we mitigate this negative impact? What is the mechanism behind it? This helps us clarify the specific pathways through which the restructuring of global value chains influences the high-quality development of innovative dimensions in reverse OFDI in Jiangsu Province. Building on the analysis of these pathways, we can outline the direction for Jiangsu Province’s future OFDI policies. This has important theoretical and practical significance for China to become a strong country in outward foreign investment and realize innovation-driven high-quality development.

The research contribution of this paper primarily includes the following aspects. Firstly, it undertakes a reverse study, inspecting the influence of high-quality OFDI advancement through the lens of GVC reconstruction. The impact of OFDI on GVC results has been extensively discussed. In addition, we approach the topic from the specific viewpoint of OFDI’s high-quality development. OFDI’s high-quality development includes many aspects. General Secretary Xi Jinping identified five key areas for high-quality progress within the new development paradigm [11]. This research analyzes the effect of GVC reconstruction on the high-quality development of reverse OFDI in Jiangsu Province from the visual angle of innovation. We will provide policy recommendations for Jiangsu Province’s future outbound investment strategies.

2. Literature Review

Current global trends reflect increasing anti-globalization sentiments within the international arena. The outlook for China’s function in the global value chain is not promising. Scholars present differing views on the future trajectory of the global value chain, debating whether it will continue to expand or contract in the coming years. In today’s era marked by volatile restructuring trends in global value chains, high-tech industries and reverse technology spillover from OFDI are significant pathways for acquiring innovation.

2.1. OFDI Reverse Technology Spillovers

Reverse technology spillovers from OFDI are crucial for accessing innovative elements vital to the high-quality progression of reverse OFDI. Numerous scholars have focused on this issue, with most emphasizing the importance of fostering innovation in environmental protection to encourage the spread of green technology spillovers [12]. Fahad et al. (2022) identified that environmental stipulations and industry-specific policies contribute to reverse spillover effects domestically [13]. Liu (2021) believed that stringent environmental stipulations and the improvement of OFDI’s knowledge transfer ability can bolster reverse green spillover [14]. Dai (2021) argued that reducing the intensity of market incentive regulations might hinder the spillover effects of green technologies. They propose that only with robust market incentives can the effective dissemination of green technologies by companies be promoted [15]. Zhou and Jiang (2019) asserted that OFDI facilitates green spillovers to China, yet the impact of OFDI initiatives on advancing local green economic transformation remains limited. Therefore, government policies are needed to reinforce reverse OFDI green spillovers and expedite local green economic transitions [16]. Bai et al. (2020) discovered that OFDI from developed nations can trigger reverse spillovers, promoting environmental innovation within the parent firms. This impact on environmental innovation can be further enhanced by the host countries’ knowledge base and multinational strategies [17]. Zhang (2022) argued that OFDI reverse technology spillovers can visibly reduce energy intensity [18]. Contemporary scholars consistently recommend strengthening our country’s technological capabilities through reverse technology spillovers from developed nations. At the same time, they emphasize adopting environmentally sustainable practices to ensure a development path that harmonizes technological advancement with environmental protection. The reverse technology spillover also presents regional characteristics. Pan et al. (2020) found that there is some zonal heterogeneity in the consequences of OFDI reverse spillover on carbon productivity. The impact of this factor is more visible in advanced regions located in the eastern areas, leading to a notable enhancement in carbon productivity. Conversely, its influence is found to be insignificant in less-developed regions such as the central and western areas [19]. The above studies all point out the influence of different factors on reverse technology spillovers of OFDI. Researchers focus on enhancing reverse technology spillovers of OFDI to gain more technological innovation. However, enterprises conducting OFDI may also squeeze R&D costs. The reduction in R&D costs adversely affects innovation development. This presents certain limitations to the conclusions of the above studies. Researchers lack a detailed examination of this perspective. Future research should conduct an in-depth analysis of this perspective to further improve research in this field.

2.2. Global Value Chain Reconstruction

Global value chain restructuring can change value distribution within the original industry chain. This is achieved by implementing technological revolutions and adjusting resource allocation, thereby upgrading the global value chain. The restructured value chain’s advantages include geographical segmentation. This allows for positioning various production activities in the most efficient locations and with the most effective participants [20]. However, since the 2008 financial crisis, the process of globalization has slowed down. The COVID-19 pandemic has once again impacted global trade. Geopolitical conflicts and climate change have driven the reconfiguration of global value chains. Economies are seeking self-centered regional value chains [21]. Therefore, China needs to elevate its position in the value chain by embedding itself to form a high-tech value chain network centered around itself. Scholars have highlighted the importance of the current reconfiguration of value chains and proposed targeted measures. Dai et al. (2022) noted that digital technology represents the intelligent demand for value chain restructuring in the new information technology industrial revolution. This may provide an opportunity for China to participate in this restructuring [22]. However, with the improvement of digitalization, information security cannot be fully guaranteed. This study only raises the issue of information security without providing specific solutions. This is a problem that scholars should consider in future research. Qu et al. (2020) found that most manufacturing enterprises in China remain embedded in the downstream segments of global value chains. China should adopt differentiated policies to reach high-quality and rapid economic growth [23]. This study focuses solely on manufacturing enterprises. There is also a lack of specific exploration into enhancing value chain integration in other industries such as agriculture and construction. Future research can broaden the perspective of other sub-sectors, contributing to the development of other industries. High-tech industries are the cornerstone of innovation. Song (2021) proposed that developed countries concentrate on global value chain governance. Developing countries work on global value chain upgrading. High-tech industries with significant competitive edges can lead global value chain development. In contrast, those industries that have not yet achieved leading competitiveness can develop national value chains to establish strong regional advantages [24]. Wang et al. (2021) found a link between the degree of participation in global value chains and technological advancement in emerging nations [25]. In many developing areas, limited participation in global value chains has obstructed the acquisition of technological capabilities. This has reduced them to simply being “manufacturing hubs” for more advanced economies. However, this literature only selects the BRIC countries and Mexico as examples to represent the value chain situation of developing countries. This presents a limitation in extending the study’s strategies to all developing economies. Ambos (2021) advocated for a holistic understanding of innovation actors and their actions within a unified framework. Evaluating innovation dynamics necessitates a GVC-centric approach rather than a narrow focus on individual participants’ activities [26]. With rapid advancements in artificial intelligence, certain functions at the lower rungs of the value chain face displacement by AI-driven automation. It poses a significant threat to their development trajectory.

In summary, scholars have pointed out that the ascent of countries in the value chain and the construction of their own core value chains are important processes in the restructuring of value chains. The literature emphasizes that technological innovation is a key requirement for this global value chain restructuring. However, they have not yet proposed solutions to a series of issues such as cybersecurity and data privacy during the current wave of digital technology. At present, scholars focus more on the development of the manufacturing industry, while related research in other industries such as agriculture and construction is relatively scarce. In future research, we also need to focus on related studies in other sub-sectors. In addition, as our country seeks green development, how to pursue the green restructuring of value chains is also one of the key points. Future research should incorporate this perspective to enhance the quality of value chain restructuring [21].

2.3. OFDI and Global Value Chain Upgrading

Contemporary research extensively examines the impact of OFDI on value chain upgrading. The majority of contemporary scholars concur that OFDI promotes GVC upgrading. OFDI in emerging economies has a larger catalytic effect on GVC upgrading than in advanced economies [27]. Li (2021) found that through empirical analyses, OFDI has a facilitating effect on the enhancement of the GVC by acquiring technological upgrading through reverse technology spillovers. The beneficial impact of emerging economies is greater than that of developed countries [3]. Li (2023) argued that developing countries can achieve technological progress through GVC production networks and capture technological spillovers [28]. Song (2022) found that OFDI has contributed to the advancement of the quality of the domestic nation’s global value chain. However, there is still room for improvement when compared to more developed nations. The efficiency of core industries, such as manufacturing, requires further enhancement to align with the high-quality evolution of trade [29]. Dai (2021) discovered OFDI’s role in forging value chain linkages with host countries, thereby fortifying the value chain’s stability and resilience [30]. Since the overall value chain position of emerging economies is lower than that of developed economies, emerging economies have a more urgent need to climb the value chain. Therefore, researchers focus more on detailed analyses of the role of OFDI in upgrading the value chains of developing countries. They mostly do not include an in-depth analysis of developed countries. This study, which takes developed countries as the research object, can compensate for the above research gaps.

2.4. High-Quality Development of OFDI

To achieve the important goal of high-quality economic growth and ensure the lasting success of the “going out” initiative, China should focus on finding better opportunities for OFDI. China’s outward investment amount is relatively high. However, our share in the world is not large. The trade tensions between China and the United States have affected the degree of obtaining high-quality OFDI. Various scholars have proposed prospective strategies aimed at fostering the high-quality advancement of OFDI. Based on the background of China’s high-quality advancement, Wang (2023) pointed out that OFDI is a vital countermeasure for comprehensive high-quality advancement. Innovation ability is the most important method of reflecting economic strength. A robust and all-encompassing encouragement mechanism should be established. Additionally, the capacity to absorb OFDI reverse technological spillovers should be enhanced by leveraging high-end human capital. On these bases, we can accelerate the realization of the high-quality development goal [5]. Li (2021) emphasized that safeguarding intellectual property rights (IPR) greatly impacts the investment strategies of Chinese companies engaged in OFDI. Establishing a robust IPR protection system tailored to China’s OFDI will boost both the quality and efficiency of its development [31]. This literature discusses measures for the high-quality development of OFDI from the perspective of national intellectual property protection. However, the levels of development across provinces in our country differ significantly. The current situation of OFDI varies across provinces. The exploration of high-quality OFDI development in each province is a gap in the research. Therefore, future research should examine the actual conditions of various provinces to explore high-quality OFDI development in more detail. From the direction of energy policy, Zhang (2021) put forward that both traditional and emerging energy policies can support enterprises’ OFDI activities. Achieving balanced progress in both conventional and new energy sectors is essential for highly efficient OFDI outcomes [32]. Liu (2024) highlighted that OFDI activities can help reduce environmental degradation, aligning with current priorities on fostering a high-quality development environment. Enhancing sustainable investment practices can further alleviate environmental pressures in highly industrialized cities in the future [33]. Although China’s total OFDI has been numerically increasing, it has not yet secured significant advantages in the high-tech sector. Most OFDI is concentrated in financial centers instead of being adequately invested in the real economy. This has resulted in a significant gap between China and more advanced economies. The above scholars have provided policy recommendations for the high-quality development of OFDI from the perspectives of intellectual property, energy policies, and other angles. However, these studies all take a national perspective and lack focused research on specific provinces. Our study takes Jiangsu Province as the research object and provides more specific high-quality development recommendations based on the current situation of OFDI in Jiangsu Province. Our research can compensate for the limitations of the aforementioned studies.

In summary, scholars have raised their views on the future development trend of GVC restructuring and the prospect of high-quality progress in OFDI. In addition, most of the literature is still oriented towards GVC upgrading. They try their best to probe the pathway mechanism of OFDI on GVC upgrading. There are few studies that consider how global value chain reconstruction affects the development of OFDI. There is limited consideration of how OFDI can achieve high-quality advancement. Additionally, this paper only contemplates the innovation dimension of the high-quality advancement of Jiangsu Province’s reverse OFDI for deep analysis. Jiangsu Province is chosen as the research target to seek the impact of today’s trend of global value chain restructuring on the innovative direction of the high-quality development indicator score system of Jiangsu Province’s reverse OFDI.

3. Theoretical Analysis and Research Hypothesis

In 1985, Porter pointed out that a company’s overall production and business activities could be divided into two segments: primary activities and support activities. Basic activities include production, marketing, transportation, and after-sales services while support activities cover raw material supply, technology, human resources, finance, and more. Each activity segment contributes to creating and delivering value. Through reciprocal interconnection and mutual impact, these components collectively form the firm’s value chain. Developed countries benefit from primitive capital accumulation and advanced production technology originating from the Industrial Revolution. They manage resources, produce on a global scale, and trade the inexpensive labor of developing countries for their competitive advantages, creating a vertically integrated production system [34]. Kogut subsequently introduced the value-added theory, which views the value chain as an effective combination of elements such as technology, labor, and raw materials. In the global strategic planning of enterprises, this theory involves configuring various aspects of the value chain. It highlights the distinct advantages enterprises have in specific segments of the value chain, showcasing the vertical division of labor and the importance of co-location within the global production network. This concept forms a fundamental basis for the global value chain theory. In 1999, Gereffi introduced the notion of global commodity chains. He argued that in the modern global economy, production activities are increasingly interconnected as multinational corporations manage value-chain activities. As key players in international production networks, these transnational corporations integrate numerous production-related firms worldwide into the global production chains of goods. Gereffi first introduced the concept of global value chains, building on the idea of global commodity chains. He later developed the theory of labor division within these value chains. This theory accounts for the current phenomenon of transnational corporations dominating production activities across borders. The division in global value chains is distinguished by fragmented production processes, the relocation of production activities, and vertical specialization. This transition has altered comparative advantage among countries from traditional product-based comparisons to those focused on specific stages of production [35,36].

With the advancement of the Fourth Industrial Revolution, the primary goal of global development has shifted towards greater intelligence and digitalization. According to the comparative advantage theory, advanced economies have amassed superior innovative technologies. Consequently, they are primarily involved in high-end industries, creating core components with substantial added value. In contrast, lower-end production processes are usually transferred to developing nations, where labor costs are lower. At the same time, the Smile Curve Theory indicates that achieving high-quality economic growth involves focusing on product development and brand services. According to this theory, these areas at both ends of the curve have higher profit margins. Our analysis indicates that the middle part of the chain has the lowest added value. From this theory, it becomes evident that technology and intellectual property are pivotal in global value chains, while marketing and promotion are crucial in regional value chains. Therefore, firms must continuously innovate to secure higher positions in the value chain.

Furthermore, the monopoly advantage theory posits that multinational corporations engage in OFDI based on specific advantages. Developed nations have a considerable advantage over emerging economies in the number of innovation patents they hold. Consequently, developing nations often conduct outward FDI to developed countries, aiming to gain reverse technology spillover. This process helps to progressively close the technological gap with advanced economies. Nonetheless, there has been a rise in trade conflicts between China and the United States. Advanced economies have imposed technological restrictions on China to maintain their leading positions in the value chain. For multinational corporations in Jiangsu Province, the path is fraught with challenges and barriers. The benefits of reverse technology spillovers from OFDI will gradually decline. Additionally, it remains uncertain whether the acquired technology spillovers can be converted into new, adaptable technologies due to limitations in absorptive capacity. Businesses recognize that relying on advanced technology and skilled labor from developed countries hampers the achievement of high-quality development goals. This is especially true when they neglect independent innovation. Recently, increasing regional conflicts have disrupted the global supply chain, significantly impacting OFDI activities worldwide. The United States has imposed hostile tariffs on China, while European countries have developed regional value chains within their alliances. Both actions have negatively impacted China’s integration into the global value chain. Previously, developed countries invested heavily in manufacturing facilities in China because of its lower labor costs. This led to the hollowing out of their own industries and an increase in unemployment rates. Consequently, these countries have initiated industrial reshoring strategies. They are bringing back critical industries while moving less essential ones to countries with lower labor costs, such as Vietnam, as China’s labor cost advantage has gradually diminished. Therefore, through the above analysis, the hypothesis is formulated as follows:

H 1.

Global value chain reconstruction will have a negative influence on the innovation dimension of the high-quality development of Jiangsu’s reverse OFDI.

4. Model Construction and Variable Descriptions

4.1. Model Construction

The data in this study are panel data. Panel data consist of data from different subjects over different time periods, encompassing both time and spatial dimensions. By controlling for time and regional effects, this study can eliminate the interference of these two factors. This approach more accurately identifies the impact of explanatory variables on the dependent variable. Therefore, this article considers selecting a two-way fixed-effects model for estimation. Additionally, after conducting the Hausman test, it is found that the p-value is 0.0000. This also indicates that the fixed-effects model should be chosen for this study. Finally, we choose the two-way fixed-effects model for this study.

Based on the fact that this paper is constructed with panel data, the following treatments are applied to the data: (1) A 1% shrinkage is applied to the dependent, independent, and control variables to relieve the influence of outliers on the model analysis. (2) Due to the large disparity in data magnitude, some of the control variables are treated logarithmically to reduce the interference of multicollinearity in the model. (3) Employing a two-way fixed-effects model, this research incorporates year-level fixed effects denoted as μ_t and country-level fixed effects denoted as σ_i in the regression analysis.

In order to probe the above hypotheses and seek the influence of GVC reconstruction on the innovation direction of the high-quality development of Jiangsu Province’s reverse OFDI, a benchmark model is first established:

OFDII_it = α₀ + β₁GVC_it + γ_iControls_it + δ_i + μ_t + ε_it

(1)

OFDII_it is the indicator score of the innovative direction of the high-quality development of Jiangsu’s reverse OFDI for host country i in year t. GVC_it is the Global Value Chain Position for host country i in year t. α₀ is the constant term. β₁ is the estimated coefficient of the GVC status. Controls_it is the control variable. The control variables cover natural resource endowment (Res), the level of the labor force (Lab), foreign direct investment (Fdi), infrastructure level (Inf), the level of manufacturing development (Ind), and institutional distance (Ins). γ_i is the estimated coefficient of the control variable. σ_i is the country-level fixed effect. μ_t is the year-level fixed effect and ε_it is the residual term.

4.1.1. OFDI Innovation Dimension High-Quality Development Indicator Score

The number of overseas subsidiaries represents the scale of an enterprise’s OFDI [37]. According to Xiong (2022), high-tech companies indicate the level of innovation development within OFDI [38]. This measure is essential for assessing the innovation aspect of OFDI’s high-quality growth [24]. Additionally, reverse technology spillover from OFDI is a key channel for obtaining innovative development [39]. Through OFDI activities, enterprises acquire advanced technologies, facilitating reverse spillovers. This process reflects the high-quality development from OFDI’s innovation perspective [40]. The assessment of the high-quality advancement level in the reverse OFDI innovation dimension involves examining the number of overseas subsidiaries of high-tech firms and reverse technology spillovers. These indicators are then amalgamated into a comprehensive score for comparative analysis using the entropy method.

Firstly, this paper draws on Wang Fusheng’s (2022) [41] measure of high-tech industries and refers to the Guidelines for Industry Classification of Listed Companies (2012 Revision) to select the relevant samples contained in the manufacturing industry of category C (C26–C28, C35–C40) and the message transfer, software, and message technology service industry (I65). We also draw on Tang’s (2020) [42] paper. The study utilized A-share listed companies in Jiangsu Province from 2010 to 2021. Overseas subsidiaries of high-tech enterprises serve as an indicator of the innovative development of OFDI. Data for the study were sourced from the CSMAR. To guarantee the robustness of data, these data are processed based on the following steps: (1) Selection of non-ST enterprises; (2) Exclusion of samples whose registered office is located in “tax havens”, as well as in Hong Kong, Macao, and Taiwan.

Secondly, for the host country’s R&D capital stock of OFDI channel spillovers, this paper draws on Dong (2021) [43] and adopts the P-L (2001) [7] method to measure it, which is expressed by the formula:

$${{\mathrm{S}}^{\mathrm{f}}}_{\mathrm{it}}={\displaystyle \frac{{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i}\mathrm{t}}}{{\mathrm{K}}_{\mathrm{i}\mathrm{t}}}}{{\mathrm{S}}^{\mathrm{d}}}_{\mathrm{it}}$$

(2)

OFDI_it denotes the stock of Jiangsu province’s outward foreign direct investment in country i in period t. K_it represents the fixed capital formation amount in country i in period t. S^d_it denotes the domestic R&D capital stock in country i in period t. Firstly, the base period R&D capital stock of country i in 2007 is computed with the formula:

$${\mathrm{S}}_{\mathrm{i}2007}={\displaystyle \frac{{RD}_{i2007}}{(g+\delta )}}$$

(3)

RD_i2007 is country i’s R&D expenditure in 2007. g is the average yearly growth rate of R&D expenditure in country i. δ is the depreciation rate of R&D capital, which is taken here as 5%, chosen by most scholars. Next, the perpetual inventory method is used to compute the R&D capital stock for the remaining years in country i, using the formula:

S^d_it= (1 − δ)S^d_it-1 + RD_it

(4)

RD_it is the newly increased R&D expenses in country i in year t (deflated to constant 2007 prices based on the consumer price index).

Finally, the R&D capital stock of the host country obtained by Jiangsu Province through OFDI is calculated by the formula:

$${{\mathrm{S}}^{\mathrm{fo}}}_{\mathrm{it}}={\displaystyle \frac{{OFDI}_{it}}{{\sum }_i^1{OFDI}_{it}}}{{\mathrm{S}}^f}_{\mathrm{it}}$$

(5)

OFDI_it is the stock of Jiangsu province’s OFDI i in period t. Here, the ratio of Jiangsu Province’s OFDI stock to China’s total OFDI stock is used to sum the R&D capital stock of the host country obtained by Jiangsu Province through OFDI.

4.1.2. Entropy Weight Method

The entropy weight method is used to compute the weights of the indicators and establish the composite index. This paper draws on Zhang (2020) [44]. Initially, the polar difference approach is used to normalize the secondary indicators. All secondary indicators are positive and undergo positive standardization based on the specified formula.

$${\mathrm{X}}_{\mathrm{ij}}^{\prime }={\displaystyle \frac{X_{ij}-X_{\min }}{X_{\max }-X_{\min }}}$$

(6)

where i expresses the ith country and j denotes the jth indicator. X_ij denotes the value of the secondary index of the jth indicator for the ith country in the innovation dimension high-quality development indicator score system. X′_ij denotes the value of the jth indicator in region i after normalization. X_max represents the maximum value of the jth index. X_min represents the minimum value of the jth index.

Secondly, based on the standardized indicator values, the share p_ij of the ith country in the jth indicator is calculated, and p_ij is approximated as 0.0001 when p_ij is equal to zero.

$${\mathrm{p}}_{\mathrm{ij}}={\displaystyle \frac{X_{ij}^{\prime }}{{\displaystyle \sum _{i=1}^n{X}_{ij}^{\prime }}}}, i=1, \cdots , n; j=1, \cdots , m$$

(7)

To compute the entropy value e_j for the jth indicator.

$${\mathrm{e}}_{\mathrm{j}}=-k{\displaystyle \sum _{i=1}^n{p}_{ij}}\ln (p_{ij})$$

(8)

where k = ${\displaystyle \frac{1}{\ln (n)}} > 0, {\mathrm{e}}_j\ge 0$

To compute the information entropy redundancy of the indicator d_j.

$${\mathrm{d}}_{\mathrm{j}}=1-e_j$$

(9)

To count the indicator weights w_j.

$${\mathrm{w}}_{\mathrm{j}}={\displaystyle \frac{d_j}{{\displaystyle \sum _{j=1}^m{d}_j}}}$$

(10)

Finally, a composite score for country i is calculated.

$${\mathrm{s}}_{\mathrm{i}}={\displaystyle \sum _{j=1}^m{w}_j{p}_{ij}}$$

(11)

This paper combines the two indicators into index scores using the entropy method. These scores represent the high-quality development of Jiangsu Province’s reverse OFDI innovative direction. A higher index score signifies a more advanced level of innovation in the high-quality progression of reverse OFDI in Jiangsu Province.

Table 1. OFDI Innovation Dimension High-Quality Development System.

Primary Index	Secondary Index	Weight
The indicator score of the innovation dimension of the high-quality development of Reverse OFDI	Number of foreign subsidiaries in high-tech industries (+)	0.511
	R&D capital stock in host countries with spillovers from the OFDI channel (+)	0.489

shows the indicator system for constructing the indicator score of the innovation dimension of the high-quality development of reverse OFDI.

4.1.3. Global Value Chain Reconstruction: A Measure of Global Value Chain Status

Today’s measurement of GVC consists of two different levels: macro and micro. The macro level focuses on the country and industry levels. The micro level measures the embeddability of firms in GVCs through the measurement of their GVC foreign value-added (FVA) rates. This paper focuses on the macro level, drawing on the approach of Zhang et al. (2023) [45] based on the perspective of GVC status at the country level in GVC reconstruction. In this paper, the computational method proposed by Koopman (2010) [46] is selected for measurement.

Koopman (2010) [46] proposed the construction of a GVC position to measure whether a country is likely to be upstream or downstream of the global value chain. Koopman (2010) [46] gauged the ratio of the value-added exports used indirectly within a country compared to the value-added exports used by other countries, which is calculated by the following formula:

$${\mathrm{GVC}}_{\mathrm{it}}=\ln (1+{\displaystyle \frac{{\mathrm{IV}}_{\mathrm{it}}}{{\mathrm{E}}_{\mathrm{it}}}})-\ln (1+{\displaystyle \frac{{\mathrm{FV}}_{\mathrm{it}}}{{\mathrm{E}}_{\mathrm{it}}}})$$

(12)

In Equation (12), IV_it denotes the value added by indirect exports of country i in year t. FV_it denotes the value added by country i’s exports abroad in year t. E_it is country i’s total exports in year t. The greater the GVC_it, the more advanced the nation’s status in the global value chain is.

4.2. Variable Descriptions

4.2.1. Selection of Indicators

• Dependent variable. The innovative direction of the high-quality development of reverse OFDI indicator score (OFDII) [38,39]. We construct the above innovative direction of the high-quality development of Jiangsu’s reverse OFDI measurement indicator system and use the entropy weight method to appraise the score of the index. The consequences are displayed in Table 1.
• Independent variable. Global Value Chain Position (GVC). It was calculated using the methodology of Koopman (2010) [46], which is described in Equation (12).
• Control variables. This study selects six control variables and explains the selection reasons and measurement methods for each variable. First, Jiangsu Province has a relatively developed economy and contains some resource-based enterprises with excess production capacity. However, due to the scarcity of domestic resources, these enterprises invest abroad to acquire foreign natural resources, thereby supporting their high-quality development. Therefore, this paper controls for this variable to consider its effect. Natural resource endowment (Res) is evaluated using the ratio of fuel exports to merchandise exports, drawing on Cheung (2009) [47]. Res signifies the abundance of natural resources in a country, with a predicted positive impact. The daily activities of enterprises are closely related to the level of the labor force. A more abundant labor force means that employers have more choices. At this time, labor costs are reduced. Cross-border enterprises in Jiangsu Province can acquire the high-end human capital they need from abroad while reducing costs. This promotes high-quality development. Therefore, we control for the number of laborers. According to Herzer (2012) [48], the labor force level (Lab) is measured by the number of individuals in each country’s labor force. A larger labor force indicates more plentiful labor resources, leading to an expected positive outcome. Foreign direct investment represents the extent of market activities. The greater the extent of market activities in a country, the more intense the competition. This will impact Jiangsu Province’s overseas enterprises, which lack competitive advantages. It is detrimental to the high-quality development of OFDI. Therefore, we control for the amount of inflow of foreign direct investment. Foreign direct investment (FDI) is assessed using the net inflows of FDI in each country, following Siddica (2017) [49]. It may potentially produce a result with a negative coefficient. Technological innovation is often concentrated in high-end manufacturing industries. In countries where manufacturing is more developed, the level of innovation may be higher. Their innovation models are more diverse, allowing overseas subsidiaries to learn and absorb more knowledge. The development level of manufacturing affects the high-quality development of Jiangsu Province’s reverse OFDI innovation direction. Therefore, we control for the impact of the manufacturing development level. The manufacturing development level (Ind), based on Megbowon (2019) [50], is assessed by the share of manufacturing value added to the host country’s GDP, with a predicted positive coefficient. Infrastructure construction is characterized by large investment scales and slow benefits. Overseas enterprises from Jiangsu Province also need to fulfill relevant social responsibilities locally. Therefore, they need to incur higher infrastructure costs. The increase in such costs may squeeze out research and development innovation costs, affecting the high-quality development of the innovation dimension of OFDI. Therefore, we control for the level of infrastructure. Infrastructure level (Inf), drawing on Forte (2023) [51], is assessed by the number of fixed broadband subscribers per hundred people in the host nation. It might lead to a negative coefficient. By logarithmically transforming the above control variables, the potential heteroscedasticity problem in the econometric analysis caused by data volatility can be mitigated. Institutional distance represents the differences in political backgrounds between countries. These differences can make it challenging for enterprises from different countries to reach a consensus on cooperation. As a result, the acquisition of reverse technology spillovers may be hindered. This may negatively impact the innovative development of reverse OFDI. Therefore, we control for institutional distance. Institutional distance (Ins) is measured using the absolute value method [52] for calculation with the following formula:

  $${\mathrm{Ins}}_{\mathrm{ci}}={\displaystyle \frac{\left\{{\displaystyle {\sum }_{k=1}^N}\right|I_{kc}-I_{ki}\left|\right\}}{N}}$$

  (13)

  Ins_ci denotes China’s institutional differences from country i. These include six dimensions, namely public sound and accountability, political stability and the elimination of violence and terrorism, government efficiency, regulatory quality, rule of law level, and corruption control. I_kc is China’s score on institutional dimension k, I_ki is country i’s score on institutional dimension k, and N is the number of institutional environment dimensions, which is 6. This coefficient may be negative.

4.2.2. Data Sources

The dependent variable is the innovative direction of the high-quality development of Jiangsu’s reverse OFDI indicator score. The data of overseas subsidiaries in high-tech sectors in the host nation are selected from the database of CSMAR. The number of overseas subsidiaries of A-share listed firms in high-tech industries in Jiangsu Province from 2007 to 2021 is selected as the basis for innovative OFDI measurement. Data on R&D inputs required for the OFDI channel’s foreign R&D capital stock are derived from the World Bank, and the interpolation method is used to fill in the very few missing values. Meanwhile, a comparative analysis was conducted utilizing outbound investment data from the Jiangsu Provincial Bureau of Statistics. The selection of research subjects from among 16 developed countries, including Spain, Poland, Germany, Denmark, Austria, Finland, France, Italy, Japan, Singapore, the United States, Netherlands, Korea, Luxembourg, Belgium, and England, was based on the criteria of data availability and comprehensiveness.

The independent variable of this paper is the Global Value Chain Position, whose relevant data are selected from the UIBE Global Value Chain Database. Input–output data are selected from the Asian Development Bank database (ADB database). The ADB database contains input–output tables for 61 countries and 35 industry sectors, which ensures that the time and country data required for this paper are well-established. The data in the ADB database span from 2007 to 2021. Therefore, we have selected the period from 2007 to 2021 as the study timeframe. The data related to the control variables in this article are obtained from the World Bank and the World Governance Index (WGI). The data for the cultural distance variables are obtained from the official Hofstede website (Hofstede). The specific measurement methods of each variable are shown in

Table 2. Meaning of variables, calculation methods, and data sources.

Type		Mean	Method	Data Sources
Dependent variable	OFDII	The indicator score of the innovative direction of the high-quality development of Jiangsu Province’s Reverse OFDI	The entropy weighting method was used to compute the score based on the sub-indicators presented in Table 1	CSMAR, World Bank database
Independent variable	GVC	Global Value Chain position	Ratio of the value-added exports used indirectly within a country to the value-added exports used by other countries	UIBE database
Control variable	Res	Natural resource endowment	The percentage of fuel export to merchandise exports, in ln logarithms	World Bank
	Lab	The level of the labor force	The total labor force is treated in ln logarithmic terms (Unit: 100 million people)	World Bank
	Fdi	Foreign direct investment	Net inflows of foreign direct investment are treated in ln logarithmic terms (Unit: USD 100 million)	World Bank
	Inf	Infrastructure level	Number of fixed broadband subscribers per one hundred people, ln logarithmic processing	World Bank
	Ind	The level of manufacturing development	Host country manufacturing increment as a proportion of GDP is taken in ln logarithmic terms	World Bank
	Ins	Institutional distance	Calculated using the absolute value method based on six dimensions	WGI

.

Descriptive statistics are examined and presented in

Table 3. Descriptive statistics.

Variable	Obs	Mean	Std	Min	Max
OFDII	207	0.004	0.009	0	0.051
GVC	207	−0.003	0.028	−0.057	0.088
Ins	207	1.732	0.404	0.908	2.439
Res	207	1.527	0.916	−2.360	2.900
Inf	207	3.420	0.256	2.675	3.833
Ind	207	2.628	0.392	1.537	3.325
Fdi	207	5.763	1.361	1.699	8.503
Lab	207	−1.960	1.515	−6.041	0.505

. This study applies panel data from 16 developed countries spanning from 2007 to 2021. OFDII stands for the score indicating the innovative direction of high-quality progress in reverse OFDI. There is a positive correlation between this score and the level of high-quality development; as the score increases, so does the development level. GVC refers to the global value chain position. A higher GVC index signifies a stronger position within the value chain. The mean of the OFDII is 0.004, with a standard deviation of 0.009. The mean of GVC is −0.003, with a standard deviation of 0.028. The standard deviations of both OFDII and GVC are less than 0.1. This indicates that the differences in the high-quality development level of innovative dimensions in reverse OFDI and the global value chain positions of various countries are relatively minor. The standard deviations of institutional distance (Ins), natural resources (Res), manufacturing industry level (Ind), and infrastructure level (Inf) are all between 0.2 and 1. This indicates that the impact of these macro factors among the 16 developed countries mentioned is relatively stable. There are differences between countries, but the magnitude of these differences is not large. The maximum value of foreign direct investment (Fdi) is 8.503, with a standard deviation of 1.361. The maximum value of labor level (Lab) is 0.505, with a standard deviation of 1.515. This indicates significant differences in foreign investment and labor levels among these 16 developed countries. For some small developed economies in Europe, there is indeed a significant gap in their foreign investment capabilities and domestic labor force compared to large developed economies such as the United States.

Table 4. VIF.

Variable	Lab	Fdi	GVC	Ind	Ins	Res	Inf
VIF	2.150	1.820	1.750	1.740	1.640	1.270	1.130

shows the VIF test results of variables. The variance inflation factor test is also performed to avoid interference from multicollinearity. All values of the VIF of the variables are less than 5. We can exclude the possibility of multicollinearity interference from this result.

5. Empirical Analysis

5.1. Benchmark Regression

This article uses a two-way fixed-effects model for the regressions.

Table 5. Benchmark regression.

	(1)	(2)
	OFDII	OFDII
GVC		−0.068 ***
		(−2.74)
Ins	−0.011 *	−0.010
	(−1.83)	(−1.61)
Res	0.014 ***	0.014 ***
	(5.21)	(5.32)
Inf	−0.006	−0.006
	(−0.89)	(−0.86)
Ind	0.014 **	0.016 **
	(2.15)	(2.30)
Fdi	−0.000	−0.000
	(−0.32)	(−0.42)
Lab	0.067 ***	0.067 ***
	(5.68)	(5.66)
_cons	0.117 ***	0.109 ***
	(4.16)	(3.95)
year	Yes	Yes
country	Yes	Yes
N	207	207
R2	0.775	0.783
F	8.090	8.211

Notes: ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively.

reports the consequences of the benchmark regressions.

In Table 5, column (1), the association between the dependent variable and the control variables is illustrated. Institutional distance (Ins) reflects the similarity of different national backgrounds. Institutional distance impedes the innovative direction of the high-quality development of Jiangsu’s reverse OFDI. Greater disparities in institutional differences correspond to higher barriers of entry for OFDI activities among enterprises in Jiangsu Province. Therefore, enterprises tend to prioritize host countries with similar institutional distances to their home country to increase profits, thereby forming a solidified circle of cooperation. This runs counter to China’s long-standing policy of high-level opening-up. At this point, the government provides more information on the institutions and risks of host nations for overseas enterprises. This encourages them to engage in extensive cooperation. Simultaneously, the government fosters collaborations between domestic and foreign enterprises to create harmonious chambers of commerce. It promotes self-management and mutual assistance within the industry. On the other hand, natural resource endowment (Res) exerts a notably positive influence on the innovation dimension of high-quality development in Jiangsu Province’s reverse OFDI at a 1% level. Amidst a period of rapid economic expansion, China’s demand for natural resources is experiencing an upsurge. Therefore, natural resource-seeking enterprises in Jiangsu Province are facing a shortage of natural resources or significantly increased costs domestically. The abundance of foreign natural resources will further stimulate resource-seeking enterprises to invest abroad. This will promote high-quality development of outward investment. However, when confronted with the resource advantages of the host country, they may also make impulsive investments. At this point, the risk of corporate investment will increase. This is not conducive to the State Council’s policy of stabilizing scale and optimizing structure. The government in Jiangsu Province is providing more policies on the natural resources of host countries to help them improve their fault tolerance in the actual process of outward foreign investment.

The extent of infrastructure (Inf) development reflects a region’s capacity for industrialization. The infrastructure level in developed host countries is generally high. Enterprises in Jiangsu Province may struggle to achieve competitiveness in nations with advanced industrialization. This challenge may potentially impede their progress. The infrastructure costs for Jiangsu Province’s overseas enterprises in developed host nations will also increase. This may inhibit the high-quality development of reverse OFDI in Jiangsu Province. This is consistent with the benchmark regression results. Therefore, the Jiangsu provincial government has appropriately increased the subsidies for Jiangsu Province enterprises’ overseas operations in developed nations. Increased financial support is also the direction pointed out by the State Council in recent years. Moreover, the high-end manufacturing (Ind) sector serves as a pivotal avenue for technology acquisition in the context of innovative and high-quality development in OFDI. It produces a clearly significant positive result at a 5% level. Manufacturing is a major industry in technological development. The higher the level of manufacturing in the host country, the more beneficial it is for Jiangsu Province to achieve high-quality development in the innovative dimension of reverse OFDI. The Jiangsu provincial government is encouraging enterprises to invest in the manufacturing industry abroad. However, developed nations have transferred a large number of manufacturing plants to developing economies, leading to industrial hollowing out. Therefore, the Jiangsu government is also paying attention to balancing domestic employment and social stability to prevent the same industrial hollowing out results as in developed nations. The government has ensured the stability of domestic employment by accelerating the construction of manufacturing parks by large state-owned enterprises. By leveraging its own credit, the government can provide higher security for the parks, thereby stabilizing domestic employment levels.

The FDI can demonstrate the market competition in the host country. It is difficult for enterprises in Jiangsu Province to maintain sufficient competitive advantages in developed countries. The ultimate outcome indicates a restraining result. In the process of restructuring the global value chain, developed nations do not want to absorb low-quality foreign investment in order to climb to a higher position. However, most of the investments made by Jiangsu’s multinational enterprises in developed nations still pursue quantity rather than quality. Therefore, it is difficult to enter the high-end industries of developed economies. Nowadays, the government has shifted its foreign investment focus to key areas such as high-tech and modern services. This strategy aims to help enterprises integrate into developed economies’ markets more quickly. Furthermore, an excess of labor (Lab) in the host country can lead to a decrease in labor costs. Some overseas talents may find it difficult to obtain the expected salary in their domestic enterprises. Jiangsu’s overseas enterprises have been seeking high-end human capital from developed economies. Cross-border subsidiaries can attract this kind of overseas talent, leading enterprises towards a path of high-level and high-quality development. The analysis results show a 1% positive significance. This indicates that the labor force will contribute to promoting high-quality development of the reverse OFDI innovation dimension in Jiangsu Province. This will help China achieve its goal of building a strong talented country. The symbols representing each variable align with theoretical expectations. The coefficient of the OFDI innovative direction of high-quality development of Jiangsu Province’s reverse OFDI in column (2) of Table 5 is −0.068, which is significantly negative at the one percent level. Ye (2022) noted that nations with stronger positions in global value chains tend to have superior patent protection. The diffusion of technology is notably more challenging during OFDI events, which can consequently impede the innovation and quality enhancement of OFDI activities [53]. This observation aligns with the conclusions drawn in this study. The control variables in columns (1) and (2) of Table 5 exhibit the same orientation. Notably, the endowment of natural resources shows a significant positive effect, which corroborates the findings of Wang (2014) [54] and Ren (2020) [55]. The ongoing development of OFDI necessitates increasing resources, leading firms to agree to accept other challenges in resource-abundant nations. The consistent signs of the variables verify the robustness of the applied methodology.

According to the benchmark regression, a one percent rise in GVC status results in nearly a 0.1 percent decline in the innovative direction of the high-quality advancement of Jiangsu’s reverse OFDI. This finding supports Hypothesis H1. From the prior analysis, this study suggests that the restraining effect of global value chain restructuring on the innovative direction and high-quality progress of Jiangsu’s reverse OFDI primarily manifests in several key areas. Firstly, developed nations strive to lessen the competitive effects posed by multinational enterprises from Jiangsu Province on their economies. As a result, they impose technological barriers for the subsidiaries of these multinational firms, preventing them from accessing reverse technology spillovers that would benefit their parent companies. Secondly, digital technology has become the dominant trend in the Fourth Industrial Revolution. Developed countries have used their technological advancements and capital accumulation to dominate various related markets. This dominance puts enterprises from Jiangsu Province at a disadvantage when entering these markets. The lack of competitive advantage in the emerging technology sector has weakened the independent innovation capabilities of these enterprises. As a result, they face the risk of remaining at the lower end of the value chain. Furthermore, persistent geopolitical conflicts have led to numerous interruptions in the global supply chain network. Companies in Jiangsu Province are compelled to bear increased logistics expenses, which consequently diminish the funds allocated for research and development. This scenario impedes the upward movement at both ends of the “smile curve”, thus exerting a constraining effect on the high-quality development of innovation dimensions within Jiangsu’s reverse OFDI.

5.2. Robustness Test

This paper focuses on two main aspects of robustness testing. One approach is to use two-stage least squares in the regression model to tackle the endogeneity issue. The second step is to consider replacing explanatory variables and supplementing omitted variables for robustness testing.

5.2.1. Replacement of the Independent Variable

For the independent variables, the GVC position is re-measured using the method proposed by Wang (2017) [56,57]. The ratio of the length of the production of forward linkages to the length of the production of backward linkages is used to measure the following formula:

$${\mathrm{gvc}}_{\mathrm{it}}={\displaystyle \frac{{\mathrm{Plv}\_\mathrm{gvc}}_{\mathrm{it}}}{{ \mathrm{Ply}\_\mathrm{gvc}}_{\mathrm{it}}}}$$

(14)

where Plv_gvc_it denotes the length of forward-associated production in country i for period t, and Ply_gvc_it denotes the length of backward-associated production in country i for period t. The greater the gvc_it, the closer the position in the GVC is to the upstream, and the higher the GVC position.

Column (1) of

Table 6. Robustness test.

	(1)	(2)	(3)
	Replace the Independent Variable	Missed Variable Addition	2SLS
GVC		−0.069 ***	−0.104 **
		(−2.74)	(−2.23)
gvc	−0.035 **
	(−2.08)
Ins	−0.010 *	−0.010	−0.009
	(−1.71)	(−1.64)	(−1.38)
Res	0.014 ***	0.014 ***	0.014 ***
	(5.24)	(4.75)	(5.33)
Inf	−0.007	−0.005	−0.008
	(−1.09)	(−0.86)	(−1.04)
Ind	0.015 **	0.015 **	0.014 *
	(2.16)	(2.33)	(1.86)
Fdi	−0.000	−0.000	−0.000
	(−0.33)	(−0.40)	(−0.61)
Lab	0.068 ***	0.065 ***	0.071 ***
	(5.71)	(3.92)	(4.77)
Cul		−0.039
		(−0.19)
year	Yes	Yes	Yes
country	Yes	Yes	Yes
N	207	207	192
R2	0.779	0.783	0.606

Notes: ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively.

presents the regression results after adjusting the measurement method of the independent variable. The coefficient of gvc is −0.035, consistent with the direction of the benchmark regression. This result remains significantly negatively correlated at the 5% level. This illustrates that the restructuring of the global value chain still inhibits the innovation dimension of the high-quality development level of Jiangsu Province’s reverse OFDI. Different measurement methods for the explanatory variable do not affect the reliability of the research conclusions. There are no significant changes in the significance and direction of the coefficients of the control variables. This further emphasizes the reliability and consistency of the results.

5.2.2. Missed Variable Addition

Moreover, alongside the pertinent control variables included in the benchmark regression, cultural distance (Cul) might exert an impact on the innovative direction of high-quality development within Jiangsu’s reverse OFDI. Cultural distance (Cul) refers to the cultural disparities between China and other countries. In this paper, the cultural distance is further added as a control variable based on the benchmark regression. Cultural distance (Cul) draws on Yan Bing et al. (2022) [58] to subdivide culture into four dimensions, which are calculated as follows:

$${\mathrm{Cul}}_{\mathrm{ci}}={\displaystyle \frac{1}{4}}{\displaystyle \sum _{i=1}^4\frac{{(I_{kc}-I_{ki})}^2}{V_k}}+{\displaystyle \frac{1}{T_{ci}}}$$

(15)

where k denotes the kth dimension. V_k denotes the variance of the cultural distance in the kth dimension. I_kc represents China’s cultural score on the kth dimension, I_ki represents the host country i’s cultural score on the kth dimension. T_ci denotes the time when the host country i established diplomatic relations with China.

The findings are displayed in column (2) of Table 6, presenting the regression results subsequent to the inclusion of cultural distance as a control variable. Increased cultural disparity corresponds to diminished effectiveness in inter-country enterprise collaborations. The expected outcome of this variable is negative. The regression analysis shows that after adding cultural distance as a control variable, the coefficient of the explanatory variable GVC is −0.069. Regression analysis reveals a notably negative association at the 1% significance level. The significance levels and coefficient directions of the other control variables are roughly the same as in the benchmark regression. After adding additional control variables, the restructuring of the global value chain still has an inhibitory effect on the high-quality development of the innovation dimension of Jiangsu Province’s reverse OFDI. This observation reaffirms the reliability of the findings.

5.2.3. Endogeneity Test

This article uses a two-way fixed-effects model with panel data, which to some extent avoids the endogeneity problem caused by omitted variables. However, the problem of endogeneity may also arise from the mutual causation of the dependent and independent variables. Therefore, this paper uses the explanatory variables lagged by one period as tool variables and employs the 2SLS method for estimation. Column (3) of Table 6 presents the outcomes of 2SLS regression. The regression results show that the GVC coefficient is −0.104, consistent with the direction of the benchmark regression coefficient. This result is significantly negative at the 5% level. The sign and significance of the variable coefficients remain unchanged, consistent with the initial regression results.

In addition, we further adopt the difference-in-differences method to mitigate potential endogeneity problems. In 2013, General Secretary Xi Jinping proposed the Belt and Road Initiative, which actively promotes economic cooperation with partners through this initiative. The process of global economic cooperation is also part of the restructuring of the global value chain. Considering that this is a government-formulated policy, it can be regarded as an exogenous shock. According to the China Belt and Road website, among the developed nations studied in this paper, South Korea, Singapore, Austria, Poland, Luxembourg, and Italy have participated in the Belt and Road Initiative, while other countries are non-participating countries.

This paper follows the approach of Mugerman [59], employing the difference-in-differences method to address endogeneity issues. The sample participating in the Belt and Road Initiative is the experimental group, while the sample not participating in the Belt and Road Initiative is the control group. This paper defines the identification indicator Treat_it for the experimental and control groups. The experimental group sample is assigned a value of 1, and the control group is assigned a value of 0. The specific DID model is as follows:

$$OFDI{I}_{\mathrm{it}}={\alpha }_0+{\beta }_{1}Trea{t}_{it}\times Pos{t}_{it}+{\gamma }_{i}Control{s}_{it}+{\delta }_i+{\mu }_t+{\epsilon }_{it}$$

(16)

where Treat_it is the identifier for the experimental group, and Post_it is a time point variable. Post_it indicates whether country i joined the Belt and Road Initiative in year t. It takes the value 0 before the Belt and Road Initiative was proposed in 2013, and 1 afterwards. α₀ is the constant term. The coefficient β₁ of Treat_it × Post_it captures the policy impact. γ_i is the coefficient of control variables. δ_i and μ_t represent the country-level fixed effects and time-level fixed effects, respectively. ε_it is the error term.

Column (1) of

Table 7. Difference-In-Differences method.

	Jiangsu Province	Zhejiang Province
	(1)	(2)
	OFDII	OFDII
Treat × Post	−0.005 ***	−0.005 ***
	(−3.59)	(−3.94)
_cons	0.145 ***	0.127 ***
	(5.00)	(4.31)
Control variables	Yes	Yes
year	Yes	Yes
country	Yes	Yes
N	207	207
R2	0.786	0.780

Notes: *** represent significance at the 1% statistical levels.

shows the effect of the Belt and Road Initiative on the innovation direction of the high-quality development of reverse OFDI in Jiangsu Province. The coefficient of the Treat × Post interaction term is negative. The result is significantly negatively correlated at the 1% level. This indicates that the Belt and Road Initiative has had an inhibitory effect on the innovative direction of the high-quality development of reverse OFDI in Jiangsu Province. This may be because most of the developed economies involved in the Belt and Road Initiative are smaller countries. Non-participating large developed nations such as the United States have higher technological levels. Jiangsu Province is one of the more economically developed provinces in China, with a smaller technological gap with some small developed nations. Therefore, it is difficult to achieve higher levels of innovation development through external cooperation with these small developed economies.

We further selected Zhejiang Province as an example for comparison to support this view. Zhejiang Province, along with Jiangsu Province, is part of the Jiangsu–Zhejiang–Shanghai region. This region, including Jiangsu Province, Zhejiang Province, and Shanghai City, is the most economically developed area in China. Since Shanghai is only a city, comparing a single city with a province might introduce bias. Therefore, we selected Zhejiang Province for comparison with Jiangsu Province. Column (2) of Table 7 reports the impact of the Belt and Road Initiative on the innovative direction of high-quality development of reverse OFDI in Zhejiang Province. The results show that the coefficient of the interaction term is significantly negative at the 1% level. This supports our findings.

The parallel trends test is a key prerequisite for estimating the DID model. Figure 1 shows the parallel trends test image based on the DID model analysis using data from Jiangsu Province. Figure 2 shows the parallel trends test image based on the DID model analysis using data from Zhejiang Province. From the Figures, it can be observed that before the Belt and Road Initiative was proposed, the 95% confidence interval included the value of 0. There is no significant difference in the trend changes, which is consistent with the parallel trends assumption. The coefficient becomes negative in the first year after the policy, and significantly negative in the second year after the policy. This may be due to the lag effect of the policy.

Based on the consistency of the robustness test coefficients above, the findings of this paper are robust.

5.3. Heterogeneity Analysis

The process of GVC reconfiguration is affected by countries’ levels of urbanization. The urbanization standard is assessed in terms of the proportion of the host country’s urban population to the total population (City). The interaction term GVC × City is constructed to further delve into the heterogeneous roles of the level of urbanization in the relationship between the reconstruction of the global value chain and the innovation dimension of the high-quality development of Jiangsu’s reverse OFDI. Detailed findings are outlined in

Table 8. Heterogeneity analysis.

	(1)	(2)
	OFDII	OFDII
GVC	−0.068 ***	−0.418 **
	(−2.74)	(−2.26)
City		0.037
		(0.83)
GVC × City		0.405 *
		(1.93)
_cons	0.109 ***	0.066 *
	(3.95)	(1.68)
Control variables	Yes	Yes
Year	Yes	Yes
Country	Yes	Yes
N	207	207
R2	0.783	0.788

Notes: ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively.

.

We examine the mechanism by which the level of urbanization mitigates the negative impact of global value chain restructuring on the high-quality development of Jiangsu Province’s reverse OFDI innovation dimension through a moderating effect. Column (1) of Table 7 indicates a significantly negative coefficient for the baseline regression of GVC. In column (2) of Table 7, we simultaneously include the moderation variable City and the interaction term between the explanatory variable GVC and the moderation variable City, GVC × City. The results in column (2) of Table 7 show that the coefficient of GVC is −0.418, which is significantly negatively correlated at the 5% level. This is consistent with the findings in Column 1. The coefficient of the interaction term is 0.405, which is significantly positively correlated at the 10% level. The opposite signs of the interaction term and the explanatory variable coefficients indicate that the level of urbanization can weaken the negative impact of global value chain restructuring on the high-quality development of the innovation dimension of Jiangsu Province’s reverse OFDI.

In the process of global value chain restructuring, countries are striving to climb to higher-end positions. This situation has significantly driven technological development. Most of these technology-holding large enterprises are located in highly industrialized cities. Consequently, the level of urbanization increases as many rural laborers move to cities in search of higher incomes. However, as the urban population continues to grow, the trend of labor moving from rural to urban areas will eventually decline. This reduced population mobility will lead to a decrease in employment demand. When workers migrate from rural areas to cities, their wage expectations also rise, increasing the labor costs in the host country. Therefore, developed host countries may choose to collaborate with Jiangsu Province’s cross-border subsidiaries to gain relative labor cost advantages. When Jiangsu Province’s cross-border subsidiaries collaborate with these high-tech enterprises, they can achieve innovative high-quality development through the reverse technology spillover effect of OFDI. This can help mitigate the negative impact on the high-quality development of the innovation dimension of Jiangsu Province’s reverse OFDI caused by the technological barriers set by developed countries.

6. Conclusions and Recommendations

China is currently working to elevate its position in the global value chain. This effort aims to reduce its reliance on technology from developed countries, which remains a significant challenge. Consequently, the primary objective of OFDI in Jiangsu Province is to foster independent innovation and reach high-quality development. In recent years, Jiangsu’s OFDI has encountered substantial challenges due to COVID-19 and uncertain global economic trends.

This article establishes an index system for the innovation dimension of high-quality development in Jiangsu Province’s reverse OFDI. It investigates the impact of GVC restructuring from 2007 to 2021 on the level of the innovation dimension of high-quality advancement in Jiangsu Province’s reverse OFDI. The results suggest that the positioning of the global value chain has negatively affected the innovation dimension of high-quality advancement in the reverse OFDI within Jiangsu Province. A 1% decrease in the global value chain position will lead to a nearly 0.1% increase in the dimension of innovation in the high-quality advancement of the reverse OFDI in Jiangsu Province. The effectiveness of selecting developed host countries with relatively lower positions in the global value chain appears to enhance the innovative direction of high-quality development within Jiangsu’s reverse OFDI. Presently, the reshaping of the value chain remains predominantly orchestrated by developed nations. Led by the United States, these developed nations are forming collaborations to establish regional value chains. This strategy aims to maintain their dominant status in the value chain hierarchy and curb China’s rise. Their overarching objective is to facilitate a process of de-sinicization. Nonetheless, the ongoing restructuring of global value chains denotes this work in progress. This transitional phase unveils a spectrum of opportunities and challenges for Jiangsu Province. Capitalizing on the reconfiguration of GVC as a crucial opportunity, the concurrent implementation of diverse strategies will continue to foster the establishment of a novel developmental paradigm. These initiatives aim to drive the innovative dimension of high-quality reverse OFDI advancement.

6.1. Continuous Promotion of Independent Innovation

In Jiangsu Province, the high-tech sectors necessitate ongoing progress in autonomous innovation to augment their technological prowess. Leveraging reverse technology diffusion from developed nations serves as a viable avenue for technology acquisition, thereby fortifying their position in the value chain. Presently, developed nations have erected technological barriers against our nation. Establishing overseas collaborative research and development centers presents an opportunity to attract foreign enterprises for joint innovation endeavors, leveraging our inherent strengths. This strategy empowers us to fully harness their elite human capital and technical proficiencies.

Concurrently, our focus lies on the “Belt and Road”. We aim to study the research and development systems of international high-tech enterprises in external cooperation along the “Overseas Silk Road”. This can help improve the quality of supervision in all areas and the comprehensiveness of reward and punishment policies. In the subsequent stages of development, we can enhance our research and development mechanisms to advance the process of independent innovation.

6.2. Deep Integration of Domestic and Global Value Chains

We aim to leverage current international trade dynamics and our inherent capabilities to strengthen both domestic and global value chains by utilizing complementary strengths. This endeavor facilitates the formation of a resilient dual circulation framework. Within this framework, we can perpetually stimulate domestic consumption to reconfigure external trade, thereby bolstering our resilience against external influences.

To deepen its engagement in value chain evolution, Jiangsu Province ought to bolster multilateral collaboration frameworks. This endeavor will facilitate the creation of an interconnected supply chain network with other nations, consequently fostering greater receptiveness towards Jiangsu-based enterprises. Furthermore, efforts should concentrate on strengthening the integration of the real economy to reduce financial bubble vulnerabilities. This will help solidify the dual circulation mechanism.

6.3. Active Participation in Multilateral Governance of Global Trade

Enterprises are urged to embrace social responsibilities to support China’s efforts in advancing the global multilateral trade governance system. Continuous collaboration with other nations is vital in constructing overseas economic corridors, disrupting the value chain dominance maintained by developed nations. Moreover, establishing various overseas zones supports the advancement of the “Belt and Road” initiative. This fosters greater cooperation among enterprises in Jiangsu Province and helps numerous companies expand their international presence. Expanding the trade market within Jiangsu Province is crucial for enhancing the international competitiveness of enterprises not only within the province but also across China as a whole.

Enterprises should accurately evaluate their strengths and locate their own advantages. They can use their advantages to occupy an important position in the appropriate industry. Subsequent firms in Jiangsu Province can also strengthen cooperation with the local government to lower the entry threshold. Foreign trade in enterprises in Jiangsu Province is developing rapidly. As an important force of Chinese enterprises, this has also increased the voice of Chinese companies in global trade.

6.4. Increased Advocacy for China’s Outward Foreign Direct Investment

The media in developed nations still has some misleading reports about our country. China’s outward foreign direct investment is often reported as China’s attempt to monopolize the world, which damages our international image. Therefore, Jiangsu Province should increase propaganda efforts when conducting outward foreign direct investment. Our country can set up a positive and friendly cooperative reputation. Jiangsu’s companies can establish a sound propaganda and cultural system to show China’s great power status. The promotion of Chinese culture and the spirit of friendship can be included in staff training. Relevant people from all walks of life and local government departments can be actively invited to visit the place. In this way, we can promote cooperation with enterprises in developed countries. Then, it will gradually break the status quo of technology blockades in developed countries and obtain the innovative direction of high-quality development of reverse OFDI.

6.5. Limitations and Future Research

This article’s sample primarily consists of A-share listed companies in Jiangsu Province, which limits the study’s generalizability. It is suggested that future studies should expand the sample source. Additionally, to ensure the relative sustainability and integrity of data, only 16 developed countries with comprehensive data were chosen as research subjects. It is recommended to explore other indicators with better data quality in future studies. By including more indicators, the innovation dimension of the high-quality development of Jiangsu Province’s reverse OFDI can be more comprehensively represented. While this study focuses on the national level, future research could refine the analysis to the enterprise and industry levels. This would provide a more detailed direction for the innovation dimension of Jiangsu Province’s high-quality reverse OFDI development.