A Research on the Sustainable Impact of FTA Strategy on the Global Value Chain Embedding of Listed Enterprises in China

Abstract

The Free Trade Area (FTA) strategy and the participation of enterprises in global value chains (GVCs) are important aspects of China’s high-quality economic development stage. This study matches trade data from the China Customs Import and Export database with information from listed firms in the CSMAR database, calculating the firms’ GVC embeddedness and the depth of trade agreements at the firm level. On this basis, this research employs a gravity model with fixed effects to empirically analyze the impact and mechanism of the FTA strategy on the embedding of Chinese listed firms in GVCs, utilizing data from 2000 to 2006. The results demonstrate that the FTA strategy substantially enhances the embeddedness of Chinese listed enterprises in GVCs. The heterogeneity analysis indicates that state-owned enterprises, those located in the central and western regions, manufacturing firms, and high-tech industry enterprises derive greater advantages from the FTA strategy in terms of their embeddedness in GVCs. Moreover, the mechanism analysis indicates that the FTA strategy enhances the embeddedness of enterprises in GVCs by increasing their technological innovation levels. Additionally, the internal control costs of enterprises negatively moderate the impact of the FTA strategy on their embedding in GVCs, and a “substitution effect” exists between asset operating efficiency and the FTA strategy in promoting the GVC embedding of listed firms. These findings provide empirical evidence and policy recommendations for the Chinese government to enhance the FTA strategy and sustainably improve the embeddedness of Chinese listed enterprises in GVCs.

1. Introduction

The 14th Five-Year Plan for National Economic and Social Development of the People’s Republic of China proposes a strategy to advance the establishment of “high-standard, expansive, and profound” free trade areas. As of March 2025, China has signed 23 free trade agreements (FTAs) with 30 nations and regions, thereby creating a three-dimensional network characterized by “peripheral radiation and global layout”. Meanwhile, China is actively participating in the global industrial division of labor, playing a crucial role in the advancement of global value chains (GVCs). At the macro level, the advantages of GVCs for economic growth, employment, and reducing poverty are expected to be sustainable, driven by favorable policies and technological changes [1,2]. At the micro level, GVC embeddedness, as an essential channel for international technological spillovers, significantly promotes firm innovation [3,4]. The FTA strategy has greatly contributed to China’s extensive opening to the global market, fostering enterprise innovation and bolstering foreign trade competitiveness, thereby serving as a crucial initiative for enhancing China’s role in the international division of labor. Moreover, the FTA strategy facilitates the enhancement of Chinese enterprises within the GVCs via the effects of research and development (R and D) innovation, which significantly contributes to overcoming the “low-end lock-in dilemma” in the value chain advancement of Chinese enterprises. This article therefore examines the impact and mechanisms of the FTA strategy on the GVC embeddedness of Chinese listed enterprises, thereby providing a novel perspective for optimizing China’s FTA strategy and proposing fresh approaches to enhance export competitiveness and GVC positioning.

The continuous deepening of the global industrial division of labor has drawn significant attention to the interaction between free trade agreements and global value chains. A large number of studies have confirmed the significant impact of rising trade liberalization and the rapid growth of FTAs on GVCs [5,6,7,8]. Certain studies have found that an increased depth of free trade agreements (FTAs) enhances the vertical specialization of member nations [9,10,11], thereby demonstrating the upgrading effect of FTA depth on the positioning of GVCs at both national and industry levels. Simultaneously, certain studies have examined the impact of global production fragmentation on the design of trade agreements [12,13,14,15]. Bondi et al. (2025) [16] argue that firms involved in GVC trade play an important role in driving deeper agreements. They found that trade associated with GVC activities increases the possibility that agreements will include investment and competition provisions, as well as chapters on labor rights and environmental standards. Technological advancement can enhance GVC participation by mediating external environmental factors [17]. Nonetheless, there is still a lack of research on the mechanisms by which the FTA strategy influences the integration of publicly listed enterprises within GVCs, and current studies do not comprehensively summarize the heterogeneous effects on various companies’ engagement in GVCs.

Investigating the influence of the FTA strategy on the embeddedness of Chinese listed firms within GVCs holds dual significance: theoretically, it addresses the shortcomings of conventional research that emphasizes macro over micro perspectives; practically, it helps refine the design of FTA provisions and develop specific policies to support enterprises. Competition policy can diminish market access barriers, while environmental rules can foster green technical innovation. These behind-the-border measures are essential for the long-term competitiveness of enterprises. Peng et al. (2020) [18] emphasize the positive effect of Chinese FTAs on the upgrading of partner countries within GVCs, based on China’s trade agreements between 2010 and 2015. Nevertheless, the majority of current research has concentrated on conventional advantages, such as tariff reductions, while paying insufficient attention to “hidden dividends,” including the enhancement of firms’ positions in GVCs. The position of enterprises within GVCs influences not only the sustainability of their competitive advantages but is also correlated with the quality of China’s economic development. Consequently, it is necessary to develop an analytical framework that includes both institutions and firms in order to evaluate the impacts of the FTA strategy systematically.

This study, utilizing data at the firm–destination country level from 2000 to 2016 matched from the China Stock Market and Accounting Research (CSMAR) Database and the China Customs Database, elucidates the various paths of the FTA strategy on the embeddedness of Chinese listed firms in GVCs. The research findings reveal that the FTA strategy greatly enhances the integration of listed firms in China into GVCs, with notable heterogeneity across various enterprise types. The article additionally examines how the FTA strategy influences enterprises’ GVC embedding by enhancing innovation levels and managing internal costs, while also addressing the substitution effect of asset operation efficiency and the FTA strategy on firms’ integration into GVCs.

The contributions of the study are as follows. First, it develops indicators of GVC embeddedness utilizing microdata from publicly listed companies, thereby overcoming the constraints of conventional macro measurement. Second, this study employs a two-way fixed-effects model to assess the causal impact of the FTA strategy on the GVC integration of China’s listed enterprises and explain its operational mechanism by examining mediating and moderating effects. Third, it examines the impact of the ownership structure, regional distribution, industry affiliation, and the heterogeneity of industry technological attributes on enterprises, providing a basis for designing differentiated support policies. This research framework enhances the theoretical understanding of the micro effects of FTAs. It provides a practical reference for Chinese listed firms to transition from “precise embedding” to “efficient upgrading” within global value chains.

2. Literature Review and Hypotheses

2.1. Literature Review

2.1.1. FTAs and GVC Embeddedness

The first strand of literature relevant to this topic is research into the influence of free trade agreements on global value chains. Prior research has focused on the overarching connections between regional economic integration and enhanced embeddedness in GVCs. Kawai and Wignaraja (2013) [5] examined trade agreements established in Asia, arguing that FTAs have emerged as a vital instrument of regional economic policy. They analyzed the positive correlation between the increasing number of FTAs and the growth of GVCs. Kowalski et al. (2015) [6] investigated the positive correlation between the rise in the number of FTAs and the growth of GVCs, utilizing the OECD TiVA value-added trade database, and explored the determinants that facilitate the advancement of GVCs. Their findings suggest that trade agreements can foster the vertical specialization of intraregional production and increase the extent of forward and backward links among member nations within GVCs. Zhang et al. (2021) [19] found that FTAs not only promote the export levels of simple and complex value chains but also enhance value-added exports of all GVC sub-items at various levels. Mohanty (2024) [20] evaluated 15 FTAs implemented by India from 2000 to 2023, highlighting the important role that FTAs play in enabling the deeper integration of countries such as India into GVCs.

On this basis, certain studies have enriched the investigation of the interaction between FTAs and GVC embeddedness from different perspectives. Laget et al. (2020) [21] examined the implementation of RTAs, employing the approach of Horn et al. (2010) [22] to develop a depth index of RTAs and assess the impact of FTA depth on GVC trade. Mattoo et al. (2022) [11] categorized FTAs into deep and shallow agreements, discovering that deep agreements, which include comprehensive provisions like digital trade and support for small- and medium-sized enterprises (SMEs), enhance the vertical specialization of member countries, augment their trade, and mitigate the crowding-out effect on non-member countries. Focusing on GVC-related services, Díaz et al. (2022) [23] examined the effect of FTAs with services provisions on the enhancement of trade in value-added services among partner countries. Fan et al. (2023) [24] utilized panel data from 43 countries to investigate how the promotional effect of trade agreements on GVC participation varies with the distance in production position between member countries in GVCs from 2000 to 2014. They argue that production position distance has an inverted U-shaped effect on the link between FTAs and GVC participation. Aly and Zaki (2025) confirmed the positive relationship between the depth of trade agreements and GVCs while also highlighting the positive impact of improvements in institutional quality on this relationship [25].

In summary, previous research has analyzed the influence of FTAs on value chain trade and the extent of GVC embeddedness at the macro level; however, there is a lack of research investigating the effects of FTAs on firms’ GVC embeddedness at the micro level [26,27]. Using only country- or industry-level data on GVC embeddedness makes it challenging to examine the micro-transmission mechanisms that influence enterprises’ participation in the global division of labor.

2.1.2. The Measurement of GVCs

Another category of literature relevant to this topic is research focused on quantifying the positioning of GVCs. Traditional trade statistics, based on gross trade calculations, overlook trade value-added components and double counting. In contrast, trade value-added statistics account for the double counting of intermediary items and facilitate the examination of nations’ participation in global value chains. Hummels et al. (2001) [28] introduced the Hummels–Ishii–Yi (HIY) method to assess the extent of countries’ embeddedness in GVCs. The HIY method relies on total exports, with the proportion of domestic re-exports of intermediates in exports representing the forward vertical specialization rate, the proportion of foreign intermediates in exports representing the backward vertical specialization rate, and the aggregate of both figures reflecting the extent of GVC embeddedness. Based on this foundation, Koopman et al. (2010) [29] introduced the KPWW method to assess GVC positioning through the export value-added decomposition framework. Wang et al. (2017) [30] categorized global production activities into three classifications: purely domestic, traditional Ricardian, and GVC, and determined the sector’s relative position within GVCs from both supply and demand perspectives. Nonetheless, all the mentioned methods evaluate GVC embeddedness using country- or industry-level data and are inapplicable at the micro-firm level.

To address this limitation and enable firm-level analysis, Upward et al. (2013) [31] refined the KPWW method, focusing on the limitation of Hummels et al. (2001) [28] in failing to differentiate between processing trade and general trade. They assessed the foreign value added of Chinese enterprises using China’s customs data, alongside aggregated information from industrial enterprises, and introduced a framework for measuring foreign value added applicable at the enterprise level. Zhang et al. (2013) [32] enhanced the above measuring method by integrating variables such as trade agents and indirect imports, revealing a decreasing trend in the domestic value-added rate of Chinese manufacturing companies from 2000 to 2006. Utilizing the technique of Upward et al. [31] (2013), Lv et al. [33] (2015) investigated the influence of efficiency and finance limitations on firms’ participation in GVCs, revealing that firms with high efficiency are more likely to participate in GVCs. These studies have developed a measurement framework applicable to the involvement of Chinese enterprises in GVCs; nonetheless, the study of the position of global value chains predominantly remains at the national or industrial level [8].

2.2. Hypotheses

The FTA strategy mainly promotes the integration of firms in GVCs via three distinct pathways. (1) The specialization trend founded on comparative advantage: free trade agreements reduce trade costs and facilitate the allocation of resources towards industries exhibiting a comparative advantage [34]. Subsequently, listed firms participate in global production through the exchange of intermediate goods and the vertical division of labor [35], hence enhancing their level of GVC integration [28]. (2) Efficiency enhancement through transaction cost optimization: FTAs reduce institutional transaction costs through investment facilitation, regulatory coordination, and other institutional mechanisms [36,37,38], improve the adaptive capacity and responsiveness of listed enterprises within the international division of labor, and bolster the stability of participation in GVCs. (3) Value chain upgrading based on economies of scale: FTAs expand market demand via regional economic integration [39], enabling publicly traded firms to leverage large-scale production to diminish unit costs and develop technological advantages [40,41], which facilitates the transition from low-value-added parts to high-value-added parts, thereby strengthening their position as leaders in the GVCs. The above studies indicate that an increased coverage of FTA provisions correlates with a more pronounced embedding of enterprises within GVCs. This study posits the following hypothesis based on the preceding analysis:

Hypothesis 1.

The FTA strategy significantly improves the embeddedness of GVCs for listed firms in China, with greater coverage of FTA provisions correlating to a more pronounced impact on this embedding.

The level of technical innovation in enterprises significantly mediates the integration of firms into GVCs through FTAs. The incentive mechanism of free trade agreements on the technological innovation capacity of firms is reflected through a dual pathway. On the one hand, market expansion encourages firms to increase their investment in innovation. Due to tariff reductions and the easing of market access restrictions from FTAs, firms’ foreign market scale expands, and the increased marginal returns improve the viability of R and D investment [42]. Meanwhile, in response to heightened demands of the international market, firms will enhance product quality and technical standards via technological innovation [43]. On the other hand, “behind-the-border” measures, including technical regulations, quality certification, environmental protection, and labor standards within FTAs, represent institutional constraints. These external regulations encourage enterprises to transition from process innovation to product innovation in order to comply with international standards [11,44].

The enhancement of enterprises’ technological innovation levels facilitates their integration into GVCs in three ways. First, the acquisition of patents and significant technological advances by companies renders them indispensable in the global division of labor, enabling them to engage in high-value-added activities such as research and development, design, and core component manufacturing [45]. Second, technological superiority fosters collaboration among firms, multinational corporations, and research institutions, enabling them to access advanced technologies and high-end orders by embedding them into the global innovation network [46]. Finally, innovation-driven improvements in the production process reduce marginal costs, enhance the price–performance ratio of products, and increase the negotiating power of firms within the global supply chain [47].

The mentioned mechanism follows the rationale of “institutional environment-innovation incentive-value chain upgrading”. This study proposes the subsequent hypothesis based on the preceding analysis:

Hypothesis 2.

The FTA strategy promotes the embeddedness of China’s listed enterprises in GVCs through the mediated effect of firms’ technical innovation level.

The efficiency with which enterprises convert external institutional dividends is determined by the rationality of their internal resource allocation [36]. Firms with low internal control costs have low management redundancy, short decision-making processes, and can identify and utilize FTA policy instruments quickly. The resources saved by such firms can be directly invested in R and D innovation and GVC upgrading, which strengthens the mediating role of technology [48]. On the contrary, firms with high internal control costs are more vulnerable to information lags and resource mismatches, limiting the impact of technical innovation in improving GVC embedding [49,50]. According to the firm’s resource-based theory, internal organizational efficiency determines the real effectiveness of the external institutional environment in transforming its competitive advantage [51].

Hypothesis 3.

Enterprises’ internal control costs have a negative moderating impact on the connection between FTA strategy and embeddedness in GVCs.

Enterprises with highly efficient asset operations tend to improve inventory turnover and return on assets through lean manufacturing and digital management. Even in the absence of external policy support for FTAs, these firms can still gain a cost advantage in GVCs through efficient asset turnover [52], implying that efficient asset operations can “substitute” for the value chain improvement effect of the FTA strategy. Low asset efficiency firms, on the other hand, face the dual constraints of internal inefficiencies offsetting the cost savings from policy dividends and limited cash flow to support GVC upgrading due to slow inventory turnover and high asset idleness [53]. Tariff reductions and customs facilitation through FTAs are more effective for these firms, providing incentives to release capital for technological innovation or global production deployment. This replacement link between asset operation efficiency and FTA strategy is simply a dynamic balance between the enterprise’s capability and external policy: asset operation efficiency reflects the enterprise’s ability to allocate resources, and efficient operation can minimize reliance on FTA strategy dividends, whereas inefficient operation can increase reliance on FTA strategy dividends. This rationale is in line with the “structure-behavior-performance” paradigm in industrial organization theory [54], which states that a firm’s internal structural qualities influence its behavioral reaction to external policies as well as its performance output. Thus, this paper proposes the following hypothesis:

Hypothesis 4.

Enterprises’ asset operation efficiency has a positive effect on their GVC embeddedness, and the effects of asset operation efficiency and FTA strategy on GVC embeddedness are substituted.

3. Research Design

3.1. Data and Sample

The financial information disclosed by listed enterprises is more comprehensive and reliable, and as a market entities with excellent resource integration ability, they are more affected by the division of labor within GVCs. Therefore, this paper uses Chinese listed companies from the China Stock Market and Accounting Research (CSMAR) database as subjects for research. Given that the company information data in the China Customs Import and Export database has only been updated to 2016, this study focuses on the period from 2000 to 2016. It employs Lv’s [33] method of conducting empirical tests by matching the information of listed enterprises in the CSMAR database with the trade data of enterprises in the China Customs Import and Export database.

The specific matching procedure is as follows. Initially, we collated customs import and export data. Customs trade data is in monthly units, yet the CSMAR database contains annual data. So, we first summarized the import and export data by year and removed observation samples with missing key variables or abnormal assignments (e.g., records with negative import and export values). Second, screening of firms in the CSMAR database was conducted. We eliminated (1) enterprises that may be financially risky (ST or *ST), (2) samples with incorrect data (e.g., negative employee wages), and (3) samples missing key financial indicators. Third, data matching was performed. (1) Information on listed firms in the CSMAR database was extracted, including basic information fields such as firm name, legal representative, zip code, and contact phone number. (2) Initial matching was performed using information that is not easy to change in the short term and has high stability (e.g., firm name and legal representative, etc.). (3) A second matching phase was carried out using the zip code and contact phone number for firms that were unsuccessful in the initial match. (4) The results of the two matches were combined and the matching errors caused by the enterprise’s name change using the enterprise’s stock code (which is unique and unchanged unless delisted) were calibrated.

By following the method outlined above, a consolidated dataset, which includes company details, financial indicators, and import/export records, was generated. The final sample consisted of 12,879 items.

This research used the methods of Hofmann et al. [55] (2017) and Hou et al. [27] (2023) to assess the level of execution of the Free Trade Area strategy in terms of the depth of provisions in FTAs signed by China. The World Bank Deep Trade Agreements (DTA) database provided information on free trade agreements, while the CSMAR and the United Nations Commodity Trade Statistics (UN Comtrade) databases were utilized to determine firm weights.

3.2. Measures

3.2.1. The Dependent Variable

The dependent variable in this study is enterprises’ GVC embeddedness ($GV{C}_{it}$), indicating the degree to which enterprises are embedded in GVCs. This research builds on Upward et al. [31] and Lv et al. [33] (2015) to describe enterprises’ GVC embeddedness. To begin with, it is assumed that all imports serve as intermediate inputs. The intermediate inputs of processing trade imports are entirely utilized for processing trade exports, while the intermediate inputs of general trade imports are used in equal parts for domestic sales and general trade exports. The foreign value-added rate for an enterprise’s exports is

$$FVA{R}_{it}={\displaystyle \frac{V_F}{X}}={\displaystyle \frac{M^P+X^O\left[M^O/\left(D+X^O\right)\right]}{X}}$$

(1)

where, $FVA{R}_{it}$ represents the foreign value-added rate of the enterprise’s exports, $V_F$ denotes the foreign value added of the enterprise’s exports, $X$ indicates the enterprise’s total exports, $M^P$ indicates the enterprise’s processing trade imports, $M^O$ refers to the enterprise’s general trade imports, $X^O$ refers to the enterprise’s general trade exports, and $D$ represents the enterprise’s domestic sales. For an enterprise with a total sales value less than the export delivery value, the foreign value added of the enterprise’s exports is presumed to equal the sum of the enterprise’s processing trade imports and general trade imports, denoted as $V_F=M^P+M^O$. If the foreign value added to the enterprise’s exports exceeds the total exports ($V_F>X$), the foreign value-added rate (FVAR) is established at 1, while the domestic value-added rate (DVAR) is established at 0. Given that the total of the exports’ foreign and domestic value-added rates is 1 ($DVA{R}_{it}+FVA{R}_{it}=1$), the enterprise’s GVC embeddedness, A, can be calculated as follows:

$$GV{C}_{it}=\ln \left(1+DVA{R}_{it}\right)-\ln \left(1+FVA{R}_{it}\right).$$

(2)

Figure 1 shows the total and mean trends of GVC embeddedness of Chinese listed enterprises from 2000 to 2016.

3.2.2. The Independent Variables

This study evaluates the strength of the FTA strategy by assessing the total depth ($Total\_FT{A}_{it}^f$) and the core depth ($Core\_FT{A}_{it}^f$) of free trade agreements (FTAs) [27,55]. The specific steps are as follows: The 52 clauses of the FTAs (14 WTO+ clauses and 38 WTO-X clauses) are evaluated. If the FTAs include a specific provision k, a value of 1 is assigned to $\mathrm{Pr}ovisio{n}_k$; otherwise, it is assigned a value of 0. The quantity of provisions encompassed by the FTAs is summed, and their proportion to the overall number of provisions (52) is utilized to assess the coverage of the FTAs:

$$FT{A}_{it}={\displaystyle \sum _{k=1}^{52}\frac{\mathrm{Pr}ovisio{n}_k}{52}}$$

(3)

where j denotes partner countries that have signed FTAs with China.

Secondly, considering the differences in trade volume with various partner nations, the study employs the ratio of China’s exports to partner countries j ($Expor{t}_{jt}$) relative to total exports ($Expor{t}_t$) to weight the indicators mentioned above, thereby obtaining the average depth level of all FTAs signed by China in year t, calculated using the following formula:

$$Avg\_FT{A}_t={\displaystyle \sum _j^N\frac{Expor{t}_{jt}}{Expor{t}_t}}\times FT{A}_{jt}$$

(4)

N is a set of partner countries, encompassing all nations with which China established FTAs in year t. This step can elucidate the importance of trade size in evaluating the effects of FTAs, which aligns more closely with the improvement of the actual impact of FTAs by trade size in economic reality. This study additionally uses the ratio of exports ($X_{it}$) to total sales ($Sale{s}_{it}$) of listed firms to further weight and deconstruct the macroeconomic effects of FTAs into microeconomic impacts on individual enterprises, thereby addressing the issue of aligning the dimensions of macro-FTA data with micro-enterprise data. The specific formula is as follows:

$$Total\_FT{A}_{it}^f={\displaystyle \frac{X_{it}}{Sale{s}_{it}}}\times Avg\_FT{A}_t$$

(5)

As a result, the firm-level independent variable of the total depth of FTAs ($Total\_FT{A}_{it}^f$) is identified. Following the methods outlined above, the 52 provisions are further filtered down to 18 core provisions to generate the core depth ($Core\_FT{A}_{it}^f$) of FTAs.

3.2.3. Mediating Variables

This study employed the methodologies of Yuan et al. [56] (2015) and Huang Bo et al. [57] (2023) to examine the impact mechanism of the FTA strategy on the GVC integration of Chinese listed enterprises. It utilizes the volume of patent applications as proxy variables to evaluate these enterprises’ technological innovation levels. This is calculated by taking the natural logarithm of the number of patent applications plus one. Data were sourced from the China Enterprise Patent Database, encompassing total patent applications ($Patent\_T_{it}$), invention patent applications ($Patent\_I_{it}$), utility model applications ($Patent\_U_{it}$), and design applications ($Patent\_D_{it}$).

3.2.4. Moderating Variables

This study employed the management expense ratio ($Mgt\_Fe{e}_{it}$) from the CSMAR database to assess the internal control costs of listed firms, and the inventory turnover ratio ($Stock\_Tur{n}_{it}$) to evaluate the operational efficiency of enterprise assets. The management expense ratio is the proportion of management expenses to operating income, indicating the management costs a company incurs to generate a unit of operating income. A higher management expense ratio indicates greater resource consumption in organizational management operations and increased internal control expenditures. The inventory turnover ratio is the quotient of operating expenses and average inventory, indicating the rate at which a company’s inventory is utilized in production and sales.

3.2.5. Control Variables

(1) The age of the enterprise ($\ln \left(ag{e}_{it}\right)$) is measured by the natural logarithm of the number of years since its establishment, calculated as the difference between the current year and the year of establishment, plus one. (2) Factor intensity ($\ln \left(k{l}_{it}\right)$) is measured by the natural logarithm of the ratio of fixed assets to employment. (3) Financing constraint ($\ln \left(financ{e}_{it}\right)$), calculated by the natural logarithm of the ratio of interest expenditure to total fixed assets; a higher value indicates a reduced financing constraint for the firm. (4) Firm size ($\ln \left(Siz{e}_{it}\right)$) is calculated by the natural logarithm of the enterprise’s net fixed assets. (5) Industry concentration ($hh{i}_{it}$) is quantified using the Herfindahl–Hirschman Index (HHI), calculated as the sum of the squares of each firm’s sales percentage relative to the overall industry sales for the respective year. The descriptive statistics of the variables are presented in

Table 1. Descriptive statistics and data sources.

Variable Type	Symbol	Data Resource	Mean	Standard
Dependent variable	$GV{C}_{it}$	CSMAR database, the China Customs Import and Export database	0.382	0.526
Independent variable	$Total\_FT{A}_{it}^f$	DTA database UN Comtrade CSMAR database	0.005	0.013
	$Core\_FT{A}_{it}^f$		0.005	0.012
Mediating Variables	$Patent\_T_{it}$	China Enterprise Patent Library	3.457	1.617
	$Patent\_I_{it}$		2.867	1.642
	$Patent\_U_{it}$		1.663	1.776
	$Patent\_D_{it}$		0.838	1.469
Moderating Variables	$Mgt\_Fe{e}_{it}$	CSMAR database	0.119	1.536
	$Stock\_Tur{n}_{it}$		15.024	737.447
Control Variables	$\ln \left(ag{e}_{it}\right)$	CSMAR database	1.875	0.659
	$\ln \left(k{l}_{it}\right)$		12.411	0.934
	$\ln \left(financ{e}_{it}\right)$		−3.575	0.254
	$\ln \left(Siz{e}_{it}\right)$		20.028	1.219
	$hh{i}_{it}$		0.095	0.102

.

3.3. Model Specification

This study developed a fixed effects model to evaluate the influence of FTA strategy on embeddedness in the global value chain of China’s listed enterprises:

$$GV{C}_{it}={\delta }_0+{\delta }_{1}FT{A}_{it}^f+{\delta }_2{X}_{it}+{\theta }_i+{\theta }_t+{\epsilon }_{it}$$

(6)

where the subscripts i and t represent Chinese listed firms and years, respectively; $GV{C}_{it}$ signifies the GVCs embeddedness of enterprise i in year t; $FT{A}_{it}^f$ serves as a proxy for the FTA strategy, measured by the total depth of FTAs ($Total\_FT{A}_{it}^f$) and the core depth of FTAs ($Core\_FT{A}_{it}^f$), adjusted at the firm level; $X_{it}$ denotes a set of control variables; ${\theta }_i$ and ${\theta }_t$ indicate the enterprise fixed effect and year fixed effects, respectively; and ${\epsilon }_{it}$ represents the error term. According to previous research, a significant positive ${\delta }_1$ suggests that the FTA strategy substantially enhances the GVC embeddedness of the listed firms in China.

4. Empirical Results

4.1. Benchmark Results

The research employed a fixed-effects model utilizing ordinary least squares (OLS) regression to analyze the influence of FTA strategy on the embeddedness of Chinese listed enterprises in GVCs.

Table 2. Benchmark estimation test.

Variables	$\mathit{G}\mathit{V}{\mathit{C}}_{\mathit{i}\mathit{t}}$
	(1)	(2)	(3)	(4)
$Total\_FT{A}_{it}^f$	5.8295 *** (0.6618)		6.9256 *** (0.8287)
$Core\_FT{A}_{it}^f$		8.0931 *** (1.0017)		9.4255 *** (1.2739)
$\ln \left(ag{e}_{it}\right)$			−0.0359 * (0.0206)	−0.0342 * (0.0205)
$\ln \left(k{l}_{it}\right)$			−0.0426 *** (0.0112)	−0.0429 *** (0.0111)
$\ln \left(financ{e}_{it}\right)$			0.2469 *** (0.0941)	0.2404 ** (0.0939)
$\ln \left(Siz{e}_{it}\right)$			0.0238 * (0.0128)	0.0247 * (0.0129)
$hh{i}_{it}$			0.1582 ** (0.0723)	0.1525 ** (0.0723)
Constant	0.3546 *** (0.0044)	0.3466 *** (0.0054)	1.3379 *** (0.3468)	1.2883 *** (0.3468)
Enterprise fixed	YES	YES	YES	YES
Year fixed	YES	YES	YES	YES
Observations	12,879	12,879	10,276	10,276
R-squared	0.644	0.645	0.646	0.647

Note: Values in parentheses are clustering robust standard errors. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

presents the results of the benchmark regressions: column (1) presents the regression results for total FTA depth $Total\_FT{A}_{it}^f$, while column (2) shows the regression results for core depth $Core\_FT{A}_{it}^f$; columns (3) and (4) provide the estimated coefficients for total depth and core depth, respectively, when the control variables are included. The regression coefficients for the total depth of FTAs and the core depth of FTAs are both statistically positive at the 1% level. The findings indicate that the FTA strategy significantly improves the GVC integration of Chinese trading firms. The promotional effect is especially pronounced in core depth, suggesting that the core provisions, such as market access, investment protection, and competition policy, are more direct and effective in motivating enterprises to engage in GVCs; thus, the quality of provisions is more crucial than their quantity.

Increasing the depth of FTAs essentially enhances the institutionalized openness and regulatory harmonization of the partner country. Trade liberalization diminishes cross-border transaction costs, and enhances trade volume and participation in GVCs [58]; investment protection provisions motivate firms’ outward foreign direct investment (OFDI) and integration into GVC via international production; additionally, “behind-the-border” measures, including intellectual property rights (IPR), safeguard enterprises’ innovative achievements and their outcomes [59], while also facilitating the transition of firms from processing and manufacturing to technological innovation. Furthermore, as a vehicle for policy convergence among member nations, FTAs can significantly diminish information asymmetry and institutional barriers in transnational cooperation [60], while also offering institutional protection for firms.

4.2. Robustness Test

This study adopted the following methods to test the robustness of the benchmark results:

• Exclude external shocks. The global financial crisis of 2008–2009 had a significant impact on the global trade network; therefore, this study excludes the regression samples of 2008 and 2009 to ensure that the episodic crisis events do not influence the identification of the effects of the FTAs strategy. The regression findings are displayed in columns (1) and (2) of

  Table 3. Robustness test.

  Variables	$\mathit{G}\mathit{V}{\mathit{C}}_{\mathit{i}\mathit{t}}$
  	Exclude External Shocks		Exclude Extreme Values		Sample Selection Bias
  	(1)	(2)	(3)	(4)	(5)	(6)
  $Total\_FT{A}_{it}^f$	6.6258 *** (0.8371)		6.9256 *** (0.8287)		9.7481 *** (1.0885)
  $Core\_FT{A}_{it}^f$		9.0797 *** (1.3145)		9.4255 *** (1.2739)		12.9467 *** (1.2953)
  $\ln \left(ag{e}_{it}\right)$	−0.0543 ** (0.0223)	−0.0529 ** (0.0223)	−0.0359 * (0.0206)	−0.0342 * (0.0205)	−0.0374 (0.0288)	−0.0357 (0.0287)
  $\ln \left(k{l}_{it}\right)$	−0.0449 *** (0.0123)	−0.0452 *** (0.0122)	−0.0426 *** (0.0112)	−0.0429 *** (0.0111)	−0.0427 *** (0.0136)	−0.0435 *** (0.0136)
  $\ln \left(financ{e}_{it}\right)$	0.2371 ** (0.1002)	0.2310 ** (0.1000)	0.2469 *** (0.0941)	0.2404 ** (0.0939)	0.2333 * (0.1202)	0.2339 * (0.1201)
  $\ln \left(Siz{e}_{it}\right)$	0.0310 ** (0.0136)	0.0319 ** (0.0136)	0.0238 * (0.0128)	0.0247 * (0.0129)	0.0169 (0.0152)	0.0185 (0.0152)
  $hh{i}_{it}$	0.1459 * (0.0801)	0.1409 * (0.0801)	0.1582 ** (0.0723)	0.1525 ** (0.0723)	0.2082 ** (0.0909)	0.2007 ** (0.0908)
  Constant	1.2245 *** (0.3705)	1.1779 *** (0.3705)	1.3379 *** (0.3468)	1.2883 *** (0.3468)	1.4287 *** (0.4462)	1.3951 *** (0.4459)
  Enterprise fixed	YES	YES	YES	YES	YES	YES
  Year fixed	YES	YES	YES	YES	YES	YES
  Observations	9198	9198	10,276	10,276	5830	5830
  R-squared	0.653	0.654	0.646	0.647	0.574	0.575

  Note: Values in parentheses are clustering robust standard errors. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

  , and the size of the regression coefficients and the significance levels of the independent variables are mostly consistent with those of the benchmark regression results;
• Exclude extreme values. Special trade patterns, data gathering failures, or unexpected cross-border mergers and acquisitions might cause enterprises’ GVC embeddedness to reach extreme levels. This study deflates the 5% quartile of firms’ GVC embeddedness to reduce the impact of extreme values on estimation results, removing values below the 5% quartile and above the 95% quartile. The regression results are displayed in columns (3) and (4) of Table 3;
• Sample selection bias. Firms with minimal export experience may be more unsure about GVC participation. To avoid estimation bias due to the short-term sample, this article eliminates enterprises that exported for less than ten years between 2000 and 2016. The results in Columns (5) and (6) of Table 3 are consistent with the benchmark regression.

According to the above robustness tests, the regression coefficients for the depth of FTAs are consistently positive and significant, affirming the robustness of the benchmark regression results.

5. Heterogeneity Analysis

5.1. Ownership Structure

To examine whether the FTA strategy has a heterogeneous effect on firms with different ownership structures, this research separated the sample into two groups: state-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs) for testing.

Table 4. Heterogeneity test: ownership structure.

Variables	(1)	(2)	(3)	(4)
	SOEs	Non-SOEs	SOEs	Non-SOEs
$Total\_FT{A}_{it}^f$	9.9380 *** (1.4667)	4.0015 *** (0.8344)
$Core\_FT{A}_{it}^f$			13.0127 *** (1.6812)	5.4955 *** (1.3136)
Constant	0.0309 (0.0333)	−0.0439 (0.0343)	0.0313 (0.0333)	−0.0428 (0.0343)
Controls	YES	YES	YES	YES
Observations	4632	4480	4632	4480
R-squared	0.656	0.716	0.657	0.717

Note: Values in parentheses are clustering robust standard errors. *** denote significance at the 1% levels.

shows the regression results: The effect is stronger for SOEs, although the FTA strategy has a large impact on GVC embeddedness for both types of firms. The reason for this is that SOEs have advantages in capital allocation and policy response. (1) SOEs usually carry out national strategic industrial layout responsibilities. They can accurately address trade and investment policies under FTAs frameworks, particularly in crucial areas such as infrastructure and energy resource cooperation. (2) SOEs have the government’s credit support, combined with the policy support of the FTAs core provisions, making it easier for them to overcome the host country’s institutional barriers, reduce the cost of market access and other aspects of the process, and thus expand the scale of intermediate trade and cross-border production.

5.2. Regional Differences

To find out if the FTA strategy has a heterogeneous effect on firms in various regions, this article divided the sample into three groups: eastern, central, and western enterprises for testing.

Table 5. Heterogeneity test: regional differences.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	Eastern	Central	Western	Eastern	Central	Western
$Total\_FT{A}_{it}^f$	6.2541 *** (0.8920)	10.2130 *** (2.4811)	10.2256 *** (2.8364)
$Core\_FT{A}_{it}^f$				8.4795 *** (1.3690)	14.3090 *** (3.0793)	14.1798 *** (3.4408)
Constant	0.6455 (0.4332)	2.0886 ** (0.9573)	2.8462 *** (0.7743)	0.5903 (0.4334)	2.0693 ** (0.9558)	2.8237 *** (0.7754)
Controls	YES	YES	YES	YES	YES	YES
Observations	7322	1669	1285	7322	1669	1285
R-squared	0.664	0.625	0.602	0.665	0.626	0.603

Note: Values in parentheses are clustering robust standard errors. ** and *** denote significance at the 5% and 1% levels, respectively.

shows the findings of the regional heterogeneity regressions. The FTA strategy can improve GVC embeddedness in all regions, but it shows a greater positive impact on firms in the central and western regions. As an area with a relatively weak export-oriented economy, firms in Central and Western China have long been confined by geography and degree of openness, resulting in minimal involvement in GVCs. As a result, the policy rewards provided by the FTA strategy, including tariff reduction, market access liberalization, and investment facilitation, have a greater marginal effect on breaking down trade barriers and lowering cross-border transaction costs.

5.3. Industrial Differences

To evaluate if the FTA strategy has a heterogeneous effect on enterprises in different industries, this study divided the sample into two groups: manufacturing and non-manufacturing.

Table 6. Heterogeneity test: industrial differences.

Variables	(1)	(2)	(3)	(4)
	Manufacturing	Non-Manufacturing	Manufacturing	Non-Manufacturing
$Total\_FT{A}_{it}^f$	9.9380 *** (1.4667)	4.0015 *** (0.8344)
$Core\_FT{A}_{it}^f$			13.0127 *** (1.6812)	5.4955 *** (1.3136)
Constant	1.6604 *** (0.3879)	0.9702 (0.8138)	1.5842 *** (0.3884)	0.9574 (0.8140)
Controls	YES	YES	YES	YES
Observations	8689	1565	8689	1565
R-squared	0.629	0.728	0.630	0.728

Note: Values in parentheses are clustering robust standard errors. *** denote significance at the 1% levels.

shows the results of the heterogeneity regressions: manufacturing firms’ GVC embeddedness is more promoted by the depth of FTAs than non-manufacturing enterprises. Manufacturing, as the core sector of intermediate goods trade and transnational production, depends on globalization for cross-border purchasing, component processing, and the final assembly of products. Policy instruments within the FTA strategy, including tariff reductions and exemptions, optimization of rules of origin, and the alleviation of technical trade barriers, may reduce transaction costs and enhance the participation in GVCs of manufacturing firms. Conversely, non-manufacturing sectors exhibit less reactivity to FTAs policies owing to the non-tradability of service and variations in the regulatory environments. Furthermore, the value creation in non-manufacturing sectors exhibits less demand for cross-border production, resulting in a lesser effect of the FTA strategy on enhancing their GVC integration.

5.4. Industry Technology Differences

High-technology industries denote technology-intensive sectors characterized by substantial R and D investment, high value-added products, and favorable international market prospects, characterized by intelligence, innovation, strategic planning, and minimal resource consumption. High-technology industries, in statistical terms, denote manufacturing sectors characterized by a significant intensity of R and D investment within the national economy.

To research whether the FTA strategy has a heterogeneous impact on firms with diverse industry technology requirements, the sample was divided into high-tech and low-tech industries for group testing. This article identifies six industries as high-tech, based on the “Classification of High-Tech Industries (Manufacturing) (2017)” from the National Bureau of Statistics and the “Guidelines for the Classification of Listed Companies by Industry (2012 Revision)” from the China Securities Regulatory Commission, as well as other research [61]: Chemical Raw Materials and Chemical Products Manufacturing (C26), Pharmaceutical Manufacturing (C27), Chemical Fiber Manufacturing (C28), Railway, Shipbuilding, Aerospace, and Other Transportation Equipment Manufacturing (C37), Computer, Communication, and Other Electronic Equipment Manufacturing (C39), and Instrument and Meter Manufacturing (C40).

Table 7. Heterogeneity test: industry technology differences.

Variables	(1)	(2)	(3)	(4)
	High-Tech	Low-Tech	High-Tech	Low-Tech
$Total\_FT{A}_{it}^f$	7.0174 *** (1.1367)	6.3231 *** (1.1100)
$Core\_FT{A}_{it}^f$			9.7069 *** (1.3943)	8.6710 *** (1.8079)
Constant	1.3860 *** (0.4795)	1.0124 ** (0.5005)	1.3056 *** (0.4782)	0.9821 ** (0.5006)
Controls	YES	YES	YES	YES
Observations	4074	6169	4074	6169
R-squared	0.634	0.662	0.635	0.662

Note: Values in parentheses are clustering robust standard errors. ** and *** denote significance at the 5%, and 1% levels, respectively.

shows that the positive impact of FTA depth on the embeddedness of listed firms in GVCs in high-tech industries is greater than that in low-technology sectors. Value creation in high-tech industries is highly dependent on cross-border flows of knowledge-intensive factors and trade in high-value intermediate goods, and there is a higher demand for deeper provisions such as intellectual property protection and mutual recognition of technical standards, and the core provisions of FTAs meet their needs.

6. Mechanism Analysis

6.1. Mediating Effects Test

Based on Jiang’s [62] (2022) research, this study uses a two-step strategy to analyze the mediating effect of technical innovation on the GVC embedding of listed firms in China that are impacted by the FTA strategy. First, it empirically examines the impacts of FTA depth on the technical innovation level of Chinese listed firms; second, it theoretically analyzes the impact of listed enterprises’ technological innovation level on their GVC embeddedness. The mediation effect model built on the foundation of the benchmark model is as follows:

$$Paten{t}_{it}={\alpha }_0+{\alpha }_{1}FT{A}_{it}^f+{\alpha }_2{X}_{it}+{\theta }_i+{\theta }_t+{\epsilon }_{it}$$

(7)

where, $Paten{t}_{it}$ represents the index of technological innovation for listed companies in China, calculated by taking the natural logarithm of the sum of patent applications plus one. This index comprises four components: the total number of patent applications ($Patent\_T_{it}$), invention patent applications ($Patent\_I_{it}$), utility model applications ($Patent\_U_{it}$), and design applications ($Patent\_D_{it}$).

Table 8. Mediating effects test.

Variables	$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}\_{\mathit{T}}_{\mathit{i}\mathit{t}}$		$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}\_{\mathit{I}}_{\mathit{i}\mathit{t}}$		$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}\_{\mathit{U}}_{\mathit{i}\mathit{t}}$		$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}\_{\mathit{D}}_{\mathit{i}\mathit{t}}$
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
$Total\_FT{A}_{it}^f$	6.0570 *** (1.7755)		7.2029 *** (1.8551)		7.1861 *** (2.1665)		6.1121 *** (1.8745)
$Core\_FT{A}_{it}^f$		8.7056 *** (2.3455)		9.6580 *** (2.4153)		9.9724 *** (2.7852)		8.4119 *** (2.2609)
$\ln \left(ag{e}_{it}\right)$	0.1768 *** (0.0461)	0.1782 *** (0.0461)	0.0955 ** (0.0479)	0.0978 ** (0.0479)	0.0613 (0.0576)	0.0634 (0.0576)	−0.0131 (0.0467)	−0.0113 (0.0467)
$\ln \left(k{l}_{it}\right)$	−0.1205 *** (0.0278)	−0.1213 *** (0.0277)	−0.1664 *** (0.0281)	−0.1669 *** (0.0281)	−0.1009 *** (0.0318)	−0.1017 *** (0.0318)	−0.1672 *** (0.0271)	−0.1678 *** (0.0271)
$\ln \left(financ{e}_{it}\right)$	0.6167 *** (0.2274)	0.6129 *** (0.2271)	0.7557 *** (0.2346)	0.7503 *** (0.2345)	1.9175 *** (0.2641)	1.9125 *** (0.2638)	0.9572 *** (0.2548)	0.9529 *** (0.2546)
$\ln \left(Siz{e}_{it}\right)$	0.1814 *** (0.0327)	0.1829 *** (0.0326)	0.2421 *** (0.0319)	0.2431 *** (0.0318)	0.2248 *** (0.0375)	0.2261 *** (0.0375)	0.1657 *** (0.0310)	0.1668 *** (0.0309)
$hh{i}_{it}$	−0.4975 ** (0.1944)	−0.5031 *** (0.1943)	−0.1988 (0.1900)	−0.2039 (0.1899)	−0.3030 (0.2317)	−0.3088 (0.2317)	−0.4926 ** (0.1964)	−0.4974 ** (0.1964)
Constant	3.2505 *** (0.8231)	3.2040 *** (0.8222)	2.6078 *** (0.8530)	2.5635 *** (0.8527)	5.3031 *** (0.9668)	5.2536 *** (0.9662)	3.1511 *** (0.8610)	3.1101 *** (0.8612)
N	8182	8182	8182	8182	8182	8182	8182	8182
R-squared	0.891	0.891	0.879	0.879	0.862	0.862	0.878	0.878

Note: Values in parentheses are clustering robust standard errors. ** and *** denote significance at the 5%, and 1% levels, respectively.

presents the regression outcomes of the mediation effect analysis. The regression coefficients for the total depth of FTAs and the core depth of FTAs are both significantly positive, suggesting that the FTA strategy can substantially improve the technological innovation level of listed enterprises. The empirical section has confirmed the impact of the independent variables on the mediating variables. This study analyzes the facilitating impacts of mediating variables on the embeddedness in GVCs of firms through theoretical examination.

According to global value chain governance theory, technology accumulation—a key factor for firms participating in global division of labor—requires overcoming technological barriers and building knowledge advantages. Enterprise technology innovation strengthens its integration in GVCs through three channels: (1) Technological innovation by enterprises enhances product value addition. The breakthrough of the enterprise’s key technology can improve its oversight of critical production processes, thus establishing an irreplaceable advantage [45]. This technological advantage drives firms to transition from low-value-added assembly to high-end activities such as research and development, design, and core input manufacturing consequently achieving the transition from “manufacturing participant” to “technology leader” and enhancing the position of firms within the GVCs. (2) Technological innovation in enterprises enhances the capacity for knowledge chain reconstruction. Technological advancement signifies the capacity of firms to integrate into the global knowledge network, hence amplifying the spillover impact of technology and increasing the bargaining power of firms within the value chain [47]. The provisions for intellectual property rights and the mutual recognition of technical standards in the FTA strategy can offer institutional safeguards for transnational knowledge flows, diminish the costs of technical cooperation, and empower enterprises to transition from being rule-takers to rule-makers within GVCs. (3) Enterprise technical innovation improves market competitiveness. Enterprises with strong technological innovation may break the technology standards barrier through innovative breakthroughs [63]. On the one hand, enterprises can increase profits through technology licensing, patent transfer, and so on; on the other hand, enterprises can participate in international standard setting, convert technological solutions into industry rules, and thus dominate the distribution of value in the division of labor system.

Empirical analyses provide additional support for the cited influencing mechanism. Hu et al. [64], based on an analysis of 16 manufacturing industries, found that green process innovation mediates the relationship between GVC participation and product innovation, supporting the path of technological innovation to promote GVC upgrading. Kang et al. [65] (2001) claim that breakthroughs are critical in the process of technical power impacting the reconfiguration of knowledge networks, emphasizing that technological innovation has a significant impact on GVC position. To summarize, firms’ technological innovation considerably improves their embeddedness in GVCs via three pathways: technological capability upgrading, knowledge chain reconfiguration, and competitive advantage strengthening. This mechanism is consistent with Gereffi’s [63] (2015) transmission path of “institutional quality-technological innovation-value chain status”. It provides theoretical support for the FTA strategy to improve enterprise embeddedness in GVCs through the channel of technological innovation.

6.2. Moderating Effects Test

The FTA strategy significantly improves enterprises’ integration into GVCs by enhancing their technical innovation skills; nevertheless, the extent to which firms can efficiently utilize this institutional advantage is also limited by their governance capacities and operational efficiency. This paper presents two significant moderating variables—internal control costs (management expense ratio, $Mgt\_Fe{e}_{it}$) and the operational efficiency of firms’ assets (inventory turnover ratio, $Stock\_Tur{n}_{it}$)—and incorporates their interaction terms with the explanatory variables in the benchmark model to evaluate their moderating effects:

$$GV{C}_{it}={\alpha }_0+{\alpha }_{1}FT{A}_{it}^f+{\alpha }_{2}FT{A}_{it}^f\times Mgt\_Fe{e}_{it}+{\alpha }_3{X}_{it}+{\theta }_i+{\theta }_t+{\epsilon }_{it}$$

(8)

$$GV{C}_{it}={\alpha }_0+{\alpha }_{1}FT{A}_{it}^f+{\alpha }_{2}FT{A}_{it}^f\times Stock\_Tur{n}_{it}+{\alpha }_3{X}_{it}+{\theta }_i+{\theta }_t+{\epsilon }_{it}$$

(9)

6.2.1. Internal Control Costs

Columns (1) and (2) of

Table 9. Moderating effects test.

Variables	$\mathit{G}\mathit{V}{\mathit{C}}_{\mathit{i}\mathit{t}}$
	(1)	(2)	(3)	(4)
$Total\_FT{A}_{it}^f\times Mgt\_Fe{e}_{it}$	−9.6891 ** (4.1294)
$Core\_FT{A}_{it}^f\times Mgt\_Fe{e}_{it}$		−12.1308 ** (5.2862)
$Total\_FT{A}_{it}^f\times Stock\_Tur{n}_{it}$			−0.0296 *** (0.0067)
$Core\_FT{A}_{it}^f\times Stock\_Tur{n}_{it}$				−0.0406 *** (0.0065)
$Total\_FT{A}_{it}^f$	8.1657 *** (0.8806)		7.7328 *** (0.7428)
$Core\_FT{A}_{it}^f$		11.0389 *** (1.0986)		11.0116 *** (0.9434)
$Mgt\_Fe{e}_{it}$	−0.0011 (0.0018)	−0.0005 (0.0021)
$Stock\_Tur{n}_{it}$			0.0003 *** (0.0001)	0.0003 *** (0.0000)
Constant	1.3825 *** (0.3532)	1.3408 *** (0.3539)	1.1892 *** (0.3625)	1.1409 *** (0.3628)
Controls	YES	YES	YES	YES
Observations	10,276	10,276	10,265	10,265
R-squared	0.646	0.647	0.646	0.647

Note: Values in parentheses are clustering robust standard errors. ** and *** denote significance at the 5%, and 1% levels, respectively.

show the moderating impact of the management expense ratio ($Mgt\_Fe{e}_{it}$): the coefficients of the interaction terms are significantly negative, suggesting that a reduction in internal control costs enhances the FTA strategy’s capacity to foster the embeddedness in GVCs of listed firms in China. The findings signify that the internal governance ability of firms serves as an “amplifier” of institutional dividends. It significantly moderates the process of translating FTA policies into corporate behaviors, verifying hypothesis 3. Redundant organizational structures and decision-making processes will hinder firms’ responsiveness to foreign market opportunities and diminish their capacity for rapid adjustments to FTA provisions and conditions. The “organizational relaxation theory” posits that excessive overhead costs diminish resources allocated for critical activities like technological development and capacity optimization, thereby undermining the beneficial impacts of the FTA strategy on the firm’s integration in GVCs [48].

6.2.2. Operational Efficiency of Assets

Columns (3) and (4) of Table 9 show the moderating effect of the inventory turnover ratio (B): the regression coefficient for the inventory turnover ratio is significantly positive, signifying its beneficial impact on the embeddedness in GVCs of listed firms; conversely, the coefficients of the interaction term between the inventory turnover ratio and the depth of FTAs are significantly negative, further substantiating the existence of a substitution effect between the two and verifying hypothesis 4. For enterprises with high asset operation efficiency, inherent advantages in trade costs and customs clearance diminish the marginal returns of FTA benefits (e.g., tariff reductions). This finding aligns with the key concepts of dynamic capability theory: the higher the operational efficiency of a firm’s assets, the more its GVC embeddedness depends on the effective functioning of its internal system rather than on external institutional advantages. When firms achieve effective resource allocation via inventory turnover, the positive impact of FTA depth on their GVC embeddedness is largely replaced, indicating a substitution effect between firms’ intrinsic capabilities and external regulatory mechanisms.

Table 10. Summary of mechanism tests.

Variables	Measurement Methods	Mechanisms	Specific Impact
Technical Innovation Level	The natural logarithm of the sum of patent applications plus one	Mediating effect	The FTA strategy could improve the embeddedness of Chinese listed enterprises in GVCs by promoting their technological innovation.
Internal Control Costs	Management expense ratio of firms	Moderating effect	Their internal control costs negatively moderate the positive impact of the FTA strategy on firms’ GVC embeddedness.
Operational Efficiency of Assets	Inventory turnover ratio of firms	Moderating effect	Higher asset turnover reduces firms’ reliance on FTA benefits, thereby diminishing the strategy’s positive effect on improving GVC embeddedness.

summarizes the main components of the mechanism analysis.

7. Conclusions and Discussion

This article has empirically explored the impact of the FTA strategy on the embedding of listed firms in China in global value chains, the heterogeneous effects, and the mechanism. Its conclusions are as follows: (1) The FTA strategy has a significant promoting effect on the GVC embedding of listed enterprises in China, and the wider the coverage of FTA provisions, the stronger the promoting effect. (2) The effect of the FTA strategy on GVC embeddedness varies by company ownership structure, geographical distribution, industry type, and technical characteristics. The positive impact of the FTA strategy is most noticeable in the GVC embeddedness of state-owned businesses (SOEs), enterprises in central and western regions, manufacturing enterprises, and high-tech firms. (3) The FTA strategy indirectly boosts the GVC embeddedness of Chinese listed firms in China by increasing their technical innovation level, allowing them to increase their competitive advantages and break free from low-end lock-in. (4) Lower internal control costs in enterprises can positively regulate the impact of the FTA strategy on their GVC embeddedness. In contrast, there is a substitution effect between asset operation efficiency and the depth of FTAs, and enterprises with higher asset operation efficiency have a lower marginal demand for political dividends.

The findings of this study provide the Chinese government with references to optimize its FTA strategy and formulate targeted policies to enhance the global competitiveness of enterprises. The conclusions also offer businesses in developing countries insights on how to sustainably integrate into global value chains. The recommendations to the government are as follows: (1) Optimize the strategic arrangement of the FTA. China should prioritize promoting FTA negotiations with developed countries and critical areas, with an emphasis on high-standard provisions such as intellectual property protection and mutual acceptance of technical standards, as well as breaking down institutional barriers in high-value chains. (2) Design tailored support policies for various firms. State-owned enterprises play a dominating role in cross-border infrastructure investment and industry chain integration, and governments should encourage them to drive non-SOEs into global value chains. The government may consider building a policy-sharing platform for East, Central, and West China to assist the transfer of value chain governance experience from coastal areas to central and western areas. The government should increase R and D subsidies for manufacturing and high-tech firms and support enterprise high-end transformation through the technological spillover impact of FTAs. (3) Internal technical innovation should be prioritized as well as improving regional policy synergy externally. On the one hand, the government should depend on the FTA’s core provisions to encourage the flow and efficient distribution of innovative factors inside the free-trade area. On the other hand, the government should encourage enterprises to build technical alliances with overseas R and D institutes and other organizations and engage in the development of international standards and mutual recognition systems. (4) The government should encourage firms with a high inventory turnover to use FTA market access rules to expand into new international markets.

Targeted recommendations for different enterprises are provided below: (1) State-owned enterprises should actively use their platform advantages to strengthen their voice in international trade and participate in the development of international rules; non-state-owned enterprises should focus more on international standard certification requirements and key trade agreement policy directions. (2) Enterprises in Central and Western China can speed up the development of local and international logistics hubs and take advantage of the tariff advantages of FTAs to lower trade costs and expand international markets. (3) Manufacturing and high-tech firms may boost patent conversion efficiency in order to remain competitive in global value chains. (4) Enterprises should reduce internal control costs, increase asset operation efficiency, and maximize the use and transformation of FTA policy dividends.

With the development of globalization, increasing enterprise participation in global value chains has become increasingly important. This study’s sample consists mainly of Chinese listed enterprises, which limits its generalizability. Future study could broaden the sample to include unlisted trading firms. Additionally, this article focuses only on the total depth and core depth of the FTAs, without precisely examining the role of any specific provision or chapter within them. Future studies could examine which provisions in free trade agreements are essential in promoting firm integration into GVCs and their potential impact on national competitiveness.