High-value-added forest product exports, the digital economy, and the efficiency of forest ecological product value realization

1 Introduction

Amid stronger global climate governance, ecological-civilization initiatives, and evolving green trade rules, forests are increasingly recognized not only as sources of timber and raw materials but also as natural capital that provides multiple services, including climate regulation, landscape recreation, and support for green industrial development (Costanza et al., 1997; Daily, 1997). As a major carrier of forest ecosystem services, forest ecological products not only maintain ecological security and provide public welfare but also have the potential to be converted into economic value through market transactions, industrial development, ecological compensation, green finance, and international trade (Millennium Ecosystem Assessment, 2005; Ouyang et al., 2020). However, externalities, public-good characteristics, and valuation difficulties often result in high ecological value but limited market conversion, as well as strong resource endowments but low industrial returns. Improving forest ecological-product value realization efficiency (FEPVRE) has become a key issue in high-quality forestry development, ecological product value realization, and regional green transformation. In recent years, China has continuously improved institutional arrangements related to ecological product value realization and green trade development. With the issuance of the “Opinions on Establishing and Improving the Mechanism for Ecological Product Value Realization,” establishing mechanisms for ecological product survey and monitoring, value assessment, operation and development, protection, and compensation has become a crucial approach to transforming regional ecological advantages into economic benefits (General Office of the CPC Central Committee and General Office of the State Council, 2021). The “Implementation Opinions on Expanding Green Trade” further emphasizes enhancing the green and low-carbon development capacity of foreign trade enterprises, supporting green design, green product certification, green supply chain construction, carbon footprint accounting, and the export of green low-carbon products (Ministry of Commerce of the People’s Republic of China et al., 2025). Ecological product value realization is no longer solely an issue in the resource and environmental domain but is also closely linked to trade structure optimization and digital transformation. FAO reports show that trade in timber, paper products, wooden furniture, and other processed forest products provides an important indicator of changes in the global forest industry and its trade structure (FAO, 2024a,b). Therefore, explanations of FEPVRE based solely on resource endowments or ecological investment can no longer adequately address practical challenges, making it necessary to incorporate external market demand, high-value-added forest product exports (HVFPE), and digital economy development into a unified analytical framework.

HVFPE differ from the expansion of general forest product exports. Their core lies in increasing the economic value per unit of forest resources through deep processing, quality control, green certification, and value chain extension. Heterogeneous firm trade theory posits that market entry for exports can improve industry productivity through firm selection and resource reallocation (Melitz, 2003). Export learning theory indicates that firms entering international markets may achieve productivity gains through technological learning, quality improvement, and management optimization (De Loecker, 2013). Meanwhile, the technological content and income characteristics of export products affect long-term economic growth and industrial upgrading (Hausmann et al., 2007). HVFPE may further enhance the efficiency of converting forest ecological products from ecological to economic value by expanding demand for high-quality green forest products, driving the application of green processing technologies, and promoting quality standard improvements. Compared with primary forest product exports, HVFPE involve deeper processing and longer value chains and may improve market recognition of forest ecological products through certification, traceability, standard mutual recognition, and cross-border marketing (Xiong et al., 2022; Jiang and Su, 2023; Wan et al., 2023). However, the positive impact of HVFPE on FEPVRE is not guaranteed. If export growth mainly relies on resource consumption or intensive processing, it may create forest resource pressure and regional factor misallocation, thereby weakening ecological capital allocation efficiency (Peng et al., 2022). With the continuous improvement of international green trade rules, forest certification, eco-labeling, and zero-deforestation supply chain requirements are profoundly influencing the competitive methods in forest product trade. Existing studies indicate that forest certification can both enhance international competitiveness of forest products through quality signals and green reputation, and create new trade barriers due to certification costs and compliance requirements (Cashore et al., 2004; Zhang et al., 2022). This implies that whether HVFPE can truly enhance FEPVRE depends not only on the processing depth and market premium of the export products themselves, but also on regional responsiveness to green standards and international market information.

The digital economy offers a new analytical perspective for understanding this issue. The development of digital technologies is transforming the production, circulation, trade, and regulation of ecological products (Bukht and Heeks, 2018; Goldfarb and Tucker, 2019; Wang et al., 2024). The digital economy can reduce information search and transaction costs, improve supply–demand matching efficiency, enhance forest resource monitoring and green certification credibility, and expand market boundaries through e-commerce and digital trade platforms (López González et al., 2023; UNCTAD, 2023). Accordingly, the digital economy may not only directly improve FEPVRE but also strengthen its association with HVFPE (Kong et al., 2023; Liang and Tan, 2024).

Existing studies explain FEPVRE mainly through within-region factors, whereas research on forest-product trade increasingly emphasizes export quality and value addition. The two literatures, however, remain weakly connected. The relationship between exports of high-value-added forest products and FEPVRE, as well as how this process is affected by industrial and digital conditions, remains to be examined. Against this background, this study uses panel data for 31 Chinese provinces from 2015 to 2023 to examine the effects of HVFPE and the digital economy on FEPVRE, and further analyzes the channel of forestry industrial-structure upgrading, the reinforcing role and stage-dependent characteristics of the digital economy, and differences across regions, levels of environmental regulation, and forest-resource endowments. This study makes three contributions. First, this study extends research on forest ecological product value realization from an open-economy perspective by examining the impact of HVFPE on FEPVRE. Second, from the perspective of industrial linkages, this study examines the role of forestry industrial-structure upgrading in converting export-related value added into forest ecological value, thereby deepening understanding of the relationship between external market demand and ecological value realization. Third, from the perspective of enabling conditions and boundaries, this study examines the reinforcing role of the digital economy and differences across stages of digital development and regional resource and environmental conditions, thereby further identifying the conditions under which HVFPE exert their value-conversion effects.

2 Literature review

2.1 Research on FEPVRE

Ecosystem-service valuation provides a conceptual foundation for understanding how the value of forest ecological products can be identified and ultimately realized. Costanza et al. (1997) estimated the global value of ecosystem services and natural capital. Daily (1997) then explained how nature’s services underpin human well-being, and the Millennium Ecosystem Assessment (2005) developed a systematic classification of ecosystem services. Building on these advances, accounting frameworks such as Gross Ecosystem Product (GEP) have been developed to quantify the contributions of ecosystems to economic and social development (Ouyang et al., 2020). Collectively, these approaches have substantially advanced the identification and quantification of forest ecosystem values, but accounting estimates alone do not indicate how effectively ecological resources are converted into tangible economic and social benefits. Accordingly, research has increasingly moved beyond ecological-value accounting to examine how efficiently ecological value is realized. Using Zhejiang Province as a case, Wang et al. (2022) developed an accounting framework for forest ecological products and assessed the efficiency with which their ecological value is converted into realized benefits, together with potential pathways for improvement. Lou et al. (2024) further conceptualized forest ecological-product value realization as a two-stage process comprising product supply and value conversion, identifying value-conversion capacity as a key component shaping overall efficiency. These advances shifted attention from the value of forest resources per se towards the way resources are connected with industrial and market systems. Against this background, industrial structure and institutional conditions have emerged as important explanations for variation in FEPVRE. Drawing on 26 mountainous counties in Zhejiang Province, Xu et al. (2024) examined how forestry industrial upgrading is associated with FEPVRE and further assessed the roles of rural digitalization and environmental regulation. Their findings suggest that industrial upgrading can strengthen value conversion as forest resources become integrated into processing, service, and market systems, while digitalization and environmental governance condition the effectiveness of this process. Zhang et al. (2025) extended this line of inquiry through the lens of new quality productive forces, applying dynamic QCA to examine configurations of innovation, technology, resource, environmental, and digital-productivity conditions and showing that high FEPVRE can emerge through multiple combinations of these factors.

Existing studies have found that resource endowments, industrial upgrading, digitalization, and environmental governance are important factors affecting FEPVRE. However, higher FEPVRE is not determined by any single factor, and different conditions may exhibit complementary or substitutive relationships. Existing studies mainly examine resource allocation, industrial structure, and the institutional environment within regions, while paying relatively limited attention to how external market demand affects FEPVRE.

2.2 Export quality and value addition in Forest products

International trade research has long explored how participation in export markets shapes firm productivity and industrial development. Building on firm heterogeneity, Melitz (2003) showed that access to export markets can raise aggregate industry productivity through the selection of more productive firms and the reallocation of resources. De Loecker (2013) further showed that firms may experience productivity gains after entering international markets through learning, improvements in production processes, and better management practices. As the literature evolved, export volume ceased to be the sole metric of trade performance. Hausmann et al. (2007) highlighted the importance of the technological sophistication and income content of exports for long-term economic growth, while Baldwin and Harrigan (2011) linked firms’ international competitiveness to differences in product quality. Manova and Zhang (2012) further documented systematic relationships among export prices, product quality, and destination-market characteristics. Collectively, these studies shifted the analysis of trade upgrading beyond export volumes towards product quality and positioning within value chains.

A similar shift is evident in research on forest-product trade. Given the resource-intensive nature of forestry, alternative processing pathways can differ markedly in resource productivity, technological intensity, and the market value they generate. Wan et al. (2023) showed that the expansion of China’s wood-based forest-product exports was accompanied by substantial quality differences, suggesting that the gains associated with export growth depend partly on product quality. From a global value-chain perspective, Xiong et al. (2022) showed that gross trade measures do not fully capture the value added retained by China’s forest-product sector within international production networks. Measures based solely on export value or volume may therefore obscure gains arising from deeper processing and movement towards higher-value positions in the value chain. Recent research has increasingly focused on this dimension of trade upgrading. Zhu et al. (2025) examined the relationship between global value-chain participation and the quality of China’s forest-product exports, finding that greater forward participation is associated with significantly higher export quality. At the same time, forest certification and evolving green trade rules have reshaped the conditions under which forest products compete internationally. Forest certification can enhance market recognition by signalling environmental performance and product quality, while simultaneously imposing certification and compliance costs on firms (Chen et al., 2020); evidence from China further suggests that certification can affect export prices and quantities in different ways (Zhang et al., 2022).

Existing studies show that processing intensity, product quality, value-chain position, and forest certification all affect forest product export performance, although their effects are not entirely consistent. Forest certification may enhance market recognition by signaling product quality and environmental performance, but certification costs and compliance requirements may also affect export volumes; moreover, changes in export prices and export volumes do not necessarily occur in parallel. More importantly, most related studies focus on trade competitiveness, export quality, or value-chain position as outcomes, with relatively little attention to the relationship between HVFPE and FEPVRE.

2.3 Research on digital economy and ecological value transformation

Digitalization reshapes the organization of economic activity by lowering the costs of information search, market matching, and transaction coordination. Goldfarb and Tucker (2019) conceptualized the economic effects of digital technologies in terms of changes in key economic costs, while Bukht and Heeks (2018) provided a systematic account of the scope and measurement of the digital economy. In international trade, digital connectivity can reduce information frictions and the economic costs associated with distance, enabling firms to access foreign markets and coordinate cross-border transactions more efficiently (López González et al., 2023). These cost reductions may be particularly important for forest ecological products because their dispersed origins and difficult-to-verify quality can restrict market access and the efficient integration of ecological resources into market systems.

A growing body of research has begun to examine how digitalization directly shapes forest ecological-product value realization. Digital technologies can support multiple stages of value realization, including resource inventories, ecological-value accounting, and market transactions (Wang et al., 2024). Kong et al. (2023) found that digital-economy development is associated with greater efficiency in converting forest ecological-product value and, through this process, with changes in urban–rural income distribution. Zheng et al. (2025) further examined the role of digital technologies in FEPVRE, showing that their contributions vary across regions with different digital conditions. Evidence linking digitalization to export quality provides a further basis for connecting digital development with ecological-value realization through trade. Digital technologies can improve firms’ access to information on foreign demand, facilitate coordination across production and supply chains, and increase their capacity to meet heterogeneous quality requirements across destination markets. Empirical evidence suggests that digitalization is associated with higher export-product quality (Zhang and Duan, 2023; Liang and Tan, 2024). In the context of green trade, Ou et al. (2025) combined Chinese industrial-firm and customs data and found that digital-economy development improves green-product export quality by facilitating the allocation of capital, labour, and technological inputs. Evidence from the forest-product sector similarly indicates that regional digital development and disparities in digital access are associated with China’s forest-product export performance (Jing et al., 2026).

Existing studies have separately examined the direct effects of the digital economy on forest ecological product value realization and the role of digitalization in promoting export trade, providing a basis for understanding the link between the two. However, these two strands of research remain relatively separate, and there is still a lack of systematic evidence on whether the digital economy merely exerts a direct influence on regional development or can improve information matching and transaction conditions, thereby strengthening the effects of HVFPE. Whether this strengthening effect varies across different stages of digital development also requires further examination.

2.4 Synthesis of the literature and research gap

Existing research provides three foundations for this study: FEPVRE is influenced by factors such as resource endowments, industrial structure, digitalization, and environmental governance; research on forest product trade has expanded from export scale to processing-based value addition, product quality, and value-chain position; and digital technologies can improve information access, resource allocation, and cross-border transaction conditions. However, these strands of research remain insufficiently connected. Research on forest ecological product value realization has paid relatively little attention to external market demand, studies of forest product trade generally stop at export performance, and research on the digital economy has not sufficiently distinguished between its direct effects and its reinforcing role in export activities. Against this background, this study examines the relationship between HVFPE and FEPVRE, analyzes the potential channel of forestry industrial structure upgrading, and further tests the reinforcing role of the digital economy, its stage-specific changes, and related heterogeneity.

3 Theoretical analysis and research hypotheses

3.1 Theoretical framework

This study focuses on the relationship between HVFPE and FEPVRE, and further examines the roles of forestry industrial structure upgrading and the digital economy in this relationship. HVFPE can help expand external markets and increase the level of processing and utilization of forest resources. Requirements in export markets for intensive processing, green certification, and stable supply may also be transmitted to the forestry production system, promoting the development of processing, manufacturing, and related service activities; thus, forestry industrial structure upgrading may serve as a channel through which this relationship operates. The digital economy can improve access to information, supply–demand matching, and transaction conditions, while reducing the costs for firms of entering external markets and organizing cross-border transactions. This may not only help improve FEPVRE but also strengthen the positive relationship between HVFPE and FEPVRE. As the level of digital economy development increases, this reinforcing effect may also exhibit stage-specific changes. Differences across regions in development foundations, environmental regulation intensity, and forest resource endowments may also cause the above relationships to exhibit different characteristics. Accordingly, this study constructs a conceptual framework and proposes corresponding research hypotheses, as shown in Figure 1.

3.2 HVFPE and FEPVRE

FEPVRE is not equivalent to forest resource endowment or the total value of ecosystem services. Its core lies in whether ecological products can be identified, measured, managed, and converted into economic returns, social welfare, and ecological-protection investment under given ecological-capital, labor, land, and financial constraints of capital, labor, and land inputs. It measures the effective output capacity of ecological capital after entering the socio-economic production system (Ouyang et al., 2020). The supply capacity of forest ecological products and their value conversion capacity are not necessarily synchronized, and industrial support and income distribution in the value conversion process often determine overall efficiency levels (Lou et al., 2024). Therefore, explaining FEPVRE cannot remain at the level of forest resource abundance, but should further examine whether resource factors can be transformed into high-quality outputs through industrial organization and market mechanisms (Wang et al., 2022).

From the perspective of transaction costs, forest ecological products exhibit strong externalities, public-good characteristics, and opaque quality information, which weaken market participants’ willingness to recognize and pay for ecological value. Property rights definition and transaction arrangements affect resource allocation efficiency (Coase, 1937), while incomplete information, asset specificity, and opportunistic behavior increase market transaction costs (Williamson, 1985). The HVFPE examined in this study mainly include wood-based panels, wood products, wooden furniture, and related products. Exports of these products can expand international market demand for processed forest products and promote the development of related activities such as processing, certification, logistics, and marketing. Export markets select more productive firms and improve overall industry efficiency through resource reallocation (Melitz, 2003). De Loecker (2013) found at the firm level that enterprises entering export markets may gain productivity benefits through technological learning, quality improvement, and management optimization.

For the forestry sector, the external demand generated by HVFPE must be absorbed by regional industries before it can be further translated into FEPVRE. Gereffi (1999) pointed out that after firms and regions are embedded in international markets, their ability to move toward higher value-added segments is key to industrial upgrading. Studies show that China’s wood-based forest-product exports are shifting from quantity expansion toward quality upgrading (Wan et al., 2023). A higher position in the value chain is also associated with stronger value-added trade performance (Xiong et al., 2022).

High-value-added exports also improve ecological value conversion through green constraints and reputation mechanisms. The natural resource-based view posits that firms’ competitive advantage can stem from pollution prevention, product management, and sustainable development capabilities (Hart, 1995). In forest product trade, forest certification, legal timber traceability, and eco-labeling are both compliance requirements for high-standard markets and important means of signaling green quality to consumers (Chen et al., 2020). When export products shift from low-end processing to green design, branded trade, and advanced manufacturing, the ecological attributes of forest resources are more easily recognized by the market and reflected in product prices. Therefore, compared with general export scale expansion, high-value-added exports better reflect the forestry industry’s deep development and effective utilization of FEPVRE, thereby helping to enhance FEPVRE. Based on the above analysis, the following hypothesis is proposed:

H1: HVFPE can significantly enhance FEPVRE.

3.3 Digital economy and FEPVRE

The digital economy affects FEPVRE primarily through reshaping information structures and transaction methods. At the same time, digital technologies possess strong penetrative power and broad complementarity (Bresnahan and Trajtenberg, 1995). For forest ecological products, with significant quality differences and ecological attributes that are difficult to observe directly, the cost-reducing effect of digital technologies can alleviate insufficient value recognition and suboptimal supply–demand matching. The direct impact of the digital economy on FEPVRE mainly occurs in four areas: resource monitoring, value accounting, product transactions, and income distribution. Remote sensing can enhance information acquisition on forest resource quality, ecological functions, and management status, making ecological value easier to account for and compare (Wang et al., 2024). Digital finance can expand the market reach of ecological products, lower entry barriers for operators, and improve consumer access to ecological products (Kong et al., 2024).

In the operation of the forestry industry, the digital economy also improves factor allocation and industrial coordination efficiency. Internet use can promote international trade growth, mainly by reducing information search and matching costs in cross-border transactions (Freund and Weinhold, 2004). Katz and Shapiro’s (1985) research on network externalities shows that network value increases as user scale and connection density rise. The realization of FEPVRE requires coordinated participation among farmers, markets, and government agencies, and the network externalities of digital platforms help connect dispersed actors into more stable market networks. Existing forestry studies provide empirical support these proposed relationships. However, the empowering effect of the digital economy exhibits clear stage characteristics. Digital infrastructure construction, data factor circulation, platform usage capacity, and digital skills all involve fixed costs; at low development levels, digital platforms may merely serve as online substitutes for traditional transaction methods and may not yet generate systematic efficiency improvements. Differences in digital access, digital skills, and digital usage capacity affect the distribution of digital dividends (van Dijk, 2006). In forest product exports, the level of digital economy development and the digital divide in destination countries affect trade costs, trade facilitation, and institutional compatibility; when the digital gap becomes too large, the trade-promoting effect of the digital economy weakens (Jing et al., 2026).

The digital economy can directly affect the process of forest ecological product value realization by improving resource monitoring, market transactions, and factor allocation; it can also complement HVFPE by reducing information and transaction costs associated with demand identification, green certification, and supply-chain coordination. The former reflects the direct effect of the digital economy, whereas the latter reflects its moderating role in the value-conversion relationship associated with HVFPE. Thus, the digital economy may not only directly improve FEPVRE, but also generate threshold effects in the process through which HVFPE influence such efficiency (Hansen, 1999; Kong et al., 2023). When the level of digital economy development is below the threshold, incompatible information systems, inadequate platform services, and insufficient digital capabilities among market participants may limit the efficiency gains associated with HVFPE. Once the level of digital economy development exceeds the threshold, network externalities, data sharing, supply-chain coordination, and the digitalization of green certification may generate cumulative effects, making it easier for the market demand and green attributes of high-value-added forest products to be translated into market returns. Accordingly, the following hypotheses are proposed:

H2a: Digital economy development can significantly improve FEPVRE.

H2b: The digital economy exhibits a threshold effect in the process by which HVFPE affect FEPVRE.

3.4 The reinforcing role of the digital economy in the effects of HVFPE

For HVFPE to improve FEPVRE, they must operate through multiple stages, including international demand identification, green standard compliance, quality certification, cross-border logistics, brand marketing, and revenue return. Information asymmetry or coordination failures at any stage may cause export gains to remain at the firm level rather than being effectively translated into ecological value realization at the regional level. Digital connectivity and digital trade policies can promote cross-border transactions by reducing trade costs (López González et al., 2023). Digital trade provides new opportunities for developing economies to participate in international markets, but its effects are constrained by digital infrastructure, digital skills, and institutional environments (UNCTAD, 2023). The digital economy may moderate this relationship by improving demand identification and market matching. High-value-added forest products often possess differentiation, customization, and green attributes, requiring export firms to continuously identify changes in overseas demand for certified timber, eco-friendly furniture, low-carbon packaging, and green building materials. Digital platforms, big data analytics, and cross-border e-commerce can shorten the information gap between firms and overseas buyers, enhancing the matching efficiency of product design, price negotiation, and order organization (Liang and Tan, 2024). In studies on digitalization of forest product trade, the digital economy promotes forest product exports by reducing trade costs, enhancing trade facilitation, and improving institutional quality, further affects the actual performance of HVFPE (Jing et al., 2026) (Figure 1).

The moderating role of the digital economy is also reflected in green compliance and supply chain transparency. With the strengthening of green trade rules, export firms must not only demonstrate product quality but also verify legal timber sourcing, traceable processing, and accountable carbon footprints. Studies on forest certification indicate that green certification can enhance market trust but can also increase firms’ compliance costs (Chen et al., 2020). If a region has strong digital traceability, online certification, and supply chain management capabilities, firms can manage certificates, batch tracking, contract fulfillment, and logistics coordination at lower costs, converting green compliance pressures into brand reputation and price premiums.

The competitive advantage of high-value-added forest products derives not only from processing depth and brand value but also from their ecological value. However, these attributes are difficult to observe directly and must be communicated through credible disclosure mechanisms. Digital technologies can record product origin, processing history, and certification data, thereby enhancing the transparency of ecological value. This facilitates market recognition of the ecological value embedded in high-value-added forest products and its conversion into economic returns. Previous studies show that digital connectivity and digital trade policies can reduce trade costs and promote international trade, although their effects depend on digital infrastructure, digital skills, and institutional conditions (López González et al., 2023; UNCTAD, 2023). Meanwhile, the digital economy can enhance the technological sophistication of exports (Liang and Tan, 2024). In the context of forest product exports, higher levels of digital economy development make product information, green certification, and supply-chain information associated with high-value-added forest products easier for international markets to recognize and absorb, thereby allowing their positive effect on FEPVRE to be more fully realized. Therefore, the effect of HVFPE on FEPVRE depends on their integration with regional digital capabilities. Based on the above analysis, the following hypothesis is proposed.

H3: The digital economy positively moderates the relationship between HVFPE and FEPVRE.

3.5 Transmission channel of HVFPE

The impact of HVFPE on FEPVRE is not limited to increased returns from expanded export volume; it also operates through changes in the forestry industrial structure. Realizing the value of forest ecological products requires an established industrial system. If the forestry sector remains concentrated in low-value-added production, forest resources, even with high ecological value, cannot be fully converted into stable market returns. Conversely, when the forestry industry shifts from resource-dependent production to deep processing, the ecological functions and economic attributes of forest resources are more easily absorbed by the market through the industrial chain. HVFPE usually entail higher product standards and more complex market requirements. After entering international markets, firms cannot maintain competitiveness solely through resource endowment and low-cost advantages; they must continuously improve production processes and product quality. Competitive pressure in export markets drives forestry firms to increase processing depth and product added value, and also pushes related industrial segments from low-end production toward advanced research and development (Fu, 2010).

The impact of forestry structural upgrading on FEPVRE is mainly reflected in resource-use efficiency and value-conversion capacity. Industrial upgrading increases the output generated per unit of forest resources, allowing the same resource input to produce higher economic returns and reducing the low returns associated with primary processing and extensive utilization (Xu et al., 2024). Meanwhile, as the forestry value chain extends, forest resources enter the market not only as primary products but also through diverse value-realization channels such as green certification and ecotourism. A more advanced industrial structure may diversify the market forms of forest ecological products and strengthen the channels through which ecological resources generate economic returns and social welfare (Gereffi, 1999). From a resource-allocation perspective, HVFPE also promote the concentration of capital, technology, and labor in more efficient forestry sectors. Advanced processing, green manufacturing, and brand trade require higher technical and management capabilities, attracting more production factors to high-value-added segments and addressing short supply chains and weak value conversion in traditional forestry. As forestry structures are optimized, forest resource utilization becomes more refined, enhancing the industrial absorption and market monetization of ecological products (Hart, 1995). Therefore, HVFPE influence FEPVRE not only by expanding external market demand but more importantly by promoting forestry structural upgrading, altering how forest resources engage in market division of labor and value distribution. Based on the above analysis, the following hypothesis is proposed.

H4: HVFPE can promote the upgrading of the forestry industrial structure, thereby providing a potential industrial-structure transmission channel through which they affect FEPVRE.

3.6 Heterogeneous effects of regional conditions, environmental regulation, and forest resource endowment

The relationships between HVFPE, digital development, and FEPVRE may depend on regional development conditions and resource and environmental characteristics. Forest ecological-product value realization is shaped not by any single economic factor, but by the interplay of multiple economic, institutional, and resource conditions. Regional differences in location, environmental-regulation intensity, and forest-resource endowments can create distinct capacities and opportunities for value realization; consequently, comparable levels of HVFPE and digital development may yield systematically different marginal associations across regions.

Differences in locational conditions may affect the roles of HVFPE and the digital economy. The eastern, central, and western regions differ in their industrial development foundations, degree of openness, and stages of digital development, and these regional differences may lead to different estimated results for HVFPE and the digital economy. Compared with the central and western regions, the eastern region has a greater concentration of export-oriented industries and supporting export services, and the positive relationship between HVFPE and FEPVRE may therefore be stronger; by contrast, digital development in the central and western regions may have greater scope for marginal improvement (López González et al., 2023). Therefore, this study groups the sample into eastern, central, and western regions to examine regional differences in the relevant estimation results.

Environmental-regulation intensity may also condition the relationship between HVFPE and FEPVRE. HVFPE are increasingly tied to requirements for cleaner production, legal sourcing, forest certification, eco-labeling, and broader environmental compliance. Environmental regulation can shape firms’ production practices, patterns of resource use, and incentives for green innovation. Well-designed environmental regulation may induce innovation that partially offsets the costs of compliance. In forest-product trade, certification, proof of legal origin, and environmental standards have become increasingly important determinants of market access and product acceptance (Cashore et al., 2004; Chen et al., 2020; Zhang et al., 2022; Wu et al., 2024). Differences in regulatory stringency may therefore alter the balance between compliance costs and incentives for green innovation, leading the contribution of HVFPE to FEPVRE to vary across regulatory contexts. Relative to weakly regulated regions, stronger environmental regulation may make it more likely that the value-added attributes of forest-product exports are translated into gains in resource-use efficiency and ecological-value realization by reinforcing cleaner-production requirements and incentives for technological upgrading.

Forest-resource endowments shape the underlying supply capacity for forest ecological-product value realization. Variation in the quantity, quality, and ecological functions of forest resources determines the types, scale, and quality of ecological products that regions are able to supply. Regions with richer forest-resource endowments generally possess a stronger ecological-product supply base and greater scope for value realization. Where forest resources are more limited, however, ecological-product development and value conversion may remain resource-constrained even when export capacity and digital infrastructure are relatively well developed (Zheng et al., 2025). Richer forest resources can therefore support a more stable supply of both forest products and ecological products, while allowing high-value-added exports to sustain value realization through greater resource availability, higher product quality, and stronger ecological branding. Digitalization cannot substitute for forest natural capital itself, but it can increase the value realized from limited resources by improving resource monitoring and information disclosure, extending market reach, and reducing information asymmetries and transaction costs.

Based on the above analysis, the effects of HVFPE, the digital economy, and their interaction on FEPVRE may not be uniform across regions. Locational conditions affect market connectivity and industrial absorptive capacity, environmental regulation influences green production constraints and innovation incentives, and forest resource endowments shape the supply base and value-conversion potential of ecological products. Therefore, the effects of these variables on FEPVRE may differ across regions with varying locational conditions, environmental regulatory intensity, and forest resource endowments. Based on this reasoning, the following hypothesis is proposed.

H5a: The relationships between HVFPE and the digital economy, on the one hand, and FEPVRE, on the other, differ across the eastern, central, and western regions.

H5b: The positive relationship between HVFPE and FEPVRE is stronger under more stringent environmental regulation.

H5c: HVFPE contribute more strongly to FEPVRE in regions with richer forest-resource endowments, whereas digitalization generates larger marginal gains where forest-resource endowments are relatively limited.

4 Research design and variable selection

4.1 Baseline regression model

We estimate two-way fixed-effects models with province and year fixed effects to account for time-invariant cross-provincial heterogeneity and common annual shocks. The models assess the associations of HVFPE and digital economy development with FEPVRE. To examine forestry industrial upgrading as a potential transmission channel, we follow Jiang (2022) and retain the same controls and fixed-effects structure while replacing the dependent variable with the forestry industrial upgrading index (FISU), allowing us to assess whether HVFPE are associated with industrial upgrading (Figure 2). The empirical specifications are as follows:

Y i , t = α 0 + α 1 X i , t + β Cons i , t + μ i + λ t + ε i , t (1)

In the equations, Y_i,t denotes the dependent variable, representing FEPVRE in the baseline regressions and being replaced by forestry industrial-structure upgrading (FISU) in the transmission-channel test; X_i,t denotes the core explanatory variable, representing HVFPE and the level of digital-economy development (Dig), respectively, in the baseline regressions, and HVFPE in the transmission-channel test. Cons_i,t denotes the control variables; α₀ is the constant term, while α₁ and β are the coefficients to be estimated for the core explanatory variable and control variables, respectively; i and t denote province and year, respectively; μ_i and λ_t denote province and year fixed effects, respectively, and ε_i,t is the random disturbance term.

4.2 Interaction effect model

To assess whether digital development conditions the relationship between HVFPE and FEPVRE, we augment the baseline specification with an interaction term between HVFPE and digital-economy development (Dig). A statistically meaningful interaction would indicate that the association between HVFPE and FEPVRE varies with the level of digital development, consistent with a moderating role of digitalization.

FE it = α 2 Dig i , t + ( a 1 + a 3 Dig i , t ) Hev i , t + β Cons i , t + μ i + λ t + ε it (2)

In the equation, FE denotes FEPVRE, Dig denotes the digital economy, and HEV denotes high-value-added forest products, α₃ represents the conditional marginal effect of HVFPE on FEPVRE at a given level of digital economy development.

4.3 Threshold effect model

We estimate a panel threshold model in which the digital economy index serves as the threshold variable (Hansen, 1999). The underlying idea is that when the digital economy is below or above a certain critical value, the impact of HVFPE on FEPVRE may vary, exhibiting clear nonlinear threshold characteristics.

FE it = α 0 + β 1 Hev i , t ( Dig it ≤ γ ) + β 2 Hev i , t I ( Dig it > γ ) + ∑ k δ k Control kit + μ i + λ t + ε it (3)

In the model, Dig_i,t denotes the digital economy and serves as the threshold variable; γ is the threshold value to be estimated; I(·) is an indicator function that equals 1 when the specified condition is satisfied and 0 otherwise; the remaining coefficients are defined as above. In this model, β₁ and β₂ represent the effects of HVFPE on FEPVRE under different digital economy threshold regimes. A significant difference between β₁ and β₂ indicates that the digital economy alters the effect of HVFPE, implying the existence of a threshold effect. If β₂ > β₁, the positive effect of HVFPE on FEPVRE strengthens after the digital economy surpasses the threshold; if β₂ < β₁, this effect weakens as the digital economy improves.

4.4 Variable selection

4.4.1 Dependent variable: FEPVRE

Drawing on previous studies (Kong et al., 2023; Kong et al., 2024; Xu et al., 2024), this study selects multiple indicators from the input and output dimensions (Table 1) and employs the SBM–Malmquist index to measure FEPVRE across Chinese provinces. The input indicators include the values of five forest ecological products: carbon sequestration and oxygen release, climate regulation, water conservation, soil retention, and cultural services. Their service quantities are estimated using models such as InVEST and CASA. To improve the scientific rigor and accuracy of valuation, calculations follow the recommendations of the Technical Guidelines for Ecosystem Assessment: Gross Ecosystem Product Accounting of Terrestrial Ecosystems. The remaining input indicators are forestry fixed-asset investment, forestry labor, and forest area. The output indicator is total forestry output value. Gross forestry output value covers the primary, secondary, and tertiary forestry industries and can comprehensively reflect the market-oriented output generated from forest ecological products through production, processing, and service activities under a consistent and continuous provincial statistical framework; it can therefore be used as an output indicator in the SBM–Malmquist framework. This indicator differs from HVFPE calculated from customs HS codes in terms of data sources, accounting objects, and economic meaning, and therefore does not represent a duplicate construction of the same variable.

Table 1. Measurement methods for FEPVRE.

4.4.2 Core explanatory variables

4.4.2.1 Export value of high-value-added forest products

Drawing on the methods used by Peng et al. (2022) and Wang et al. (2023), this study classifies high-value-added wood forest products to include engineered wood (HS4408–HS4413), wood products (HS4414–HS4421), and wooden furniture (HS940161, HS940169, HS940330, HS940340, HS940350, HS940360). The export level of high-value-added forest products is measured as the natural logarithm of annual provincial export values of these products. The original export values are uniformly denominated in US dollars and transformed using the natural logarithm to measure the scale of HVFPE at the provincial level. Furthermore, the share of HVFPE in total forest product exports is used as an alternative indicator to reflect the adjustment of the export product structure toward higher-value-added categories.

4.4.2.2 Digital economy

Following Fan and Wu (2021), the digital economy is measured along four dimensions: digital infrastructure, digital circulation, digital industry, and digital innovation. The symbols, economic meanings, and measurement methods of the main variables are shown in Table 2. Digital infrastructure captures access, coverage, and network capacity. Digital industry measures the scale and absorptive capacity of core digital sectors. Digital circulation reflects enterprise adoption, e-commerce penetration, and circulation efficiency. Digital innovation captures R&D inputs, personnel, and intellectual-property outputs. To eliminate differences in measurement units and magnitudes across indicators, all indicators are normalized using range standardization over the full sample; the entropy method is then used to calculate the global objective weight of each indicator. Each dimensional index is constructed as the weighted sum of its constituent indicators using weights normalized within that dimension, while Dig is calculated by weighting all indicators according to their global entropy weights.

Table 2. Composition and description of the digital economy index.

4.4.3 Control variables

To systematically analyze the impact of forest product exports on FEPVRE, this study selects control variables from social, economic, and natural dimensions. (1) Economic Development Level. The logarithm of provincial per capita GDP is used to measure economic development, reflecting differences in regional economic base, market size, and residents’ consumption capacity. (2) Urbanization Rate. Measured as the ratio of urban permanent population to total permanent population, reflecting population concentration, resource allocation efficiency, and urban expansion. (3) Industrial Structure. Measured as the share of secondary industry value-added in regional GDP, reflecting industrial composition. (4) Public Finance. The logarithm of local general public budget expenditure is used to reflect fiscal capacity and public resource allocation. (5) Openness. Measured as the ratio of regional total imports and exports to regional GDP. (6) Transportation Infrastructure. Measured as the logarithm of regional road length. (7) Annual Rainfall. Logarithm of the province-level annual average rainfall is used to capture differences in natural climate conditions. (8) Ecological Protection Intensity. Measured as the proportion of regional nature reserve area to total jurisdiction area.

4.4.4 Mechanism variable

Forestry structural upgrading reflects the transition of the forestry sector from a low-level to a high-level industrial structure, characterized by declining shares of the primary and secondary forestry industries and a rising share of the tertiary forestry industry. To capture the extent of this structural transformation, this study follows previous research and employs the structural hierarchy coefficient method to measure the level of forestry structural upgrading (Fu, 2010; Xu et al., 2024). This method assigns different hierarchical weights to the primary, secondary, and tertiary forestry industries, thereby capturing the structural upgrading effects arising from changes in their value-added shares. The calculation formula is as follows:

FISU it = ∑ x = 1 3 ( F xit F it × j ) (4)

In the equation, FISU_it represents the level of forestry structural upgrading in year t; F_xit denotes the value-added of the x-th forestry sector for the i-th unit in year t; F_it is the total forestry value-added of the i-th unit in year t; x refers to forestry sector categories corresponding to the primary, secondary, and tertiary forestry industries.

4.5 Data sources

This study examines 31 provincial-level administrative regions in China during 2015–2023. Data on HVFPE were obtained from the China Customs Statistics Database. The dataset combines socioeconomic indicators with remote-sensing data. Socioeconomic indicators were compiled from the China Statistical Yearbook, the China Forestry and Grassland Statistical Yearbook, and provincial statistical yearbooks for 2015–2023 (Table 3). Remote sensing data were mainly obtained from the Resource and Environmental Science and Data Center of the Chinese Academy of Sciences, the CLCD dataset of Wuhan University, the NASA Geospatial Data Center, the China Meteorological Data Service Centre, and the Geospatial Data Cloud platform. Before estimating the regression models, we calculated VIFs to assess multicollinearity among the explanatory variables. The results show that the VIFs of all variables range from 1.079 to 6.016, with a mean VIF of 3.038; all values are below 10, indicating that the model does not suffer from serious multicollinearity.

Table 3. Descriptive statistical analysis.

5 Empirical results and analysis

5.1 Baseline regression analysis

Table 4 reports the baseline regression results for the relationships among HVFPE, the digital economy, and FEPVRE. All models control for province and year fixed effects to minimize the influence of unobserved regional heterogeneity and common temporal shocks on the estimation results. In column (1), the estimated coefficient on HVFPE is 0.075 and is statistically significant at the 1% level, indicating that, after controlling for other variables and two-way fixed effects, a 1% increase in HVFPE is associated with an average increase of 0.00075 units in FEPVRE. In terms of economic magnitude, a 10% increase in HVFPE is associated with an increase of approximately 0.007 units in FEPVRE, equivalent to 0.71% of the sample mean. In column (2), the estimated coefficient on the digital economy is 0.962 and is statistically significant at the 1% level, indicating that a one-unit increase in the digital economy index is associated with an average increase of 0.962 units in FEPVRE. The estimated coefficients of both core explanatory variables are significantly positive, providing preliminary support for Hypotheses H1 and H2a.

Table 4. Baseline regression results.

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

5.2 Robustness checks

We conducted six robustness checks of the baseline estimates (Tables 5, 6). First, all continuous variables are winsorized at the 1st and 99th percentiles to reduce the influence of extreme values on the estimates. After winsorization, the estimated coefficients on HVFPE and the digital economy are 0.026 and 0.845, respectively. Second, after excluding 2020–2022, which were strongly affected by the COVID-19 pandemic, the coefficients on HVFPE and the digital economy are 0.026 and 0.877, respectively. Third, after excluding Tibet, Shanxi, Gansu, Ningxia, and Qinghai, five provinces or autonomous regions with relatively low exports of high-value-added forest products, the coefficients on HVFPE and the digital economy are 0.045 and 0.774, respectively. Fourth, the value of high-value-added exports is replaced with the share of high-value-added exports in total exports; under this alternative measure, the estimated coefficient on the high-value-added export share is 0.087. The estimated coefficient on the digital-economy index is 0.947. Fifth, province-specific linear time trends are included to control for potential linear trends over time within each province. After including the trend terms, the coefficients on HVFPE and the digital economy are 0.029 and 0.832, respectively, and remain significantly positive. Sixth, spatial robustness tests are conducted. Given the potential spatial correlation in interprovincial economic activities, this study applies the Pesaran CD test to the residuals of the baseline model and calculates Conley spatial HAC standard errors based on geographic distances between provinces. A Bartlett kernel function is employed, with spatial cutoff distances set at 500, 1,000, and 1,500 km, while controlling for first-order temporal correlation. The Pesaran CD test finds no significant cross-sectional dependence (p = 0.102). After further applying Conley spatial HAC standard errors, the coefficients of HVFPE and the digital economy remain stable at 0.031 and 0.835, respectively, across different cutoff distances, and both are significant at the 1% level. Across all robustness tests, the regression coefficients of Hev and Dig remain positive and statistically significant at the 1% level, and the main conclusions remain unchanged.

Table 5. Robustness check results.

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

Table 6. Spatial robustness tests.

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

5.3 Endogeneity test

Regions with higher FEPVRE may also have more developed industrial structures, stronger export capacity, and higher levels of digital economy development, potentially generating a reverse relationship between higher efficiency and the expansion of HVFPE. To examine this possibility, this study sequentially employs one-period-lagged HVFPE and the system GMM estimator. The lagged-variable results show that the estimated coefficient on L. HVFPE is 0.006 and is positive and statistically significant at the 5% level, indicating that the positive estimate remains after HVFPE are lagged by one period (Table 7). This also separates the contemporaneous feedback from current FEPVRE to current exports from the core explanatory variable, thereby reducing interference from contemporaneous reverse relationships. Considering the dynamic persistence of FEPVRE, this study further employs system GMM for testing. The dynamic panel equation includes the first-order lag of FEPVRE, with FEPVRE and HVFPE specified as GMM-style variables and their second- and third-order lags used to construct internal instruments; the digital economy, control variables, and year dummy variables are treated as conditionally exogenous IV-style variables in both the differenced and level equations. The model is estimated using one-step robust estimation with a small-sample adjustment. To limit the number of instruments, the lag window is restricted to the second and third lags, and the GMM-style instruments are collapsed. The system GMM results show that the estimated coefficient on HVFPE is 0.02179 and is positive and statistically significant at the 10% level. The p-values of the AR(1) and AR(2) tests are 0.020 and 0.145, respectively; those of the Hansen and Sargan tests are 0.175 and 0.107, respectively; and all reported p-values for the difference-in-Hansen tests are no lower than 0.144. The relevant diagnostic tests do not reject the corresponding null hypotheses regarding model specification and instrument validity, indicating that the positive estimate of the core variable remains robust under the dynamic panel specification.

Table 7. Endogeneity checks using lagged HVFPE and system GMM.

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

5.4 Moderating effect analysis

To test whether the digital economy moderates the association between HVFPE and FEPVRE, we added the mean-centered interaction term reported in Table 8. Column (1) includes the interaction term after centering the relevant variables. When the digital economy index is set at its sample mean of 0.213, the marginal effect is approximately 0.026; that is, a 10% increase in HVFPE is associated with an increase of approximately 0.0028 units in FEPVRE, equivalent to 0.28% of the sample mean. When the digital economy index increases from 0.146 at the 25th percentile to 0.266 at the 75th percentile, the marginal effect increases from 0.0286 to 0.0308, representing an increase of approximately 7.6%. This indicates that higher levels of digital economy development strengthen the positive relationship between HVFPE and FEPVRE, supporting Hypothesis H3.

Table 8. Moderating effect test.

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

The development of the digital economy can improve firms’ conditions for obtaining market information, conducting product certification, and organizing cross-border transactions, making the product information and ecological attributes of high-value-added forest products easier for markets to recognize and reducing the information and transaction costs associated with entering external markets. Therefore, better digital conditions help improve the value-conversion effect of HVFPE.

5.5 Transmission-channel test

Traditional stepwise tests of mediation effects rely on strong identification assumptions, such as the absence of unobserved confounding between the mediator and the outcome variable. Directly incorporating a mediator that is determined after treatment may also give rise to a “bad control” problem. Based on this consideration, drawing on the approach of Jiang (2022), this study examines the relationship between HVFPE and the upgrading of the forestry industrial structure from the perspective of potential transmission channels. The results in Table 9 show that the estimated coefficient on HVFPE is 0.0148 and is positive and statistically significant at the 5% level, indicating a significant positive relationship between HVFPE and the upgrading of the forestry industrial structure. The expansion of HVFPE is typically accompanied by greater requirements for advanced processing, green certification, and value-chain extension; the upgrading of the forestry industrial structure, in turn, can increase the level of processing and utilization of forest resources and output per unit of resource, while broadening the channels through which forest ecological products enter markets and generate returns. Therefore, the above results provide support for forestry industrial structure upgrading as a potential transmission channel. Hypothesis H4 is supported.

Table 9. Transmission channel test.

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

5.6 Digital economy threshold effect

The baseline regression and interaction-term results indicate that the positive relationship between HVFPE and FEPVRE strengthens as the level of digital economy development increases. Furthermore, the digital economy index is used as the threshold variable. A 5% trimming proportion is applied for grid search, and threshold significance is tested using 500 bootstrap replications. Table 10 shows that the estimated single threshold is 0.2833, with an F-statistic of 130.757 and a bootstrap p < 0.01, significantly rejecting the null hypothesis of no threshold effect. A Wald test was further conducted under the null hypothesis that the coefficients on the two sides of the threshold are equal. The null hypothesis was rejected (p < 0.001), indicating a statistically significant difference between the two coefficients. The critical values at the 10, 5, and 1% levels are 11.442, 13.074, and 17.601, respectively, indicating that the digital-economy threshold is statistically significant. Sample statistics show that the mean and median of the digital-economy index are 0.2133 and 0.2072, respectively, while the 75th percentile is 0.2659 (Figure 3); the threshold value of 0.2833 lies at approximately the 78.49th percentile of the full sample distribution. Among the 279 province–year observations, 219 lie below the threshold and 60 lie above it.

Table 10. Threshold effect of the digital economy.

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

Examining the estimated coefficients on either side of the threshold, when the digital-economy index does not exceed 0.2833, the estimated coefficient on HVFPE is 0.056; when the index exceeds 0.2833, the coefficient rises to 0.062, with both estimates statistically significant at the 1% level. The marginal coefficient on HVFPE above the threshold is approximately 0.006 higher than that below the threshold, corresponding to a relative increase of 10.67%. The results show that at higher levels of digital economy development, the positive relationship between HVFPE and FEPVRE becomes stronger, suggesting a certain degree of complementarity between HVFPE and digital conditions such as digital infrastructure, digital circulation, information matching, and cross-border logistics.

5.7 Heterogeneity analysis

Considering differences in development conditions across regions, this study further examines the regional heterogeneity of HVFPE and the digital economy. Table 11 shows that the estimated coefficients on HVFPE are 0.101, 0.057, and 0.017 in eastern, central, and western China, respectively; all are statistically significant at the 1% level and decline sequentially from east to west. A Chow test is further employed to examine the stability of coefficients across regions, and the results indicate significant regional differences in the coefficient on HVFPE (F = 52.228, p < 0.001). Pairwise comparisons further show that the coefficient differences between eastern and central China, eastern and western China, and central and western China are all statistically significant, indicating a clear regional gradient in the value-conversion effect of HVFPE. The declining HVFPE coefficients from eastern to western regions may be related to regional differences in foreign trade foundations, supporting processing and manufacturing capacity, and market connectivity. The eastern region has a greater concentration of relevant supporting conditions, and HVFPE may therefore exhibit a stronger positive relationship with FEPVRE; the lower estimated coefficient in the western region may indicate greater room for improvement in related industrial and market support. The estimated coefficients on the digital economy are 0.517, 0.770, and 0.816 in eastern, central, and western China, respectively, and all are statistically significant at the 1% level. The Chow test further indicates that the digital-economy coefficients differ significantly overall across the three regions (F = 18.839, p < 0.001). Further pairwise comparisons show that the differences between eastern and central China and between eastern and western China are statistically significant at the 1% level, whereas the difference between central and western China is not statistically significant. The digital economy exhibits larger estimated coefficients in the central and western regions, which may reflect greater marginal scope for digitalization in expanding market coverage, improving information matching, and reducing transaction costs in these regions.

Table 11. Regional heterogeneity test results.

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

Second, based on the median of province-level sample-period means, the sample is divided into high- and low-environmental-regulation groups and high- and low-forest-cover groups. Environmental-regulation intensity is measured by the share of local-government fiscal expenditure related to ecological and environmental governance in regional GDP; forest-resource endowment is measured by forest cover, defined as the share of forest area in total regional land area. Table 12 shows that the estimated coefficients on HVFPE are 0.020 and 0.050 in regions with low and high environmental regulation, respectively, and both are statistically significant at the 1% level. A further Chow test shows that the HVFPE coefficients differ significantly between the two groups (F = 35.791, p < 0.001), indicating that the positive estimated coefficient is significantly larger in the high-environmental-regulation group. By contrast, the estimated coefficients on the digital economy are 0.827 and 0.749 in regions with low and high environmental regulation, respectively, and the Chow test yields a p-value of 0.012, indicating that the difference between the two groups is statistically significant at the 5% level. This suggests that environmental regulation does not weaken the effect of export upgrading but may instead enable HVFPE to translate more effectively into FEPVRE by raising green-compliance requirements, inducing technological improvement, and enhancing the credibility of certification.

Table 12. Heterogeneity analysis by environmental regulation and forest cover.

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

In terms of forest-resource endowments, the estimated coefficients on HVFPE are 0.025 and 0.041 in regions with low and high forest cover, respectively, and both are statistically significant at the 1% level. The Chow test shows a significant difference between the coefficients of the two groups (F = 16.221, p < 0.001), indicating that the positive estimated coefficient is significantly larger in the high-forest-coverage group. Regions with stronger forest resource endowments tend to have more stable raw material supplies, stronger foundations for ecological branding, and greater potential for ecological product development, and are therefore more likely to achieve efficiency gains through HVFPE. The estimated coefficients on the digital economy are 0.893 and 0.694 in regions with low and high forest cover, respectively, and the Chow test indicates a significant difference between the two groups (F = 25.461, p < 0.001). The digital economy exhibits a larger marginal effect in regions with relatively limited forest-resource endowments, possibly because digital technologies strengthen the identification of resource information and reduce information asymmetry and transaction costs, thereby partially alleviating constraints arising from the resource base. Overall, hypotheses H5a, H5b, and H5c are supported.

6 Conclusions and policy implications

Using panel data for 31 Chinese provinces from 2015 to 2023, this study examines the relationships among HVFPE, the digital economy, and FEPVRE. The results show that forest ecological-product value realization depends not only on resource supply, but also on the capacity of the forestry industrial system to capture and convert value and on the conditions of digital development. Both HVFPE and the digital economy are significantly and positively associated with FEPVRE; HVFPE are positively associated with forestry industrial upgrading at the 5% significance level, providing support for the industrial-structure channel; and the interaction-term results show that the positive relationship between HVFPE and FEPVRE becomes stronger as the level of digital economy development increases. This suggests that the realization of forest ecological value does not depend solely on resource endowments, but is also associated with industrial upgrading and improvements in market connectivity.

Further analysis shows that when the digital economy index exceeds 0.2833, the positive estimated coefficient on HVFPE increases further. The heterogeneity results show that this positive relationship is more pronounced in the eastern region, regions with stronger environmental regulation, and regions with higher forest cover, whereas the positive estimated coefficient on the digital economy is larger in the central and western regions and in regions with lower forest cover. These findings indicate that forest ecological product value realization is clearly condition-dependent, with the estimated effects of HVFPE and the digital economy varying according to regional development conditions, environmental governance, and forest resource endowments.

This study extends research on forest ecological-product value realization from within-region factors to external-market value addition and the conditions governing its conversion, establishes a connection between research on forest-product trade and forest ecological-product value realization, and provides new empirical evidence for understanding the market transformation of forest ecological value in an open-economy context.

7 Research contributions, limitations, and prospects

The main contribution of this study lies in further extending the understanding of the factors influencing forest ecological-product value realization. Existing studies have largely explained forest ecological-product value realization from the perspective of within-region conditions, whereas this study incorporates the external-market value added generated by HVFPE and examines its relationship with FEPVRE. On this basis, the study further finds that forestry industrial restructuring, the stage of digital-economy development, and regional resource and environmental conditions all affect how this relationship manifests. Accordingly, this study connects research on forest-product trade with research on forest ecological-product value realization and further demonstrates that the conversion of external-market value added into forest ecological value is condition dependent, providing new empirical evidence for understanding forest ecological-product value realization in an open-economy context.

This study has several limitations that warrant further investigation. Owing to data availability, the analysis relies on provincial panel data to examine the relationships among HVFPE, the digital economy, and FEPVRE. Although two-way fixed effects, lagged variables, and system GMM are employed to mitigate potential endogeneity, observational data do not permit strict causal identification, and the findings should therefore be interpreted primarily as robust associations. Provincial-level data also cannot fully capture firm-level heterogeneity in export structure and digital technology adoption. Future research could combine data on forest product exporting firms, product-level customs transactions, and firm-level digitalization, together with policy shocks or credible instrumental variables, to provide stronger micro-level causal evidence. Some measurement issues also remain. HVFPE is constructed from customs export values based on HS codes, whereas FEPVRE uses ecological product value and related indicators as inputs and gross forestry output value as the output. Although the two measures differ in data sources and measurement objects, common forestry-scale factors may still generate some measurement correlation. Because disaggregated provincial forestry output data are unavailable, potentially overlapping product outputs within gross forestry output value cannot be matched item by item to customs HS classifications. More detailed product- and firm-level data would help further assess and reduce this potential overlap.

Interprovincial forest product trade, logistics networks, and digital platforms may also create spatial interdependence. The Pesaran CD test and Conley spatial HAC standard errors suggest that the baseline results remain robust after accounting for such dependence, but the direction and transmission mechanisms of spatial spillovers are not explicitly identified. Future studies could therefore employ alternative spatial weight matrices and spatial econometric models to examine the spillover effects of forest product trade and the digital economy across provinces. In addition, the digital economy index, constructed from digital infrastructure, digital circulation, digital industries, and digital innovation, captures overall regional digital conditions but only partially reflects trade-specific dimensions such as digital platform use, cross-border transactions, and digital trade services. Developing a forest-product-specific digital trade capacity indicator could provide a more precise assessment of the conditions under which digitalization generates direct and reinforcing effects. Finally, the measurement of FEPVRE depends partly on the monetary valuation of forest ecosystem services. Although multiple ecological product values are estimated according to relevant technical specifications, differences in ecological process models, valuation parameters, and price benchmarks may affect the results. Future research could further validate FEPVRE through alternative valuation approaches, parameter sensitivity analyses, and higher-resolution ecological monitoring data.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Yashwant Singh Rawat, Federal Technical and Vocational Education and Training Institute (FTVETI), Ethiopia

Reviewed by: John Adekunle Adesina, University of Lagos, Nigeria

Hao Wang, Shandong University (Qingdao), China