Forward Participation in GVCs and Its Impact on Export Quality of Forestry Products: Evidence from China

Abstract

Improving the export quality of forestry products is crucial for the development of the forestry industry in developing countries. This study focuses on China—the largest developing country and a leading exporter of forestry products—and explores how GVCs influence the export quality of their forestry products, taking into account the unique characteristics of the forestry industry as a resource-based sector. Using China’s trade data from 41 partner countries between 2000 and 2014, the study finds that forward GVC participation significantly enhances the export quality of forestry products, distinguishing this study from the previous research. The diffusion of technology within GVCs is identified as a key influencing factor. This result highlights that, as a resource-based industry, the value chains in the forestry industry are typically buyer-driven, with lead firms in these chains, which possess advanced technologies, often occupying downstream positions related to marketing and brand building. Therefore, when developing countries leverage GVCs to improve the export quality of forestry products, they should not only focus on backward GVC participation but also consider the potential for technology diffusion generated through forward GVC participation. This distinction from other production-driven value chains is one of the key findings of this research.

1. Introduction

Improving the export quality of forestry products is crucial for sustainable development in developing economies. Higher export quality can boost trade revenue, create jobs, and drive industry growth by pushing firms to innovate and improve production [1]. The forestry industry, as a sector with significant emissions, accounts for approximately 10% of carbon emissions in developing regions [2]. With rising global demand for low-carbon solutions—driven by population growth, better living standards, and eco-awareness [3]—focusing on quality can promote a shift toward greener practices, reducing energy use and pollution.

China, as the world’s largest developing country and a leading exporter of forestry products, faces an urgent need to enhance the quality of its forestry exports. The country’s forestry sector has achieved several remarkable milestones: By 2018, China’s wood processing output surpassed that of the U.S. by 1.3 times, while its papermaking output exceeded the U.S. by 1.21 times [3]. By 2020, the total forestry output reached USD 1159.44 billion, with trade surpassing USD 160 billion [4]. Yet, China’s forestry export quality still needs improvement; the quality index for their wooden furniture exports, estimated using a regression method based on Hallak and Sivadasan (2013) [5] and CEPII BACI trade data (1998–2017), ranges between 0.2 and 0.25, lower than that of paper products (above 0.35) and wood-based panels [6]. This index, standardized between 0 and 1, reflects consumer-perceived quality relative to price and quantity. It has declined slightly in the Belt and Road Initiative (BRI) markets, which include countries such as Malaysia, Singapore, the United Arab Emirates, and Vietnam in Asia, the Middle East, and Eastern Europe, since 2012 [6], partly due to unequal value distribution in global value chains (GVCs) [7]. In comparison, major European exporters like Poland and Italy continue to achieve higher export quality in wooden furniture, with revealed comparative advantage (RCA) indices consistently above 2.5, indicating stronger competitiveness [6]. Developing countries like Vietnam have also exhibited rising export quality and competitiveness, surpassing China in certain sub-categories [6].

To promote higher value-added exports, China has enacted strict policies limiting raw timber exports. Since the launch of the Natural Forest Protection Program (NFPP) in 1998—subsequently reinforced in 2015 and 2023—the commercial logging of natural forests has been banned, and raw log exports prohibited, in a bid to encourage the development and export of high-value processed exports [8,9]. The 2019 Forestry Law further distinguished ecological and commercial forests, promoting fast-growing plantations for processing industries [10]. This policy aligns with strategies in other countries, such as Indonesia’s 2001 raw log export ban to foster local processing [9], Sweden’s restrictions prioritizing FSC-certified pulp and furniture exports [11], and Russia’s high export tariffs (25%–80%) to encourage sawnwood exports [8]. These policies reflect a global trend toward enhancing downstream value chain participation to capture greater value and improve export quality.

Global value chains (GVCs)—defined as “production organized in sequential processing stages across different countries” [12,13]—are a key feature of China’s forestry product trade. China’s forestry sector heavily relies on imported raw materials (e.g., timber, pulp) for processing into finished goods (e.g., furniture, paper), which are then exported [3]. Despite high domestic value-added rates (85% in wood processing, 82% in papermaking by 2018), China’s forestry sector remains reliant on GVCs for raw material supplies [3,8], reflecting a structural dependency. Therefore, GVCs play a critical role in improving the quality of China’s forestry exports.

China’s forestry trade highlights its pivotal role within GVCs. In 2018, over 50% of China’s timber supply was imported, primarily from Russia, New Zealand, the United States, and Indonesia for logs and sawnwood, and Brazil for pulp [8,9]. Exports, predominantly wooden furniture (40%–50%), plywood (20%–25%), and paper products (15%–20%), were directed to major markets including the United States, Japan, and the European Union [8]. Unlike other developing countries’ lower-value exports [9], China places an emphasis on the export of high-value processed products. These trade dynamics underscore China’s strategic integration into GVCs, enhancing its potential for export quality upgrading.

However, China’s disadvantaged position within forestry GVCs may be a key constraint on the upgrading of its export quality. The country predominantly occupies lower-value niches, where small and medium enterprises (SMEs) operate under “captive” governance—foreign buyers dictate the pricing and standards, squeezing local firms’ profit margins [7,14]. Despite these challenges, GVCs have proven effective in enhancing China’s manufacturing exports [15], suggesting their potential role to improve forestry product quality remains a critical yet underexplored avenue. This study therefore examines whether and how strategic GVC participation could help overcome these limitations to enhance the quality of China’s forestry exports.

The structure of forestry GVCs differs fundamentally from that of manufacturing GVCs, which may shape distinct pathways for the upgrading of export quality. Manufacturing GVCs are typically “producer-driven” chains controlled by upstream technology holders [16]; however, forestry GVCs follow a buyer-driven model where downstream actors (e.g., retailers and brands) dictate product standards [17,18].

This structural contrast generates distinct participation patterns. The first is backward participation, where semi-finished goods are imported for processing and then re-exported, as seen in China’s timber imports for furniture production. The second is forward participation, where exported materials undergo foreign manufacturing and re-export, as exemplified by the use of Chinese plywood in third-country furniture [12,13]. The empirical evidence suggests that forward participation holds greater quality-upgrading potential, as demonstrated through China’s emission reduction achievements when supplying advanced economies [19] and enhanced sustainability certification rates [20], outcomes that are particularly relevant for standardizing quality in the BRI markets where Chinese furniture exports currently exhibit significant variability [6].

This research addresses a critical gap by examining whether forward participation in GVCs can likewise improve the quality of China’s forestry exports. While the existing studies focus overwhelmingly on backward participation in high-tech GVCs [21], this study examines forward participation in resource-based industries such as forestry. In the context of China, it provides insights for developing countries to improve forestry export quality, thereby supporting export transformation, green development, and low-carbon goals.

The study addresses the following two key questions: (1) Does forward participation in GVCs improve the export quality of China’s forestry products? (2) What is the underlying mechanism, and how does the trading partner’s position in the value chain influence this relationship? The analysis examines China’s bilateral trade data with 41 partners (2000–2014) from a GVC perspective.

2. Literature Review

Export quality is a key focus in international trade research, including in the forestry sector. Porter’s seminal value chain framework [22], further elaborated in his broader competitive strategy work [23], highlights firm-level activities—such as processing, design, and marketing—that enhance competitive advantage and product quality, providing a foundation for understanding export quality dynamics in forest product industries. According to Schott and Khandelwal, export quality reflects a product’s intrinsic attributes and consumer preferences [1,24]. In the forestry sector, export quality often includes durability, processing precision, environmental sustainability, and compliance with international standards such as forest certification [4,19].

While the existing studies have extensively explored factors influencing export quality, such as the share of domestic value added, sustainability standards, and economic development [25,26,27,28,29], few have directly examined the relationship between GVCs and export quality in the forestry sector. Porter’s value chain framework, extended to GVCs as formalized by Gereffi, suggests that participation in international production networks enhances quality through firm-level activities like processing and marketing [22,30]. Leontief’s input–output economics provides tools to measure value-added trade flows in GVCs, while Benckler’s network-based production theory highlights how collaborative networks, such as those enforcing FSC certification, facilitate quality standardization [31,32]. However, the role of GVCs in promoting export quality has been primarily explored in manufacturing, with limited application in agriculture and forestry. For instance, in the wooden furniture industry, a key downstream segment of forestry, competitiveness and quality upgrading in GVCs remain underexplored, particularly for developing countries [6,33].

The existing research suggests that GVC participation can improve export quality at both the macro and sectoral levels. At the micro level, firms engaged in GVCs can access higher-quality intermediate inputs at lower costs and benefit from “learning by doing” [34,35], which helps improve export quality [36,37]. At the macro level, cross-national studies have indicated that GVC participation positively impacts a country’s export quality [38,39]. However, evidence remains scarce regarding whether these mechanisms apply to resource-based sectors like forestry, where small producers often face governance constraints (e.g., “captive” relationships) that limit value capture [7,40].

However, evidence on whether forward participation in GVCs enables developing countries to improve export quality remains limited. Cross-country analyses reveal that backward participation enhances export quality in both developed and developing nations, while forward participation’s quality-enhancing effect is only observed in developed economies [38,39]. At the micro level, methodological constraints and a focus on the “imports drive exports” logic have led most studies to examine backward participation rather than forward participation [41,42].

For resource-based industries like forestry, forward GVC participation may represent a critical yet understudied pathway to quality upgrading. Unlike manufacturing where technology diffuses through backward participation [43], forestry GVCs are buyer-driven, with downstream actors (e.g., retailers, certifiers) controlling the advanced knowledge and sustainability standards [17,20]. This structural feature suggests that forward participation—exporting materials for further processing in high-standard markets—could enable developing countries to absorb green technologies and quality benchmarks, as evidenced by emission reductions when Chinese semi-processed goods enter advanced economies’ production [19]. Yet, the existing studies cannot confirm whether this mechanism actually improves export quality, creating the research gap that this study addresses.

Building on these insights, this study makes two key contributions. First, it extends the literature on GVCs and their impact on export quality by examining the forestry sector in a developing economy, demonstrating that forward GVC participation can improve export quality. Second, it empirically confirms this effect using China’s forestry export data.

3. Theoretical Analysis

The forestry value chain has unique features that shape how knowledge and technology diffuse, potentially impacting export quality. Porter’s value chain framework [22], further elaborated in his competitive strategy work [23], posits that a competitive advantage arises from optimizing firm-level activities, such as processing and marketing, which are critical for quality upgrading in forestry products like wooden furniture. This framework has informed GVCs, which emerged in the 1990s to describe production coordinated across countries, often in buyer-driven chains led by downstream firms in advanced economies [17,30]. Benckler’s concept of network-based production suggests that collaborative networks within GVCs, such as those enforcing sustainability standards like FSC certification, enable technology diffusion and quality improvements [29].

Building on these theoretical mechanisms, it can be argued that forward participation in GVCs facilitates knowledge and technology diffusion in China’s forestry sector, significantly enhancing export quality. Firstly, firms within the forestry value chain often establish long-term cooperative relationships with downstream firms in developed countries, which are equipped with advanced technology, market insights, and sustainability standards and are typically responsible for managing and coordinating the chain [16,44]. For example, in the wooden furniture sector—a major downstream area of forestry—big retailers in advanced markets often set the product and quality benchmarks [7]. Additionally, since forestry items like furniture and paper target consumers, buyers value traits like durability, quality, and eco-friendliness, including compliance with standards like the Forest Stewardship Council (FSC) [6,20]. Moreover, assessing forestry product quality with uniform measures is difficult, unlike manufactured goods, due to raw material variations (e.g., grain, density) and complex processing [18,45].

Based on these traits and the prior studies on technology diffusion in GVCs, forward GVC participation could enable firms in developing nations to gain advanced technologies and improve forestry export quality through the following routes:

First, downstream companies in advanced countries might share advanced technologies with upstream firms in developing nations. To ensure high-quality, durable, and eco-friendly forestry outputs, these firms closely monitor upstream raw inputs like logs and pulp [44,46]. Since directly assessing forestry quality is complex, downstream players often guide upstream production to meet global standards [18,47]. Long-term collaborations also prompt downstream firms to help upstream partners maintain steady, top-grade supplies [48,49]. For example, in wooden furniture, retailers in developed markets might aid Chinese suppliers with eco-labeling compliance, spurring technology diffusion [6]. Thus, downstream leaders are motivated to share innovative insights, driving quality improvement.

Second, foreign direct investments (FDIs) by multinational corporations (MNCs) in forestry GVCs facilitate the diffusion of technology to developing countries. Given the high transportation costs of timber, downstream MNCs frequently establish local production bases [50,51], enabling domestic firms to adopt advanced operational practices [52], improve input standards, and benefit from skilled labor mobility [46,53]. For instance, when global furniture manufacturers operate in China, the technologies (e.g., precision woodworking) and sustainability protocols they introduce often spill over to local exporters, elevating product quality [20]. Such knowledge diffusion may explain the growth in the quality of China’s furniture exports to certain markets [6]. These findings suggest that forward GVC participation through MNC linkages can help bridge quality gaps for developing countries, which is particularly relevant for China’s forestry sector where export standards still trail several global leaders [6].

Drawing from these insights, the following hypotheses are proposed:

Hypothesis 1: Forward participation in GVCs can improve the quality of forestry exports in developing countries.

Hypothesis 2: The diffusion of technology from leading downstream firms in advanced countries is a key pathway through which forward participation in GVCs enhances export quality.

4. Materials and Methods

4.1. Main Regression Model

To explore the effect of forward participation in GVCs on the quality of China’s forestry exports, the following statistical model is formulated:

$$\begin{array}{cc}\hfill \mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}{\mathrm{y}}_{\mathrm{i}\mathrm{c}\mathrm{t}}& ={\beta }_0+{\beta }_1\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_2\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{B}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}\hfill \\ & + {\beta }_3\mathrm{u}\mathrm{p}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_4\mathrm{d}\mathrm{o}\mathrm{w}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_5{\mathrm{X}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\mathrm{v}}_{\mathrm{i}}+{\mathrm{v}}_{{\mathrm{c}}^{\prime }}+{\mathrm{v}}_{{\mathrm{t}}^{\prime }}+{\epsilon }_{\mathrm{i}\mathrm{c}\mathrm{t}}\hfill \end{array}$$

(1)

where $i$, $c$, and $t$ denote the forestry industry (e.g., furniture, paper), partner nation, and time period, respectively. The outcome variable, $\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}{\mathrm{y}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$, measures the export quality of China’s forestry products in industry $i$ to nation $c$ at time $t$, with further explanation in Section 4.3.

Table 1. Variables in the main regression model.

Variable	Definition	Type
${\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$	Export quality of China’s forestry products in industry $\mathrm{i}$ to country $\mathrm{c}$ at time $\mathrm{t}$	Dependent variable
${\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$	Forward participation in GVCs: share of China’s exports used as intermediate goods in country c’s production for further export	Core explanatory variable
${\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{B}\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$	Backward participation in GVCs: share of imported intermediate inputs in China’s exports	Explanatory variable
${\mathrm{U}\mathrm{p}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$	Trading partner’s upstreamness: distance from final demand in country $\mathrm{c}$’s production	Explanatory variable
${\mathrm{D}\mathrm{o}\mathrm{w}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$	Trading partner’s downstreamness: distance from primary inputs in country $\mathrm{c}$’s production	Explanatory variable
${\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{c}\mathrm{t}}$	GDP of trading partner country $\mathrm{c}$ at time $\mathrm{t}$ (market size)	Control variable
${\mathrm{G}\mathrm{D}\mathrm{P}\mathrm{c}\mathrm{a}\mathrm{p}}_{\mathrm{c}\mathrm{t}}$	Per capita GDP of trading partner country $\mathrm{c}$ at time $\mathrm{t}$ (consumption level)	Control variable
${\mathrm{F}\mathrm{T}\mathrm{A}}_{\mathrm{c}\mathrm{t}}$	Dummy variable: 1 if China has an FTA with country $\mathrm{c}$ at time $\mathrm{t}$, 0 otherwise	Control variable
${\mathrm{R}\mathrm{L}}_{\mathrm{c}\mathrm{t}}$	Rule of law index of country $\mathrm{c}$ at time $\mathrm{t}$ (legal enforcement)	Control variable
${\mathrm{R}\mathrm{Q}}_{\mathrm{c}\mathrm{t}}$	Regulatory quality index of country $\mathrm{c}$ at time $\mathrm{t}$ (quality regulation)	Control variable
${\mathrm{v}}_{\mathrm{i}}$	Industry fixed effects	Fixed effect
${\mathrm{v}}_{\mathrm{c}}$	Country fixed effects	Fixed effect
${\mathrm{v}}_{\mathrm{t}}$	Time fixed effects	Fixed effect
${\mathsf{\epsilon }}_{\mathrm{i}\mathrm{c}\mathrm{t}}$	Error term	Error term

summarizes the variables in the model, including their definitions, types, and data sources.

The outcome variable, quality_ict, and the core explanatory variable, GVCFP_ict, are detailed in Section 4.2 and Section 4.3, respectively. Other GVC-related variables (GVCBP_ict, upstream_ict, downstream_ict) and control variables (X_ict) account for the GVC structure and external factors influencing export quality [3,53,54]. Although geographical distance is an important factor affecting export quality, its effect is absorbed by the country fixed effects (${\mathrm{v}}_{\mathrm{c}}$) in the model [55].

To control for unobserved heterogeneity, the model includes industry fixed effects (${\mathrm{v}}_{\mathrm{i}}$), country fixed effects (${\mathrm{v}}_{\mathrm{c}}$), and time fixed effects (${\mathrm{v}}_{\mathrm{t}}$), with ${\epsilon }_{\mathrm{i}\mathrm{c}\mathrm{t}}$ as the error term. The regression is estimated using ordinary least squares (OLS) with fixed effects. The coefficient of interest, ${\beta }_1$, captures the effect of forward participation on export quality, testing Hypothesis 1, which posits that forward participation enhances export quality by facilitating technology diffusion and sustainable practices. The inclusion of $\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{B}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$, upstreamict, and downstreamict allows us to examine the moderating role of the trading partner’s position in the value chain, as proposed in Hypothesis 2.

4.2. Main Explanatory Variable

The core explanatory variable, $\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$, measures China’s forward participation in global value chains within the forestry sector, reflecting the share of China’s exports integrated into the production processes of trading partner countries for further export. This variable is built using the inter-country input–output (ICIO) framework, which splits trade into value-added segments. The forward GVC participation amount ($EX{P}_{GVCF,ict}$) is the portion of China’s exports to country $c$ in sector $i$ that is used in $c$’s production for re-export, showing domestic value added in GVC trade. The participation share, $\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$, is this amount divided by the total exports to $c$ in $i$ at $t$, expressed as follows [56]:

$$\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}={\displaystyle \frac{EX{P}_{GVCF,ict}}{EX{P}_{ict}}}$$

(2)

A higher $\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ indicates greater forward participation, possibly allowing Chinese forestry firms to adopt advanced technologies and green practices like eco-labeling, while improving quality, especially in BRI downstream markets [6,56].

Other GVC variables are also drawn from the ICIO framework, with $\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{B}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ reflecting the foreign value added in exports, showing imported input use in sector $i$ to country $c$ at $t$ [41]. The variables upstreamict and downstreamict are calculated via the Leontief inverse matrix, assessing the mean number of production stages separating the partner country’s output from the final demand (upstreamness) or primary inputs (downstreamness) [54].

4.3. Measurement for Export Quality

Export quality is often measured by unit price, under the premise that higher prices reflect higher quality [24,55]. However, unit prices may also capture costs, exchange rates, and other factors, leading many studies to combine unit prices with indicators like market share, trade surplus, or export volume [1,27,57,58]. This study adopts the KSW method [1], which jointly determines export quality using unit price and export volume, leading to the following econometric models:

$$\mathrm{l}\mathrm{n}{\mathrm{Q}}_{\mathrm{h}\mathrm{c}\mathrm{t}}+\sigma \mathrm{l}\mathrm{n}{\mathrm{P}}_{\mathrm{h}\mathrm{c}\mathrm{t}}={\phi }_{\mathrm{c}\mathrm{t}}+{\phi }_{\mathrm{h}}+{\epsilon }_{\mathrm{h}\mathrm{c}\mathrm{t}}$$

(3)

$${\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}={\displaystyle \frac{{\epsilon }_{\mathrm{h}\mathrm{c}\mathrm{t}}}{\sigma -1}}$$

(4)

In Model (3), the subscripts $h$, $c$, and $t$ represent the product variety (HS 6-digit level), trading partner country, and time, respectively. The variables ${\mathrm{Q}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$, ${\mathrm{P}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$, and $\sigma$ denote export volume, unit price, and the substitution elasticity of products, respectively. The right-hand side includes the country-time fixed effect (${\phi }_{\mathrm{c}\mathrm{t}}$), product fixed effect (${\phi }_{\mathrm{h}}$), and error term (${\epsilon }_{\mathrm{h}\mathrm{c}\mathrm{t}}$). Following the existing literature, we set the substitution elasticity $\sigma$ to 3 for forestry products, and use $\sigma$ = 5 for robustness checks [1]. After estimating Model (3), the export quality at the product level is obtained using Model (4).

To standardize the product-level quality, we apply the following standardization [30]:

$${\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}={\displaystyle \frac{{\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}-min\cdot qualit{y}_{hct}}{max\cdot qualit{y}_{hct}-min\cdot qualit{y}_{hct}}}$$

(5)

where max ${\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$ and min ${\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$ are the maximum and minimum quality values of product $h$ across all countries and years. After standardization, ${\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$ ranges between 0 and 1. The industry-level export quality is then aggregated as follows:

$${\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{c}\mathrm{t}}={\displaystyle \frac{{\mathrm{T}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e}}_{\mathrm{h}\mathrm{c}\mathrm{t}}}{\sum _{\mathrm{h}} {\mathrm{T}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e}}_{\mathrm{h}\mathrm{c}\mathrm{t}}}}{\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$$

(6)

In this equation, ${\mathrm{T}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e}}_{\mathrm{h}\mathrm{c}\mathrm{t}}$ is the trade value of product $h$ with trading partner country $c$ at time $t$. The industry-level quality, ${\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{\mathrm{c}\mathrm{t}}$ is the weighted sum of standardized product-level quality, with trade shares as weights, and is used as the dependent variable in Model (1).

4.4. Research Methods of Impact Channels

To explore the channels through which forward participation in GVCs affects export quality, we introduce a moderating variable representing the technological level of trading partners [59,60]. Model (7) incorporates an interaction term between the moderating variable—trading partners’ labor productivity ($\mathrm{P}{\mathrm{L}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$)—and forward GVC participation ($\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$).

$$\begin{array}{cc}\hfill \mathrm{q}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}{\mathrm{y}}_{\mathrm{i}\mathrm{c}\mathrm{t}}& ={\beta }_0+{\beta }_1\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_2\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}\times \mathrm{P}{\mathrm{L}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_3\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{B}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}\hfill \\ & + {\beta }_4\mathrm{u}\mathrm{p}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_5\mathrm{d}\mathrm{o}\mathrm{w}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\beta }_6{\mathrm{X}}_{\mathrm{i}\mathrm{c}\mathrm{t}}+{\mathrm{v}}_{\mathrm{i}}+{\mathrm{v}}_{\mathrm{c}}+{\mathrm{v}}_{\mathrm{t}}+{\epsilon }_{\mathrm{i}\mathrm{c}\mathrm{t}}\hfill \end{array}$$

(7)

In this model, $\mathrm{P}{\mathrm{L}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ represents the labor productivity of the trading partner country c in industry $i$ at time $t$, calculated as the total output value of a sector divided by the number of workers in that sector, reflecting the industry’s technological level. A significant and positive coefficient on the interaction term (${\beta }_2$), alongside a significant coefficient on $\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ (${\beta }_1$), would indicate that higher technological levels in trading partner countries amplify the positive impact of forward GVC participation on export quality.

4.5. Data

The sample includes China’s panel data across the following three aspects: industry, trading partner, and year. To capture the entire value chain, we select industries spanning upstream activities (e.g., forestry and logging) and downstream activities (e.g., wood and paper product manufacturing), forming a complete industrial chain. The selected industries are based on the International Standard Industrial Classification (ISIC Rev. 4) and include Industry 2 (forestry and logging), Industry 7 (manufacture of wood and products of wood and cork, articles of straw and plaiting materials), and Industry 8 (manufacture of paper and paper products). For trading partner countries, we include China’s major forest product trading partners, as well as countries of varying sizes and development levels, to avoid sample selection bias. The time range covers 2000 to 2014, a period following China’s accession to the WTO, marked by the rapid integration into GVCs and the significant expansion of forestry trade. This period provides a stable and comprehensive dataset for analyzing GVC dynamics, as more recent WIOD data were not available at the time of this study.

The data are sourced from the 2016 edition of the World Input–Output Database (WIOD), which provides input–output data for 56 industries across 43 economies, including China and the “rest of the world” from 2000 to 2014. In comparison to the alternative input–output databases (e.g., OECD TiVA), the WIOD provides a more precise categorization of timber-related industries, rendering it appropriate for this research.

Although the WIOD database extends only to 2014, this timeframe encompasses a pivotal stage of China’s integration into GVCs and the expansion of its forestry trade, offering a solid basis for our analysis. The WIOD database is extensively utilized in GVC research owing to its detailed industry delineation, especially for forestry-related sectors, which are less thoroughly represented in more recent databases such as OECD TiVA. Subsequent studies might investigate more current data, such as those available in the OECD TiVA database (covering up to 2020), to determine whether these trends continue beyond 2014.

The data on China’s forward participation in GVCs ($\mathrm{G}\mathrm{V}\mathrm{C}\mathrm{F}{\mathrm{P}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$) are derived from the 2016 WIOD database. Using the world input–output tables, we calculate the trade value of China’s forward participation in GVCs for Industries 2, 7, and 8 across 41 trading partner countries from 2000 to 2014, following the method described in Section 4.2.

Export quality data are obtained from the United Nations Trade Database (UN Comtrade). First, we identify all HS 1996 6-digit codes corresponding to Industries 2, 7, and 8 in ISIC Rev. 4 using the HS-to-ISIC correspondence table provided by the OECD (https://stats.oecd.org/wbos/fileview2.aspx?IDFile=2bddcb44-5e74-49a0-8ac9-80ed46a2274c, accessed on 29 April 2025). Second, we extract the export value and volume of these HS 6-digit products from China to 41 trading partners from 2000 to 2014. Third, product-level export quality is calculated using the method described in Section 4.3. Finally, the product-level quality data are aggregated to the industry level using the trade-share-weighted method outlined in Section 4.3, resulting in export quality data for Industries 2, 7, and 8, which align with the GVC data from WIOD.

The moderating variable, labor productivity ($\mathrm{P}{\mathrm{L}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$), is calculated using the 2016 edition of the Socio-Economic Accounts from WIOD. We extract the nominal total output, apply the output discount rate to obtain the real output, and divide by the number of workers in each industry to compute labor productivity in the currency of each country. Exchange rate data from the CEPII database are used to convert these values into U.S. dollars, yielding labor productivity for 41 trading partners from 2000 to 2014. Since these data and the GVC-related data are sourced from the same database, no additional matching is required.

Control variables, including the GDP and per capita GDP of the trading partner countries, are sourced from the CEPII database. The FTA dummy variable is obtained from the WTO database, while the rule of law (${\mathrm{R}\mathrm{L}}_{\mathrm{c}\mathrm{t}}$) and regulatory quality (${\mathrm{R}\mathrm{Q}}_{\mathrm{c}\mathrm{t}}$) indices are derived from the Worldwide Governance Indicators (WGI). After removing the samples with missing data, the final dataset includes three industries in the upstream and downstream segments of the forestry industry chain, 41 trading partner countries, and 15 years, totaling 1845 observations.

Table 2. Descriptive statistics of key variables.

Variable	Description	Mean	SD	Min	Median	Max
quality ($\sigma$ = 3)	Export quality (KSW, $\sigma$ = 3)	0.029	0.037	0.000	0.021	0.575
GVCFP	Forward GVC participation	0.233	0.142	0.000	0.223	0.768
GVCBP	Backward GVC participation	0.138	0.035	0.053	0.138	0.193
upstream	Distance from final demand	2.917	0.514	1.095	2.956	4.245
downstream	Distance from primary inputs	2.301	0.434	1.000	2.415	3.203
PL (Value Added)	Labor productivity (Value Added, thousand USD)	48	38	1	39	205
PL (Total Output)	Labor productivity (Total Output, thousand USD)	144	130	1	108	872
gdp	Partner’s GDP (billion USD)	1061	2281	4	338	17,390
gdpcap	Partner’s per capita GDP (USD)	27,490	21,960	457	22,230	119,200
FTA	FTA dummy (1 = yes, 0 = no)	0.016	0.127	0.000	0.000	1.000
RL	Rule of law index	0.998	0.795	−1.084	1.120	2.125
RQ	Regulatory quality index	1.057	0.612	−0.866	1.138	2.025

Note: Sample size is 1845 (3 industries, 41 countries, 2000–2014).

presents the descriptive statistics for the key variables used in the regression model, covering the period 2000–2014, including means, standard deviations, and ranges.

Table 3. Top 10 trading partner countries by forestry export value.

Country	Export Value (Billion USD)	Share (%)
United States	38.829	19.56
Japan	22.574	11.37
South Korea	8.745	4.41
United Kingdom	7.487	3.77
Canada	6.968	3.51
Germany	6.312	3.18
Australia	5.056	2.55
France	3.640	1.83
Netherlands	3.427	1.73
Russia	3.097	1.56

Note: Export value is the total sum (billion USD) for 3 industries and 15 years. Share is the percentage of China’s total forestry exports to all countries worldwide. The top 10 countries account for over 60% of China’s forestry exports. Sample size is 1845 (41 countries, 3 industries, 2000–2014).

lists the top 10 trading partner countries by forestry export value, which collectively account for over 60% of China’s total forestry exports during the sample period, highlighting their significance as major markets. To further validate the data,

Table 4. T-Test Results for export quality by GVCFP Level.

Variable	Quality ($\mathit{\sigma }$ = 3)	GVCFP	GVCBP	upstream	downstream
Quality (σ = 3)	1.000
GVCFP	0.025	1.000
GVCBP	0.086 *	0.197 *	1.000
upstream	−0.054	0.231 *	0.053	1.000
downstream	−0.085 *	0.306 *	0.594 *	0.149 *	1.000

Note: Correlations are based on 1845 observations (3 industries, 41 countries, 2000–2014). * indicates p < 0.05. All correlations are below 0.6, indicating low multicollinearity.

presents t-test results comparing export quality means between countries with high and low forward GVC participation, demonstrating significant differences that support the relevance of GVCFP in explaining variations in the export quality.

5. Results

5.1. Main Regression Results

Table 5. Regression results of forward participation in GVCs on export quality.

	Quality ($\mathit{\sigma }$ = 3)	Quality ($\mathit{\sigma }$ = 3)	Quality ($\mathit{\sigma }$ = 5)	Quality ($\mathit{\sigma }$ = 5)
GVCFP	0.411 ***	0.395 ***	0.561 ***	0.537 ***
	(0.0852)	(0.0858)	(0.0909)	(0.0920)
GVCBP	0.0225 **	0.0201 **	0.0207 **	0.0194 *
	(0.00923)	(0.00952)	(0.00984)	(0.0102)
upstream	−0.0150 ***	−0.0170 ***	−0.0107 ***	−0.0121 ***
	(0.00309)	(0.00317)	(0.00329)	(0.00340)
downstream	−0.0283 ***	−0.0302 ***	−0.0326 ***	−0.0344 ***
	(0.00428)	(0.00436)	(0.00456)	(0.00467)
lngdp		−0.0664 *		−0.0263
		(0.0346)		(0.0371)
lngdpcap		0.0182		0.0021
		(0.0331)		(0.0355)
RL		0.00561		0.000713
		(0.0114)		(0.0122)
RQ		0.00171		0.00241
		(0.00926)		(0.00992)
FTA		−0.0210		−0.0251
		(0.0185)		(0.0198)
Constant	0.740 ***	2.003 ***	0.620 ***	1.156 *
	(0.0178)	(0.591)	(0.0190)	(0.634)
Observations	1845	1845	1845	1845
R-squared	0.734	0.847	0.711	0.821
Industry FE	YES	YES	YES	YES
Country FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES

Notes: The symbols ***, ** and * reflect statistical significance at the 1%, 5%, and 10% levels, respectively, the values in parentheses represent standard error clustered at the country-industry level.

reports the regression results examining the impact of forward participation in GVCs (GVCFP) on the export quality of China’s forestry products. Export quality is calculated using two values of substitution elasticity ($\sigma$ = 3 and $\sigma$ = 5) to ensure robustness. Columns (1) and (3) present the baseline specifications without control variables, while Columns (2) and (4) include the full set of control variables specified in Model (1).

Across all specifications, the coefficients of GVCFP are consistently positive and statistically significant at the 1% level, ranging from 0.395 to 0.561. For example, in Column (2) with $\sigma$ = 3, the coefficient of GVCFP is 0.395 ($p$ < 0.01), suggesting that a one-unit increase in forward participation in GVCs is associated with a 0.395-unit increase in export quality, holding other factors constant. This finding holds when using $\sigma$ = 5, as shown in Column (4), where the coefficient rises to 0.537 ($p$ < 0.01). The robustness of these results across different elasticity values and model settings provides strong support for Hypothesis 1, which posits that forward participation in GVCs improves the export quality of China’s forestry products, likely through technology diffusion and the adoption of sustainable practices such as eco-labeling [6].

These results contrast with some prior studies. For instance, Ndubuisi and Owusu found that forward participation in GVCs has no significant effect on export quality in developing countries [38]. This discrepancy may be attributed to the specific focus of our study on forestry GVCs, where leading firms are often positioned downstream in the value chain, potentially facilitating greater technology diffusion and quality upgrading through forward linkages. Moreover, our study focuses on China, a major forestry exporter and a key player in global trade, which may exhibit distinct dynamics compared to the broader sample of developing countries analyzed in Ndubuisi and Owusu [38].

The regression results also reveal several noteworthy patterns. First, the coefficient of backward participation in GVCs (GVCBP) is positive and significant, although its magnitude is smaller than that of GVCFP. In Column (2), the coefficient of GVCBP is 0.0201 ($p$ < 0.05), indicating that backward participation—through the importation of higher-quality inputs—remains an important driver of export quality in China’s forestry sector [3,41]. However, the larger and more significant coefficient of GVCFP suggests that forward participation represents a novel and impactful pathway for quality improvement, likely due to increased integration into downstream markets and exposure to advanced production standards.

Second, the coefficients of $\mathrm{u}\mathrm{p}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ and $\mathrm{d}\mathrm{o}\mathrm{w}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ are both negative and significant at the 1% level across all models. For instance, in Column (2), the coefficient of $\mathrm{u}\mathrm{p}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ is −0.0170 ($p$ < 0.01), and that of $\mathrm{d}\mathrm{o}\mathrm{w}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}{\mathrm{m}}_{\mathrm{i}\mathrm{c}\mathrm{t}}$ is −0.0302 ($p$ < 0.01). These findings suggest that as a trading partner’s position in the value chain moves farther from final demand (higher upstreamness) or primary inputs (higher downstreamness), the export quality of China’s forestry products declines. One plausible explanation is rooted in the U-shaped curve theory of value chains, which posits that technological content and value addition are typically highest at the two ends of the value chain—primary production (upstream) and final sales (downstream)—while intermediate stages often involve lower technological complexity [44]. In the forestry sector, this pattern may be particularly pronounced, as upstream activities (e.g., logging) and downstream activities (e.g., consumer-driven markets) often require higher technological inputs or quality standards, whereas intermediate stages (e.g., basic processing) may involve less innovation. Thus, trading partners positioned in the middle of the value chain may contribute to lower export quality when collaborating with China.

Regarding the control variables, the coefficient of GDP ($lngdp$) is negative and marginally significant in Column (2) (−0.0664, $p$ < 0.1), suggesting that larger market size may not necessarily lead to higher export quality, possibly due to competitive pressures in larger markets. The coefficient for per capita GDP ($lngdpcap$) is positive yet statistically insignificant (0.0182, $p$ > 0.1), suggesting that the consumption capacity of trading partners exerts a minimal direct influence on export quality in this setting. Likewise, the coefficients for rule of law (${\mathrm{R}\mathrm{L}}_{\mathrm{c}\mathrm{t}}$), regulatory quality (${\mathrm{R}\mathrm{Q}}_{\mathrm{c}\mathrm{t}}$), and the FTA dummy (${\mathrm{F}\mathrm{T}\mathrm{A}}_{\mathrm{c}\mathrm{t}}$) lack significance, implying that institutional factors and trade agreements may have a subordinate impact relative to GVC participation in promoting quality enhancements in China’s forestry exports.

Collectively, these results emphasize the substantial contribution of forward participation in GVCs to elevating the export quality of China’s forestry products, while also affirming the ongoing relevance of backward participation and the moderating influence of the trading partner’s value chain position.

5.2. Results of the Channels

To investigate the pathways by which forward participation in global value chains (GVCs) influences the export quality of China’s forestry products, this study assesses the moderating effects of trading partners’ technological capabilities and their positions within the value chain, as outlined in Model (7) in Section 4.4. The regression outcomes are presented in

Table 6. Moderating effects of technology and value chain position on relationship between forward participation in GVCs and export quality.

	Quality ($\mathit{\sigma }$ = 3)	Quality ($\mathit{\sigma }$ = 3)	Quality ($\mathit{\sigma }$ = 3)
GVCFP	0.0126 *	0.00400 *	0.237 ***
	(0.0125)	(0.0107)	(0.0480)
GVCFP × PL (Value Added)	0.000167 ***
	(0.000162)
GVCFP × PL (Total Output)		0.000148 ***
		(4.30 × 10−05)
GVCFP × downstream			−0.0471 ***
			(0.0138)
GVCFP × upstream			−0.0369 ***
			(0.0107)
GVCBP	0.394 ***	0.410 ***	0.392 ***
	(0.0859)	(0.0857)	(0.0876)
upstream	−0.0174 ***	−0.0170 ***
	(0.00321)	(0.00318)
downstream	−0.0307 ***	−0.0332 ***
	(0.00442)	(0.00441)
lngdp	−0.0667 *	−0.0627 *	−0.0474
	(0.0345)	(0.0344)	(0.0352)
lngdpcap	0.0203	0.0190	0.00540
	(0.0331)	(0.0329)	(0.0337)
RL	0.00551	0.00413	0.00916
	(0.0114)	(0.0114)	(0.0116)
RQ	0.000797	0.00118	0.000296
	(0.00925)	(0.00922)	(0.00942)
FTA	−0.0211	−0.0215	−0.0156
	(0.0185)	(0.0184)	(0.0188)
Constant	2.005 ***	1.931 ***	1.548 **
	(0.590)	(0.588)	(0.601)
Observations	1845	1845	1845
R-squared	0.849	0.854	0.818
Industry FE	YES	YES	YES
Country FE	YES	YES	YES
Year FE	YES	YES	YES

Notes: The symbols ***, ** and * reflect statistical significance at the 1%, 5%, and 10% levels, respectively; the values in parentheses represent the standard error clustered at the country–industry level.

, with the export quality derived using a substitution elasticity of $\sigma$ = 3.

Columns (1) and (2) of Table 6 report the results of the moderating effect of trading partners’ technological levels, measured by labor productivity ($PL$). The interaction term between forward participation in GVCs ($GVCFP$) and labor productivity ($GVCFP \times PL$) is positive and significant at the 1% level in both specifications. In Column (1), where labor productivity is calculated based on value added, the coefficient of the interaction term is 0.000167 ($p$ < 0.01). In Column (2), where labor productivity is derived from total output, the coefficient stands at 0.000148 ($p$ < 0.01). The consistency across different labor productivity measures shows that greater technical ability in trading partners boosts forward GVC participation’s positive effect on export quality. This supports the technology diffusion hypothesis, suggesting that forward participation lifts quality by enabling Chinese forestry firms to tap into advanced techniques and standards from tech-strong partners [59,60]. These uniform results across productivity metrics reinforce Hypothesis 2.

Column (3) examines how a partner’s value chain position moderates this, using $GVCFP \times upstream$ and $GVCFP \times downstream$ terms. Upstreamness measures a partner’s production distance from final demand (e.g., consumers) [54], while downstreamness gauges distance from primary inputs (e.g., raw materials, R&D). Higher upstreamness means farther from demand, and higher downstreamness places partners closer to the middle of the chain.

Both interaction terms are negative and significant ($GVCFP \times upstream$: −0.0369, $p$ < 0.01; $GVCFP \times downstream$: −0.0471, $p$ < 0.01). These findings indicate that the beneficial impact of forward GVC participation on export quality diminishes when trading partners are situated further from either the final demand or the primary inputs, namely nearer to the value chain’s intermediate segment. This observation corresponds with the U-shaped curve theory of value chains, which asserts that technological sophistication and value addition peak at the chain’s extremities (e.g., R&D and raw material extraction upstream, and branding and sales downstream), whereas mid-tier stages typically exhibit reduced technological complexity [44]. Within forestry GVCs, trading partners positioned centrally in the value chain—where upstreamness and downstreamness are elevated—may undertake activities of lesser technological intensity, such as rudimentary processing or assembly. As a result, partnering with such entities could constrain the technology diffusion benefits derived from forward participation in GVCs thus diminishing its favorable effect on export quality. This outcome offers additional support for the notion that technological diffusion constitutes a primary mechanism by which forward GVC participation affects export quality.

The control variable coefficients in Table 6 match those of Table 5. For instance, GVCBP stays positive and significant (e.g., 0.392 in Column (3), $p$ < 0.01), highlighting backward participation’s key role in quality gains. The upstream and downstream coefficients remain negative and significant in Columns (1) and (2), aligning with the main results in Section 5.1. Variables like $lngdp$ and $lngdpcap$ show similar trends to the primary regression, indicating limited direct quality impact. Overall, these findings confirm that partners’ technical strength significantly shapes forward GVC participation’s quality effect, supporting the technology diffusion mechanism, while mid-chain positions weaken these benefits.

5.3. Discussion

This study concludes that both forward and backward participation in GVCs by the forestry industry can promote an improvement in the quality of China’s forest product exports, which is basically consistent with the conclusions in the related literature that economic entities’ participation in GVCs improves export quality. At the macro level, Ndubuisi and Owusu concluded that economic entities’ participation in GVCs can significantly enhance export quality. At the micro level [38], the existing studies generally find that industrial enterprises in developing economies such as China, by participating in GVCs, improve the variety and quality of imported inputs, which significantly enhances firms’ export performance and improves export quality [59,61].

Regarding the role of GVCs in promoting export quality, the most significant difference between this study and the existing research lies in the finding that developing countries can promote export quality through forward participation in GVCs, which is a novel conclusion. The conclusion of Ndubuisi and Owusu is the closest to this study [59], but they used country-level data to conclude that forward participation in GVCs by developing countries does not significantly improve export quality. Meanwhile, other studies that have shown enhanced export performance through micro-level firm participation in GVCs mainly discuss the pathway through which firms expand the import of intermediate inputs—that is, backward participation in GVCs—and have less direct focus on forward participation in GVCs [60,62].

Regarding the technological diffusion enabled by GVCs, this study, along with the existing research, concludes that technological diffusion effects through “learning by doing” can enhance export performance. However, the learning pathways identified in the existing literature are at the import end of the value chain, emphasizing that developing countries should focus on importing intermediate inputs containing advanced knowledge and technology from developed countries, thereby improving the productivity of developing firms through backward participation in GVCs [21,63]. The findings of this study further indicate that, for forest product exports, forward participation in GVCs also generates technological diffusion effects, thereby enhancing export quality.

The possible reason for the differences between this study and the existing research may lie in the fact that this study focuses on forestry, which is a resource-based industry. Ndubuisi and Owusu targeted the entire industry of an economy, concluding that forward participation in GVCs does not enhance export quality in developing countries [38]. However, for resource-based industries such as forestry, the value chain leaders with advanced technology are often concentrated in the downstream part of the chain, such as market sales and brand-building. Therefore, to make their products more competitive, multinational corporations leading the value chain provide technical support to upstream producers [48], while also closely monitoring the production processes of upstream inputs to meet quality requirements [18]. Consequently, the forward participation of developing countries in GVCs at the export end also results in technological diffusion effects.

Therefore, this study—targeting resource-based industries like forestry where the value chain leaders are in the downstream segments—identifies a new pathway for developing countries to enhance export quality through forward participation in GVCs.

6. Conclusions and Policy Implications

Using China’s forestry context, this study analyzes export and GVC data with 41 partners from 2000 to 2014 to explore if forward GVC participation boosts forestry export quality and its mechanisms, offering key insights for emerging nations on using GVCs to improve quality and promote sustainable forestry growth.

• Forward participation in GVCs significantly improves the quality of China’s forestry exports, as evidenced by the consistently positive and significant coefficients of GVCFP across various model specifications.
• Technological diffusion serves as a critical channel for this effect, with forward participation facilitating quality improvements by enabling Chinese forestry firms to access advanced technologies and production standards.
• The impact of forward participation on export quality is moderated by the technological level and value chain position of trading partners; higher technological levels amplify the positive effect, while mid-chain positions weaken the technology diffusion benefits.

This study, focused on China’s forestry sector, shows that forward participation in global value chains (GVCs) significantly boosts forestry product export quality, with the diffusion of technology as a key driver, offering policy insights for emerging nations aiming to use GVCs for sustainable export growth.

• Forward GVC participation stands out as a novel, effective way to lift export quality in forestry, adding to the traditional approach of importing high-quality inputs via backward participation. Unlike sectors where backward participation typically enhances quality [38,53], forestry’s downstream-led structure aids technological spread through forward ties. For example, a Chinese firm sending pulp to a European furniture maker might adopt green methods to meet standards, enhancing export quality. This path highlights the value of global market ties at the export stage, especially in buyer-led sectors like forestry.
• Policymakers must carefully assess key firms’ value chain positions when using GVCs for quality enhancement. Data suggest that the diffusion of technology from forward participation varies by partner location, with mid-chain roles reducing benefits (Section 5.2). Developing nations should systematically analyze the technological proficiencies and value chain placements of their trading partners—whether proximate to primary inputs (e.g., R&D and raw materials) or final demand (e.g., branding and sales)—to ascertain whether forward or backward GVC participation is more conducive to technology diffusion. For example, partnering with entities at the downstream segment of the value chain, where consumer-oriented standards frequently demand advanced technologies, may optimize the advantages of forward participation [17].
• Fostering connections with tech-savvy firms in GVCs is vital for maximizing diffusion. Labor productivity’s moderating role (Section 5.2) shows advanced partners boost forward participation’s quality impact. Emerging economies should seek ties with such firms via joint ventures, tech transfers, or trade fairs to access modern methods and green practices. For instance, a Chinese forestry firm might team up with a German paper producer for energy-saving techniques, lifting quality and cutting environmental harm. Governments can assist with this through incentives like tax breaks or subsidies.
• While forward participation matters, backward participation remains the main quality driver in forestry, as seen in larger GVCBP coefficients (Section 5.1 and Section 5.2), aligning with the literature on input imports [53,59]. Importing top-grade logs from Canada, for example, directly boosts Chinese furniture quality. This study complements—not replaces—backward participation, urging policymakers to keep leveraging inputs while exploring forward potential, especially in downstream-influenced sectors like forestry.
• Beyond quality, forward GVC participation may support broader sustainability aims. Quality gains enhance competitiveness, profits, and innovation [1], enabling reinvestment in R&D and better processes. Meanwhile, technology diffusion suggests eco-benefits, like reduced energy use and emissions, advancing forestry sustainability. Developing nations should craft policies that blend economic goals (quality) with ecological aims (emissions cuts) for holistic progress.

This study leverages data from China, a leading developing country and one of the world’s largest forestry trade nations, to demonstrate that forward participation in GVCs significantly enhances export quality through technology diffusion. Conducted over 2000–2014, a period marked by China’s rapid trade growth and GVC integration following WTO accession, the findings offer valuable insights for other developing countries with resource-based industries, such as those in Southeast Asia, Africa, and Latin America. By engaging in forward GVC participation, these nations can adopt advanced technologies and standards, improving the quality of their forestry exports and promoting sustainable trade.

The relationship between forward GVC participation and environmental outcomes, such as carbon emissions, warrants further exploration. Given that GVC participation enhances export quality, future studies could investigate whether it also promotes the adoption of green technologies, contributing to sustainable development in forestry and other resource-based industries.

The study’s limitations are as follows:

• The reliance on data from the World Input–Output Database (WIOD) introduces limitations due to its time coverage, which ended in 2014, potentially affecting the timeliness of findings.
• The WIOD’s classification of forestry product processing categories is not sufficiently detailed, constraining the granularity of GVC analysis. Macro-level GVC studies depend on the time periods and industry classifications provided by input–output tables, thereby limiting the scope of investigation.