The Digital Economy Promotes the Coordinated Development of the Non-Timber Forest-Based Economy and the Ecological Environment: Empirical Evidence from China

Abstract

The digital economy offers new solutions for reconciling the growth of the non-timber forest-based economy (NTFBE) with ecological and environmental protection. Utilizing panel data from China’s provinces between 2011 and 2020, this study constructed a comprehensive indicator system for the purpose of examining the coordinated development of the NTFBE and the ecological environment. The employment of a panel of econometric methods, including Tobit models, mediated effects models, spatial Durbin models and threshold regression models, has enabled us to ascertain that the digital economy can effectively drive this coordinated development. The digital economy has a positive spillover effect in neighboring regions, although there is no discernible impact in central and northeastern China. Improvements in human capital and digital infrastructure reinforce this effect. Furthermore, the empowerment of green technology and industrial transformation, as well as the adoption of differentiated development strategies across distinct forest economic models, would be of paramount importance. These findings indicate a necessity for the standardization of the NTFBE. In conclusion, these implications offer novel solutions from China’s forested regions that reconcile socioeconomic growth and environmental protection, thereby fostering the sustainable development of forests.

1. Introduction

It is of paramount importance to achieve a harmonious balance between socioeconomic growth and environmental protection in order to fulfil the United Nations Sustainable Development Goals [1]. In light of the increasing frequency of global extreme weather events and the worsening effects of climate change [2], it is imperative to consider the potential irreversible damage that the previous model of extensive economic growth could have on the ecological environment. Some scholars have found that economic growth affects the quality of the local ecological environment, which in turn affects economic growth [3,4]. Therefore, they have devised a system to evaluate the coupled economy–environment system. The coupling coordination degree model, originally derived from physics, assesses the extent of “econo-my-environment” coupled and coordinated development across diverse regions.

As a green industry with both natural and social attributes [5], the forestry industry must also consider the trade-off between economic growth and environmental protection in the process of industrial restructuring. In recent years, the Chinese government has proposed the concept of an “all-round approach to food” and introduced a series of policies to promote the upgrading and greening of the forestry industry. This has had a significant impact on the formation and development of the non-timber forest-based economy (NTFBE) (non-timber forest-based economy is hereinafter abbreviated as “NTFBE”), facilitating a change in the development model of forest areas. The implementation of the no-deforestation policy has compelled forestry units that previously relied mainly on timber sales to undergo industrial transformation [6]. This has entailed a gradual shift in focus from timber production to the development of non-timber forest products and industries, with the objective of ensuring the continued survival, development and income security of employees [7]. Against this backdrop, China’s provinces have gradually formed a new business model centered on forest foods, while incorporating ecotourism, forest health conservation and other industries. Furthermore, the collective forest tenure reform has significantly enhanced farmers’ enthusiasm for developing the NTFBE, with an increasing number of households demonstrating a willingness to cultivate non-timber forest products on their forest land [8]. Consequently, China’s NTFBE has witnessed a period of accelerated growth.

Despite the Chinese government’s issuance of the “Opinions on Accelerating the Development of the Non-Timber Forest-based Economy” (For details, please see the following: https://www.gov.cn/gongbao/content/2012/content_2201880.htm (accessed on 11 January 2025)) in 2012, which mandates the adherence to ecological principles in the advancement of the NTFBE, the extended growth cycle of trees precludes immediate benefits for forest farmers from these forest resources in the near term [9]. This has resulted in some forest farmers disregarding the principles of sustainable development, resorting to unsuitable cultivation techniques or utilizing inferior fertilizers and pesticides in an effort to reduce costs, and introducing fast-growing non-timber forest products that are detrimental to the environment, posing a significant risk to the ecological integrity of forest areas and, indeed, the entire country [10]. This has also resulted in unsustainable economic development in forested areas, impeding China’s pursuit of modernization and the achievement of the United Nations Sustainable Development Goals [11]. Presently, the Chinese government has yet to issue comprehensive guidelines for the development of the NTFBE or to establish environmentally sustainable standards for different non-timber forest-based economic models that align with ecological environmental protection principles in diverse geographical regions. Therefore, it is imperative to establish a coherent relationship between the NTFBE and the ecological environment, in order to advance environmental protection and enhance the sustainability of the forest industry on a global scale.

Recently, the advent of new-generation digital technologies, intelligent systems and models, including big data, the Internet and cloud computing, has led to the integration of these technologies into forestry economic activities, thereby giving rise to the digital economy. This is also evident in the non-timber forest-based economy [12,13,14]. The advancement of the NTFBE is contingent upon the availability of financial resources [15]. These include intelligent financial service platforms, inclusive agricultural mortgage loans, intelligent machinery equipment subsidies, and the integration of financial resources [16]. These novel digital economy approaches have had a beneficial impact on various facets of the NTFBE, including production, circulation, distribution, and consumption [17]. This is because, to a certain extent, the digital economy has compensated for the shortcomings of the traditional economy [18], providing more convenient and higher-quality financial services to rural populations in forested areas and injecting new impetus into the development of the NTFBE [19].

However, research on the NTFBE is relatively scarce internationally, and the understanding of the concept and connotation of the NTFBE is inconsistent between China and other countries, resulting in an incomplete theoretical foundation and hindering research progress in this field. In 2018, the Chinese Forestry Society issued the “Terminology for non-timber forest-based economy” (T/CSF001-2018), which defines the NTFBE as an economic model that relies on forests, forest areas and their ecological environment, operates according to the principles of sustainability, and engages in environmentally friendly compo-site operations. It mainly includes the cultivation of non-timber forest products, the raising of non-timber forest animals, the collection and processing of related products and the use of forest landscapes [20]. The forest areas suitable for the development of the non-timber forest economy are mainly concentrated in commodity forests, economic forests and some ecologically less important public welfare forests and natural forests [21]. Apart from China, other countries do not have a clear concept of a “non-timber forest-based economy” but rather refer to similar concepts such as agroforestry [22,23], mixed agroforestry [24], non-timber forest products [25] or non-timber forest products [26], all of which refer to environmentally friendly economic models that adhere to the principles of green development and ecological priority and rely on forest lands and their forest ecological environments for composite operations. The majority of existing scholars have demonstrated that the NTFBE can effectively stimulate economic growth in forested areas. However, there is currently no consensus regarding its impact on the environment. Some scholars posit that the advancement of the NTFBE exerts a deleterious influence on the ecological environment [27], whereas others contend that it has a beneficial impact [28]. In general, the academic community has yet to provide clear answers to questions such as “What kind of NTFBE is eco-friendly?” and “What harm can unregulated development models of the NTFBE bring to ecological protection?” Furthermore, existing research has failed to consider the mutual relationship between the NTFBE and the ecological environment and has not assessed the extent of coupled and coordinated development between the two at the national and provincial levels in China.

Overall, the extant literature on the digital economy, the NTFBE and the ecological environment provides a robust foundation for research. Nevertheless, there is a paucity of studies that have investigated the impact and mechanism of the digital economy on a specific system. Furthermore, the academic community has not considered the NTFBE and the ecological environment to be a coupled and coordinated system. Accordingly, based on data from 30 provinces in China, this study initially evaluated the extent of coupled and coordinated development of the NTFBE and the ecological environment in each province, as well as its temporal and spatial evolution. Secondly, the impact of the digital economy on the coupled and coordinated development of the NTFBE and the ecological environment was explored from multiple dimensions. The transmission roles of green technological innovation and industrial structure upgrading in this process were also captured. Thirdly, this paper interpreted this effect from the perspective of regional heterogeneity, employing spatial econometric models to capture the spatial spillover effects of the digital economy on the coordinated development of the NTFBE and the ecological environment. Finally, this study discussed the threshold effects of human capital and digital infrastructure construction, with a view to providing new solutions for the sustainable development of the NTFBE in the context of the construction of digital China and ecological civilization. The specific process and steps are shown in Figure 1.

2. Theoretical Analysis and Research Hypotheses

2.1. The Direct Impact of the Digital Economy on the Coordinated Development of the Non-Timber Forest-Based Economy and Ecological Environment

As a principal driver of economic digitalization and green development, the digital economy exerts a pivotal influence on the expansion of the NTFBE and the enhancement of the ecological environment. Moreover, it effectively orchestrates the integrated growth of these two domains [29,30]. This can be specifically analyzed based on the “production–life–ecology” triple spatial theory, in which the NTFBE and the ecological environment operate [31]. Firstly, the development of new-generation technologies, such as big data, the Internet of Things, and cloud computing, supported by inclusive digital finance, enables enterprises along the non-timber product industrial chain to achieve intelligent production. This is particularly evident in the application of networked monitoring models and high-precision smart devices in non-timber production activities, which effectively controls damage to the ecological environment and additional energy consumption. This, in turn, reduces pollution emissions while increasing the benefits of the NTFBE [32]. Secondly, the NTFBE concept encompasses green consumption scenarios based on forests, including tourism, healthcare, and study tours. Digital technologies have facilitated the development of more precise and efficient growth points for green consumption scenarios. These include forest transportation, carbon credit platforms, and forest tourism planning platforms. Residents’ participation in ecological protection and environmental awareness, as well as their green consumption concepts in the NTFBE, has been enhanced by these platforms. This has effectively reduced environmental pollution and improved resource utilization efficiency [33]. Finally, with regard to the protection of the ecological environment in forest areas, the government is in a position to establish a forest ecological compensation mechanism with greater effectiveness through the digital economy, green finance and inclusive finance [34]. This would facilitate the flow of green finance capital to more efficient and environmentally friendly sectors, while also contributing to the real economy and environmental benefits. It also allows relevant departments to assess the value of forest products, manage resources better and support sustainable non-timber forestry.

Based on the above analysis, our paper proposes the following hypothesis:

H1: 

The digital economy can directly promote the coordinated development of the NTFBE and the ecological environment.

2.2. The Indirect Impact of the Digital Economy on the Coordinated Development of the Non-Timber Forest-Based Economy and Ecological Environment

The digital economy provides the basis for green technological innovation [35]. The absence of an adequate green technological innovation model for the development of the NTFBE results in low efficiency and increased pollution emissions. Furthermore, such a management model often has a negative impact on local biodiversity [36]. By contrast, the reasonable utilization of green technology in non-timber forestry activities has the potential to enhance sustainability, increasing economic benefits while employing technological means to improve resource and energy utilization efficiency, as well as effectively monitoring and avoiding damage to the ecological environment caused by non-timber forestry activities. From the perspective of enterprises operating in the NTFBE, the digital economy can facilitate the availability of a more diverse range of green financing channels, thereby reducing the likelihood of their adopting development models and management modes that are environmentally detrimental, or at least outdated, due to a lack of funds. The application of green innovative technologies and modern information means can facilitate the development of information services for the non-timber circular economy. This can provide professional technical guidance for the development of the non-timber circular economy and promote the adoption of eco-friendly standards and technical specifications for the NTFBE. The extant literature indicates that information asymmetry constrains the capacity of forestry enterprises to secure green funds, while the dearth of green funds impedes the enhancement of innovation efficiency in non-timber forestry activities [37]. The digital economy can help reduce information asymmetry in non-timber forestry credit activities, increasing the chance of green innovation projects aligning with non-timber forestry activities. Forestry enterprises must engage in green technology and business model reform to facilitate the coordinated development of the NTFBE and the ecological environment.

Based on the above analysis, this paper proposes the following hypothesis:

H2: 

The digital economy can promote the coordinated development of the NTFBE and the ecological environment by driving green technological innovation.

The positive impact of industrial structural upgrading on the integrated advancement of economic growth and the ecological environment has been substantiated by scholars in the field [38,39]. In comparison with less developed and environmentally detrimental industries, the forestry industry, as a prototypical green industry, exhibits considerable growth potential and minimal policy risk [40]. Nevertheless, the relatively traditional model of the NTFBE may still result in additional environmental damage and resource and energy waste due to insufficient technology. In contrast, the more advanced model of the NTFBE can achieve growth in a way that minimizes environmental damage and biodiversity loss, thereby creating a more efficient and environmentally friendly integrated non-timber forestry space. In the context of the digital economy, finance not only has its inherent pursuit of profit and risk avoidance but also has the dual attributes of inclusiveness and greenness. This further drives financial institutions to increase investment in emerging technology-intensive and low-carbon industries, thereby bringing about the optimization and upgrading of the forestry industrial structure. The digital economy represents an organic combination of traditional finance with information technology, which is promoting the widespread application of financial technology in related green finance scenarios of the NTFBE [41]. New technologies support the creation of green financial products, stimulating the supply and innovation of green finance for non-timber forestry products [42]. More green financial products could make people more aware of green issues, encouraging saving and spending in a greener way. This could boost consumption in the NTFBE. Furthermore, it may prompt consumers to pay closer attention to the extent to which environmental protection standards are met in non-timber forestry. During the growth phase of the NTFBE, the demand for green consumption and the supply of green finance attracts greater social capital into related NTFBE sectors, facilitating the greening of non-timber forestry activities.

Based on the above analysis, this paper proposes the following hypothesis:

H3: 

The digital economy can promote the coordinated development of the NTFBE and the ecological environment by facilitating industrial structural upgrading.

The specific theoretical analysis framework is shown in Figure 2.

3. Research Design

3.1. Model Setting

In light of the theoretical analysis and the fact that the degree of coupling and coordination between the NTFBE and the ecological environment falls between 0 and 1, the dependent variable can be classified as a censored dependent variable. The application of ordinary least squares or panel models for the estimation of parameters may result in biased and inconsistent parameter estimates [43]. Accordingly, to address the issue of the censored dependent variable, a Hausman test was conducted and a panel Tobit model with two-way fixed effects at the individual-time level was established as the benchmark regression model for estimating the impact of the digital economy on the coordinated development of the NTFBE and the ecological environment. Furthermore, this model allows for a more effective examination of the discrepancies among samples. The primary model is as follows:

$$\left\{\begin{array}{l}{\mathit{CD}}_{it}={\alpha }_0+{\alpha }_1{\mathit{DE}}_{it}+{\alpha }_i{X}_{it}+u_i+v_t+{\epsilon }_{itit},{\mathit{CD}}_{it}>0\hfill \\ 0,{\mathit{CD}}_{it}⩽0\hfill \end{array}\right.$$

(1)

In Equation (1), ${\mathit{CD}}_{it}$ is the restricted explained variable, representing the level of coordinated development between the NTFBE and the ecological environment. ${\alpha }_0$ is the constant term. ${\mathit{DE}}_{it}$ is the core explanatory variable, indicating the level of digital economy development in various regions, and its coefficient ${\alpha }_1$ reflects the impact of the digital economy on the level of coordinated development between the NTFBE and the ecological environment. $X_{it}$ represents a series of control variables, which are other factors that may affect the level of coordinated development between the NTFBE and the ecological environment, and ${\alpha }_i$ is the coefficient of each control variable. Furthermore, province fixed effects $u_i$ and year fixed effects $v_t$ are introduced to eliminate the influence of other factors varying across provinces and over time. ${\epsilon }_{it}$ is the random error term, which follows an independent and normal distribution.

3.2. Date Sources

In order to examine the impact of the digital economy on the integrated growth of the NTFBE and the ecological environment, this paper has primarily drawn upon panel data from 31 Chinese provinces over a 10-year period, from 2011 to 2020. However, due to data availability constraints and other considerations, the study has not included the Hong Kong, Macao, and Taiwan regions. The data pertaining to socioeconomic development variables have been derived from the “China Statistical Yearbook”, “the China Forestry and Grassland Statistical Yearbook”, and provincial bulletins. The data pertaining to natural geographic variables were obtained through the utilization of zonal statistics in ArcGIS 10.8. In particular, the value of ecological product services is derived from the “Spatial Distribution Dataset of Ecosystem Service Value in China’s Terrestrial Areas” (For details, please see the following: https://www.resdc.cn/DOI/doi.aspx?DOIid=48 (accessed on 11 January 2025)), a dataset provided by the Institute of Geographic Sciences and Natural Resources Research at the Chinese Academy of Sciences. The Human Activity Footprint is derived from the “Global Human Activity Footprint Distribution Map” (For details, please see the following: https://www.x-mol.com/groups/li_xuecao/news/48145 (accessed on 11 January 2025)), as calculated by Mu et al. [44]. The ecological environment quality is derived from the “High-Resolution Historical Ecological Environment Quality Dataset of China (2001–2021)” (For details, please see the following: http://www.geodata.cn/data/datadetails.html?dataguid=190747515712302&docid=0 (accessed on 11 January 2025)). The spatial resolution is adjusted to 30 m. Missing data were replaced using interpolation techniques.

3.3. Variable Description

3.3.1. Dependent Variable

The dependent variable is the level of coordinated development between the NTFBE and the environment (CD). This study has selected a number of representative indicators to construct an indicator system for the coordinated development of the NTFBE and the ecological environment [45]. A system is constructed from the dimensions of input and output, considering the economic characteristics of the NTFBE [46]. The “Pressure–State–Impact–Response” theory states that the ecological environment is constructed from four dimensions: pressure, state, impact, and response [47]. The NTFBE and the ecological environment are interlinked. The objective entropy weight method is used to assign weights to indicators in the NTFBE and the ecological environment. Table 1 shows the specific indicator names, definitions and assigned weights. The coupling coordination model (CCM) is a model used to calculate the degree of coupling and coordination between multiple systems through mathematical expression, reflecting the interaction of the parts. The degree of coupling and coordination is calculated to quantify the developmental level of each system and reflect the relationship and degree of interaction between two or more systems. This methodology is widely applied in the fields of resource and environmental protection and regional economic development [48]. The following steps are to be undertaken:

$$C_{ii}=2\times \sqrt{{\displaystyle \frac{\left(F\times E\right)}{{\left(F+E\right)}^2}}}$$

(2)

$$T_{it}=\alpha F+\beta E$$

(3)

$${\mathit{CD}}_{it}=\sqrt{C_{it}\times T_{it}}$$

(4)

In Equation (2), i and t represent province and time, respectively. $C_{ii}$ is the degree of coupling between the NTFBE and the ecological environment, and F and E are the comprehensive evaluation indices of the NTFBE system and the ecological environment system, respectively. In Equation (3), $T_{it}$ is the comprehensive coordination index between the NTFBE and the ecological environment, and $\alpha$ and $\beta$ represent the contribution rates of the NTFBE system and the ecological environment system to their coordinated development, respectively. Following the approach of existing scholars, both are set to 0.5 [49]. In Equation (4), ${\mathit{CD}}_{it}$ is the calculated degree of coupling and coordination between the NTFBE and the ecological environment (the calculation of the composite index was standardized using the extreme variance method for the indicators involved in both systems).

Table 1. Indicator frameworks of NTFBE system and ecological environment quality system.

System	Primary Indicator	Secondary Indicator	Indicator Attribute	Indicator Weight (%)
NTFBE System	Inputs	Agricultural, Forestry, Animal Husbandry, and Fishery Employees (10,000 people)	+	13.275
		Value of Ecological Product Services (10,000 RMB)	+	25.147
		Forest Stock Volume (100 million cubic meters)	+	23.666
		Forestry Land Area (10,000 hectares)	+	13.944
	Outputs	Understory Economic Output Value (10,000 RMB)	+	9.317
		Forest Tourism Revenue (10,000 RMB)	+	7.733
		Average Wage of Agricultural, Forestry, Animal Husbandry, and Fishery Employees (RMB)	+	6.919
Ecological Environment Quality System	Pressure	Total Wastewater Discharge (10,000 tons)	−	3.439
		Sulfur Dioxide Emissions (10,000 tons)	−	3.732
		Human Activity Footprint	−	13.480
	State	Afforestation Area (1000 hectares)	+	22.930
		Forest Coverage Rate (%)	+	16.080
	Impact	Total Carbon Dioxide Emissions (10,000 tons)	−	3.278
		Ecological Environment Quality	+	9.530
	Response	Investment Completed in Industrial Pollution Control (10,000 RMB)	+	27.531

Table 1. Indicator frameworks of NTFBE system and ecological environment quality system.

System	Primary Indicator	Secondary Indicator	Indicator Attribute	Indicator Weight (%)
NTFBE System	Inputs	Agricultural, Forestry, Animal Husbandry, and Fishery Employees (10,000 people)	+	13.275
		Value of Ecological Product Services (10,000 RMB)	+	25.147
		Forest Stock Volume (100 million cubic meters)	+	23.666
		Forestry Land Area (10,000 hectares)	+	13.944
	Outputs	Understory Economic Output Value (10,000 RMB)	+	9.317
		Forest Tourism Revenue (10,000 RMB)	+	7.733
		Average Wage of Agricultural, Forestry, Animal Husbandry, and Fishery Employees (RMB)	+	6.919
Ecological Environment Quality System	Pressure	Total Wastewater Discharge (10,000 tons)	−	3.439
		Sulfur Dioxide Emissions (10,000 tons)	−	3.732
		Human Activity Footprint	−	13.480
	State	Afforestation Area (1000 hectares)	+	22.930
		Forest Coverage Rate (%)	+	16.080
	Impact	Total Carbon Dioxide Emissions (10,000 tons)	−	3.278
		Ecological Environment Quality	+	9.530
	Response	Investment Completed in Industrial Pollution Control (10,000 RMB)	+	27.531

3.3.2. Core Explanatory Variable

The level of development of the digital economy is indicated by the index aggregate (IA). Given the industrial nature of the NTFBE and the compensation mechanism required for ecological environment protection, the principal channel through which the digital economy exerts an influence is green and inclusive financial services, which are characterized by a dual nature. Accordingly, this paper adopts the China Digital Inclusive Finance Index, jointly compiled by the Digital Finance Research Center of Peking University and Ant Financial Group (For details, please see the following: https://www.idf.pku.edu.cn/yjcg/zsbg/513800.htm (accessed on 11 January 2025)), as a measure of the development level of the digital economy [50]. The paper also uses coverage breadth (CB), usage depth (UD), and digitization level (DL) to represent the three dimensions of the digital economy: breadth of coverage, depth of usage, and degree of digitization, respectively.

3.3.3. Control Variables

In order to mitigate the impact of potential omitted-variable bias, this study refers to existing research and selects a series of factors that may affect the coordinated development of the NTFBE and the ecological environment as control variables. For further details, please see references [51,52]. With regard to socioeconomic factors, the following variables have been included in the analysis as control variables: Population Density (PD), Urbanization Level (UL), Per Capita Regional Gross Domestic Product (GP), Forestry Finance-Related Rate (FR), and Government Behavior (GB). In light of the potential influence of natural environmental factors on the development of forest resources and changes in the ecological environment [53], the study also controls for average annual precipitation (AP) and average annual temperature (AT).

3.3.4. Mediating Variables

In order to capture the mechanism of the digital economy driving the coordinated development of forest economy and ecological environment, this paper mainly selected two mediating variables. One is Industrial Structure Upgrade, which is mainly measured by the proportion of the sum of the output value of the secondary and tertiary industries in the total output value of forestry with reference to existing studies [54]; the other is Green Technology Innovation, which is measured by taking the number of green-related invention patents granted at the provincial level plus one and subtracting the logarithm.

3.3.5. Threshold Variables

According to the research of existing scholars [55], there may be a nonlinear relationship and a certain threshold effect in the process of the digital economy, driving a win–win situation of economic growth and environmental protection. Therefore, this paper selects two threshold variables—Human Capital Level and Level of Digital Infrastructure—to investigate whether there is a threshold effect in the process of the digital economy driving the coordinated development of the forest economy and ecological environment.

Descriptive statistics of the relevant variables are shown in

Table 2. Selection and reference sources of other variables.

Type	Symbol	Indicator	Indicator Calculation Method	Mean	Min	Max
Dependent	CD	Coupling Coordination Degree of Understory Economy and Ecological Environment	Computed Actual Value	0.562	0.337	0.768
Core Explanatory	IA	Level of Digital Economy Development	Computed Actual Value by Peking University Digital Finance Research Center	216.2352	16.22	431.9276
	CB	Breadth of Digital Economy Coverage	Computed Actual Value by Peking University Digital Finance Research Center	196.6696	1.96	397.0019
	UD	Depth of Digital Economy Usage	Computed Actual Value by Peking University Digital Finance Research Center	211.1211	6.76	488.6834
	DL	Level of Digital Economy Digitization	Computed Actual Value by Peking University Digital Finance Research Center	290.1421	7.58	462.2278
Controls	PD	Population Density	Logarithm of Population Density	7.858	5.361	8.669
	UL	Urbanization Level	Urban Population/Annual Permanent Resident Population (Ten Thousand People)	0.565	0.222	0.942
	GP	Per Capita Regional Gross Domestic Product	Regional Gross Domestic Product/Total Population at Year-end (Yuan/Person)	53,483.697	16,165.000	164,889.000
	FR	Forestry Finance-Related Rate	Ratio of the Sum of Forestry Loans and Deposits at the End of the Year to GDP	0.001	0.000	0.009
	GB	Government Behavior	Ratio of General Budgetary Expenditure of Local Finance to Regional GDP	0.000	0.000	0.001
	AP	Average annual precipitation	Average annual precipitation (m)	0.003	0.001	0.006
	AT	Average annual temperature	Average annual temperature (°C)	11.961	−4.022	25.079
Intermediate	US	Industrial Structure Upgrade	(Output Value of Forestry Secondary Industry + Output Value of Forestry Tertiary Industry)/Total Forestry Output Value	0.556	0.003	1.555
	GI	Green Technology Innovation	Number of Green Invention Patent Applications	7.300	1.099	10.382
Threshold	HC	Human Capital Level	Proportion of Regular Higher Education School Students	0.019	0.008	0.039
	DI	Level of Digital Infrastructure	Proportion of Internet Broadband Access Users (Household)	0.192	0.041	0.436

.

4. Empirical Results and Analysis

4.1. Spatiotemporal Analysis of the Coordinated Development Level of the Non-Timber Forest-Based Economy and Ecological Environment

From the perspective of subsystems, as illustrated in Figure 3, the comprehensive indices of China’s non-timber forest-based economy (FE) and ecological environment (EE) systems have demonstrated relatively stable trends over time. The comprehensive index of the NTFBE system reached its highest point in 2015 but subsequently declined to its lowest point in 2020. By the conclusion of the observation period (2020), a slight decrease was noted in comparison to the baseline period (2011). The decline observed in 2019 may have been attributable to the impact of the global pandemic caused by the SARS-CoV-2 virus, which led to an overall economic downturn and a consequent decrease in efforts towards environmental protection. This had a negative effect on the balance between the NTFBE and the ecological environment. Furthermore, the comprehensive index of the ecological environment system also demonstrated an approximate “M”-shaped trajectory over time, exhibiting a gradual increase from the base period and reaching its highest point in 2015. Thereafter, it exhibited a gradual decrease before increasing again from 2018 and eventually stabilizing. In conclusion, the end of the sample period exhibited a slight increase in comparison to the base period.

With regard to the evolution of the interrelationship between China’s NTFBE and its ecological environment, as illustrated in Figure 4, the overall trajectory exhibited an inverted “U”-shaped pattern, initially rising from 2012, reaching its peak in 2016, and subsequently declining on an annual basis. However, at the conclusion of the observation period, there was a notable surge in comparison to the initial baseline. This may be associated with “Opinions on Accelerating the Development of the Non-timber Forest-based Economy” being published by the General Office of the State Council of China in 2012. The initial impact of the policy was a surge in enthusiasm among forest peasants, the rapid entry of social capital, and a smooth industry transformation. The negative impact of the NTFBE on the surrounding ecological environment was not yet apparent. Indeed, the emerging NTFBE even helped to improve the ecological environment to some extent, with many barren forests achieving both qualitative and quantitative growth under the management of forest peasants. However, over time, in pursuit of rapid income growth and low input costs, many regions adopted inappropriate production structures that caused significant ecological damage. This resulted in irreversible environmental pollution of soil, forests, and groundwater, which severely threatened the regional ecological environment and biodiversity. Furthermore, it caused a continuous decline in the level of coupling and coordination between the NTFBE and the ecological environment. In light of the above, it is imperative to address how to standardize the development of the NTFBE and promote its coordinated development with the ecological environment. This is a pressing issue with significant implications for the high-quality development of the NTFBE and the construction of an ecological civilization.

From the provincial perspective, as illustrated in Figure 5, Figure 6, Figure 7, Figure 8, Figure 9 and Figure 10, the sample base period (2011) and the end of the sample period (2020) were selected as two typical years from the 2011–2020 sample period. The intensity and distribution of the comprehensive indices of the NTFBE system and the ecological environment system across Chinese provinces were then plotted. Furthermore, the intensity and distribution of the coordinated development level of the NTFBE and the ecological environment across Chinese provinces were plotted, thus providing a more intuitive representation of the differences between regions and the changes within the same region over different years.

In particular, during the base period, Yunnan Province exhibited the highest level of coordinated development, with a score of 0.729, which falls within the intermediate coordination level of (0.7, 0.8]. Furthermore, Yunnan ranked first among the 31 provinces in terms of its comprehensive indices for the NTFBE system and the ecological environment system. This may be attributed to the fact that Yunnan was relatively early in developing the NTFBE, with activities commencing in 2008. Following the damage caused by quarrying, numerous “prospectors” in the NTFBE were granted access to areas for the purpose of ecological restoration, which involved tree planting and moderate development. These activities were encouraged and supported by governments at all levels. The reasonable planting of fruit trees yielded both ecological and economic benefits. However, during the base period, 11 provinces exhibited a state of near-imbalance (0.4, 0.5], with no provinces falling below this level. Among them, Tianjin and Shanghai are clearly decoupled from the development of their non-agricultural industrial bases and ecological protection due to their disparate economic status. They show a regressive trend from near imbalance in 2011 (0.4, 0.5) to mild imbalance in 2020 (0.3, 0.4), with real values declining markedly from year to year.

4.2. Baseline Regression Results and Analysis

A two-way fixed-effects panel Tobit model was used to perform a baseline regression. Model (1) and Model (2) use digital inclusive finance as a proxy for the digital economy. The two models differ in that the latter includes control variables. Model (3) to Model (8) use three digital economy dimensions as explanatory variables, with and without control variables.

Table 3. Baseline regression results using the Tobit model.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	CD	CD	CD	CD	CD	CD	CD	CD
IA	0.0003 ***	0.0004 ***
	(0.0001)	(0.0001)
CB			0.0004 ***	0.0003 **
			(0.0001)	(0.0001)
UD					0.0001 ***	0.0002 **
					(0.0000)	(0.0001)
DL							0.0001 ***	0.0001 ***
							(0.0000)	(0.0000)
PD		0.0008		0.0021		−0.0018		−0.0004
		(0.0073)		(0.0074)		(0.0147)		(0.0072)
UL		0.0674		0.0881		0.0277		0.0734
		(0.0616)		(0.0631)		(0.1243)		(0.0585)
GP		−0.0000 ***		−0.0000 ***		−0.0000 *		−0.0000 ***
		(0.0000)		(0.0000)		(0.0000)		(0.0000)
FR		4.6792		4.8217		5.5054		4.3414
		(3.6544)		(3.6949)		(5.5824)		(3.5908)
GB		−3.6733		398.7791		−183.2289		−373.0040
		(556.2638)		(552.1291)		(766.8561)		(551.7414)
AP		5.9966 *		6.4116 *		5.4514		4.6075
		(3.3238)		(3.3411)		(4.7861)		(3.3004)
AT		0.0025		0.0036		0.0029		0.0020
		(0.0022)		(0.0023)		(0.0025)		(0.0021)
sigma_u	0.0921 ***	0.0789 ***	0.0941 ***	0.0844 ***	0.0901 ***	0.0743 ***	0.0888 ***	0.0774 ***
	(0.0118)	(0.0120)	(0.0122)	(0.0129)	(0.0116)	(0.0134)	(0.0114)	(0.0115)
sigma_e	0.0277 ***	0.0267 ***	0.0278 ***	0.0268 ***	0.0263 ***	0.0267 ***	0.0270 ***	0.0262 ***
	(0.0012)	(0.0012)	(0.0012)	(0.0012)	(0.0017)	(0.0019)	(0.0011)	(0.0011)
Constant	27.5592 ***	21.1162 ***	32.2095 ***	21.4760 **	0.5530 ***	0.5490 ***	15.7494 ***	8.4948 ***
	(6.0821)	(6.0059)	(8.5648)	(8.5353)	(0.0173)	(0.1349)	(2.1599)	(3.0261)
Observations	310	310	310	310	310	310	310	310
Number of ID	31	31	31	31	31	31	31	31
ProvinceFE	YES	YES	YES	YES	YES	YES	YES	YES
YearFE	YES	YES	YES	YES	YES	YES	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

shows that the digital economy and its three dimensions have a positive impact on the coordinated development of the NTFBE and the ecological environment, regardless of whether control variables are included. This confirms the robustness of the results. Enhancing the digital economy facilitates the coordinated advancement of the NTFBE and the ecological environment.

Based on the above analysis, research hypothesis H1 is supported.

4.3. Analysis of the Mechanism

The baseline regression model demonstrated that the advancement of the digital economy can facilitate the integrated growth of the non-timber forest-based enterprise (NTFBE) and the ecological environment. Nevertheless, the precise mechanism by which this occurs remains to be elucidated. In light of the aforementioned theoretical analysis and hypotheses, it is plausible to suggest that green technological innovation and industrial structure upgrading may serve as potential conduits through which the digital economy fosters the coordinated development of the NTFBE and the ecological environment. To further investigate whether the digital economy can influence the coordinated development of the NTFBE and the ecological environment through these two channels, this paper followed established practices and treated the mediating variables as the explained variables in the following model [56]:

$$M_{it}={\beta }_0+{\beta }_1{\mathit{DE}}_{it}+{\beta }_i{X}_{it}+u_i+v_t+{\epsilon }_{it}$$

(5)

In Equation (5), M represents the mediating variables, namely green technological innovation (GI) and industrial structure upgrading (US). The other symbols are interpreted similarly to the baseline regression. The regression results are shown in

Table 4. Results of the test for the effect of green technology innovation.

Variables	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)
	GI	GI	GI	GI	GI	GI	GI	GI
IA	0.0063 ***	0.0087 ***
	(0.0024)	(0.0030)
CB			0.0179 ***	0.0151 ***
			(0.0030)	(0.0030)
UD					0.0061 ***	0.0031 **
					(0.0004)	(0.0015)
DL							0.0044 ***	0.0011 ***
							(0.0002)	(0.0004)
PD		−0.0376		0.0383		−0.0499		−0.1407 *
		(0.0730)		(0.0724)		(0.0740)		(0.0733)
UL		2.0510 ***		1.3703 **		2.1980 ***		7.5475 ***
		(0.6374)		(0.6466)		(0.6298)		(0.8808)
GP		−0.0000		−0.0000		−0.0000		0.0000 ***
		(0.0000)		(0.0000)		(0.0000)		(0.0000)
FR		−85.0803 **		−86.3109 **		−79.3652 **		−102.0939 **
		(36.0907)		(35.1462)		(36.2562)		(44.3704)
GB		−37,167.0586 ***		−32,596.0277 ***		−38,279.2425 ***		−9552.4715
		(5391.3294)		(5334.2810)		(5390.3119)		(14,483.3145)
AP		88.5626 **		71.3215 **		86.6257 **		135.5786 ***
		(36.5120)		(35.4879)		(36.8388)		(46.8138)
AT		0.0206		0.0257		0.0210		0.1169 **
		(0.0198)		(0.0194)		(0.0196)		(0.0485)
Constant	5.9754 ***	6.4985 ***	5.6154 ***	5.2728 ***	6.0114 ***	6.0202 ***	6.0180 ***	1.4183 *
	(0.2271)	(0.6962)	(0.2363)	(0.7333)	(0.0799)	(0.7545)	(0.0654)	(0.7945)
Observations	310	310	310	310	310	310	310	310
Number of ID	31	31	31	31	31	31	31	31
ProvinceFE	YES	YES	YES	YES	YES	YES	YES	YES
YearFE	YES	YES	YES	YES	YES	YES	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

and

Table 5. Results of the test for the effect of forestry industry upgrading.

Variables	(17)	(18)	(19)	(20)	(21)	(22)	(23)	(25)
	US	US	US	US	US	US	US	US
IA	0.0003 ***	0.0003 **
	(0.0000)	(0.0001)
CB			0.0003 ***	0.0004 ***
			(0.0000)	(0.0001)
UD					0.0003 ***	0.0003 ***
					(0.0000)	(0.0001)
DL							0.0002 ***	0.0002 ***
							(0.0000)	(0.0000)
PD		0.0044		0.0083		−0.0334 **		0.0002
		(0.0134)		(0.0142)		(0.0164)		(0.0122)
UL		0.0686		−0.0054		−0.2393 ***		0.1216
		(0.0915)		(0.0934)		(0.0686)		(0.1013)
GP		0.0000		−0.0000		−0.0000		0.0000
		(0.0000)		(0.0000)		(0.0000)		(0.0000)
FR		−17.3417 ***		−16.9117 ***		−16.8165 ***		−18.6486 ***
		(5.0566)		(4.9514)		(2.1259)		(5.4759)
GB		−1880.6590		−1938.2091		−1149.0551 ***		−2229.1404
		(1410.5906)		(1300.0852)		(250.6382)		(1419.1534)
AP		9.2024 *		8.3452				6.6775
		(5.2518)		(5.0185)				(4.7018)
AT		−0.0017		−0.0030				−0.0044
		(0.0098)		(0.0090)				(0.0084)
Constant	0.0941 ***	0.0836	0.0970 ***	0.1240	0.1032 ***	0.5428 ***	0.0908 ***	0.1196
	(0.0104)	(0.1462)	(0.0096)	(0.1409)	(0.0101)	(0.1407)	(0.0105)	(0.1197)
Observations	310	310	310	310	310	310	310	310
R-squared	0.290	0.320	0.312	0.341	0.212	0.149	0.277	0.332
Number of ID	31	31	31	31	31		31	31
ProvinceFE	YES	YES	YES	YES	YES	YES	YES	YES
YearFE	YES	YES	YES	YES	YES	YES	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

.

Table 4 presents the regression results for Model (9) and Model (10), which examine the influence of the digital economy on the mediating variable (GI) with and without control variables. Model 11 to Model 16 show how the three digital economy dimensions affect the mediating variable. The digital economy positively influences green technology, with statistical significance levels of 1% or 5%. The digital economy can advance the NTFBE and the environment via green technology. The digital economy could provide the financial support needed for green technology in forestry, particularly for NTFBE. It can also monitor ecological damage caused by NTFBE and predict non-point source pollution, using advanced technologies for early prevention [19]. The digital economy can digitally intervene in NTFBE activities, including pre-production, production, and post-production. This can boost productivity, optimize resources, prevent waste and achieve environmental protection and a circular forestry economy.

Based on the above analysis, research hypothesis H2 is supported.

Table 5 shows the regression results for Model (17) and Model (18), which examine the influence of the digital economy on the mediating variable (US) with and without control variables. Model 19 to Model 25 show the impact of the digital economy on industrial upgrading. The digital economy and its three dimensions have a positive impact on industrial upgrading at the 1% and 5% levels of significance. This suggests that the digital economy can facilitate the coordinated development of the NTFBE and the ecological environment through the industrial structure [57]. The digital economy supports new businesses and modernizes traditional forestry. The digital economy offers new opportunities for forestry operators, leading to an innovative forestry model. This model includes forest mushrooms, medicine, poultry and fisheries. It unites the space above, within and below the forest, ensuring that there is no contradiction between the new cash crops and the original trees. Growing new cash crops alongside the original trees increases the forest’s yield. This facilitates the integrated growth of the forest economy and the ecological environment. The forest economy model uses forest landscapes to facilitate the transition of forestry into the tertiary sector, resulting in the establishment of eco-landscape-based management frameworks [58]. This extends the forest industry chain while maintaining the ecological balance, increasing added value and achieving a mutually beneficial balance between economic prosperity and environmental sustainability.

Based on the above analysis, research hypothesis H3 is supported.

4.4. Regional Heterogeneity Analysis

In the baseline regression, this study examined the positive effect of the digital economy on the overall coordinated development of the NTFBE and the ecological environment. The digital economy’s impact on the NTFBE and ecological environment may differ across China due to variations in digital infrastructure. Furthermore, existing scholars have corroborated through cross-regional comparative studies that there may be considerable regional disparities in the impact of the digital economy on the coordinated development of economic growth and the ecological environment [59,60]. Consequently, when discussing the role of the digital economy in promoting the coordinated development of the NTFBE and the ecological environment, relevant departments may not only consider the overall effect but may also conduct further comparative studies across different regions.

Table 6. Results of the test for regional heterogeneity.

Variables	(26)	(27)	(28)	(29)
	Eastern	Central	Western	Northeastern
	CD	CD	CD	CD
IA	0.0003 **	0.0002	0.0005 ***	0.0001
	(0.0001)	(0.0003)	(0.0002)	(0.0003)
PD	−0.0734 ***	0.0348	0.0058	0.0252
	(0.0234)	(0.0220)	(0.0083)	(0.0389)
UL	0.1143	0.4806 ***	0.0316	−0.0667
	(0.0958)	(0.1321)	(0.1116)	(0.1880)
GP	−0.0000 **	0.0000	−0.0000	0.0000
	(0.0000)	(0.0000)	(0.0000)	(0.0000)
FR	17.9245	9.8140	14.1908 **	20.4983
	(11.2864)	(22.7422)	(6.1853)	(15.3218)
GB	−2625.3826 **	−2378.9164	−441.2238	985.8270
	(1246.2582)	(2822.8017)	(794.2988)	(2435.4066)
AP	−1.1685	22.6851 ***	8.4460	−1.5977
	(4.7388)	(5.9983)	(8.3711)	(19.6439)
AT	0.0058	0.0045	0.0042	−0.0176 *
	(0.0041)	(0.0030)	(0.0035)	(0.0099)
sigma_u	0.0588 ***	0.0000	0.0672 ***	0.0000
	(0.0216)	(0.0000)	(0.0148)	(0.0048)
sigma_e	0.0166 ***	0.0293 ***	0.0274 ***	0.0254 ***
	(0.0018)	(0.0027)	(0.0019)	(0.0033)
Constant	22.2711 **	32.2757 *	29.5966 **	0.7758
	(9.6191)	(16.9734)	(12.2252)	(20.5273)
Observations	100	60	120	30
Number of ID	10	6	12	3
ProvinceFE	YES	YES	YES	YES
YearFE	YES	YES	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the regression results of the regional heterogeneity analysis based on the regional classification by the National Bureau of Statistics of China (The eastern region includes Beijing, Tianjin, Shanghai, Hebei Province, Shandong Province, Jiangsu Province, Zhejiang Province, Fujian Province, Guangdong Province, Hainan Province; the central region includes: Shanxi Province, Henan Province, Hubei Province, Anhui Province, Hunan Province, Jiangxi Province; the western region includes Inner Mongolia Autonomous Region, Xinjiang Uygur Autonomous Region, Ningxia Hui Autonomous Region, Shaanxi Province, Gansu Province, Qinghai Province, Chongqing, Sichuan Province, Guangxi Zhuang Autonomous Region, Guizhou Province, Yunnan Province; the northeastern region includes Heilongjiang Province, Jilin Province, Liaoning Province. (Due to data limitations, this study does not include the Hong Kong, Macao, Tibet and Taiwan regions)). The models indicated by (26), (27), (28), and (29) represent regressions utilizing samples from the eastern, central, western, and northeastern regions, respectively. The digital economy exerts a positive influence on the coordinated development of the NTFBE and the ecological environment in the eastern region, with a 5% level of statistical significance, and in the western region, with a 1% level of statistical significance. Nevertheless, the positive effects of the digital economy are not statistically significant in the central and northeastern regions. Three primary reasons contribute to the observed regional heterogeneity.

Firstly, the disparity in economic development levels and industrial structures is a significant factor. The eastern region, characterized by a higher level of economic development, boasts a more advanced secondary and tertiary forestry industry sector. Consequently, the development of digital economy is better positioned to integrate with these industries, thereby enhancing the growth of the NTFBE. In contrast, the western region, where the primary forestry industry holds a larger share, benefits from the digital economy’s ability to augment resource utilization efficiency and foster management innovation, positively influencing the NTFBE. However, the northeastern and central regions may not experience the same level of impact due to their economic development constraints and industrial structure limitations.

Secondly, the distribution of forest resources and ecological conditions vary across regions. The eastern region, with its abundant forest resources and high forest coverage, is well-suited for the digital economy to facilitate the rational exploitation of these resources and promote a diversified forest economy. Conversely, the western region, despite its rich forest resources, faces relatively harsh ecological conditions. Here, the digital economy plays a crucial role in enhancing the green technology innovation capabilities of the forestry industry, thus positively affecting the NTFBE. In contrast, the northeastern and central regions may not see as significant a boost from the digital economy in the forest economy due to the uneven distribution of forest resources or ecological limitations.

Lastly, the unbalanced degree of digitalization and technological innovation capabilities across regions is another contributing factor. The eastern region, with its higher digitalization levels and robust technological innovation capabilities, is better equipped for the digital economy to drive innovation and transformation within the NTFBE. While the western region may lag in these areas, the development of digital economy has the potential to improve digitalization and technological innovation, thereby exerting a positive impact on the forest economy. Therefore, the interplay between economic development, resource distribution, ecological conditions, and digitalization levels results in a complex and region-specific effect of the digital economy on the NTFBE.

4.5. Robustness Tests

To further examine the robustness of the positive effect of the digital economy on the coordinated development of the NTFBE and the ecological environment, this study employed two methods for robustness testing. The initial method employed was that of winsorization. In light of the potential sensitivity of the research results to outliers, this paper conducted empirical regressions after winsorizing the top and bottom 5% of all variables included in the baseline regression samples. The results are presented in

Table 7. Robustness check results (trimmed at 5% on both tails, with replacement).

Variables	(30)	(31)	(32)	(33)
	CD	CD	CD	CD
IA	0.0004 ***
	(0.0001)
CB		0.0004 ***
		(0.0001)
UD			0.0002 **
			(0.0001)
DL				0.0001 ***
				(0.0000)
PD	−0.0001	0.0021	−0.0035	−0.0016
	(0.0111)	(0.0112)	(0.0200)	(0.0110)
UL	0.0998 *	0.1017 *	0.0611	0.1300 **
	(0.0596)	(0.0614)	(0.1342)	(0.0568)
GP	−0.0000 ***	−0.0000 ***	−0.0000 *	−0.0000 ***
	(0.0000)	(0.0000)	(0.0000)	(0.0000)
FR	14.9840 **	16.4815 **	10.0188	14.1036 **
	(6.8183)	(6.8544)	(10.0503)	(6.7598)
GB	−563.7928	−225.7899	−44.6131	−730.8612
	(738.1056)	(732.5130)	(1051.5134)	(733.1472)
AP	6.1656 *	6.8180 **	4.7298	4.9042
	(3.2751)	(3.2816)	(4.7057)	(3.2706)
AT	0.0019	0.0027	0.0034	0.0018
	(0.0021)	(0.0021)	(0.0025)	(0.0021)
sigma_u	0.0764 ***	0.0786 ***	0.0777 ***	0.0780 ***
	(0.0113)	(0.0116)	(0.0140)	(0.0114)
sigma_e	0.0263 ***	0.0264 ***	0.0265 ***	0.0259 ***
	(0.0011)	(0.0011)	(0.0019)	(0.0011)
Constant	18.8774 ***	21.2729 ***	0.5373 ***	6.4180 *
	(5.8330)	(7.5434)	(0.1688)	(3.2834)
Observations	310	310	310	310
Number of ID	31	31	31	31
ProvinceFE	YES	YES	YES	YES
YearFE	YES	YES	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

. The regression results presented in Model (30) to Model (33) demonstrate that the digital economy, when considered in its three dimensions, continues to exert a positive influence on the coupling coordination degree of the NTFBE and the ecological environment, even after winsorization. These results indicate that the digital economy can still have a positive impact on the coordinated development of the NTFBE and the ecological environment at the 1% or 5% significance level. The second method employed was regression testing utilizing the GMM two-stage model. In line with existing studies, this study selected the average digital economy development level of each province (I1) as the instrumental variable to address the potential bidirectional causality between the coupling coordination degree and the digital economy. Moreover, the results are presented in

Table 8. Robustness check results (two-stage GMM regression).

Variables	(34)	(35)	(36)	(37)
	GMM (I1)	GMM (I1)	GMM (I1)	GMM (I1)
	CD	CD	CD	CD
IA	0.0010 ***
	(0.0004)
CB		0.0011 **
		(0.0004)
UD			0.0015 ***
			(0.0004)
DL				0.0003 ***
				(0.0001)
PD	0.0055	0.0096	−0.0015	0.0056
	(0.0082)	(0.0084)	(0.0135)	(0.0081)
UL	−0.2265 ***	−0.2410 ***	−0.2175 ***	−0.2048 ***
	(0.0440)	(0.0472)	(0.0822)	(0.0423)
GP	−0.0000 ***	−0.0000 ***	−0.0000 ***	−0.0000 ***
	(0.0000)	(0.0000)	(0.0000)	(0.0000)
FR	5.2908 **	3.9238	5.2266	5.2315 **
	(2.5417)	(2.6907)	(3.6501)	(2.5244)
GB	−2898.0410 ***	−2840.7771 ***	−2715.8190 ***	−2994.6396 ***
	(265.3172)	(274.0595)	(430.7149)	(266.3799)
AP	43.1881 ***	45.2916 ***	39.6292 ***	44.7398 ***
	(3.4715)	(3.5867)	(5.9060)	(3.4356)
AT	−0.0102 ***	−0.0103 ***	−0.0101 ***	−0.0097 ***
	(0.0011)	(0.0011)	(0.0017)	(0.0010)
Constant	64.4937 ***	75.9244 ***	0.6342 ***	18.6617 **
	(23.4145)	(28.4548)	(0.1410)	(9.0482)
Observations	310	310	310	310
R-squared	0.653	0.632	0.596	0.658
ProvinceFE	YES	YES	YES	YES
YearFE	YES	YES	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05.

, which shows the results of the GMM two-stage regression model, which was used to replace the bidirectional fixed-effects model. Model (34) to Model (37) present the results obtained from the GMM two-stage regression using I1 as the instrumental variable. These results indicate that when I1 is used as the instrumental variable and the model is replaced, the digital economy and its three dimensions continue to exert a positive influence on the coordinated development of the NTFBE and the ecological environment at the 1% or 5% significance level. The aforementioned methods serve to illustrate the robustness of the model results.

4.6. Endogeneity Discussion

In light of the potential endogeneity issues that may arise from omitted variables and bidirectional causality in this study, this paper employed the IV-Tobit model as a means of alleviating endogeneity. The specific results of this approach are presented in

Table 9. Instrumental variable results for addressing endogeneity (two-stage OLS regression).

Variables	(38)	(39)
	First	Second
	IA	CD
IA		0.0006 ***
		(0.000)
I1	1.0529 ***
	(1.065)
PD	13.7993 **	0.0070
	(5.722)	(0.008)
UL	0.0008 ***	−0.2081 ***
	(0.000)	(0.042)
GP	−19.4461	−0.0000 ***
	(333.266)	(0.000)
FR	−89,584.9563 **	6.8271 ***
	(35,864.763)	(2.417)
GB	2178.3532 ***	−2822.4421 ***
	(473.979)	(258.998)
AP	0.4118 ***	42.5778 ***
	(0.139)	(3.480)
AT	−3.5095 ***	−0.0093 ***
	(0.952)	(0.001)
athrho2_1	−0.0207 (0.064)
lnsigma1	−2.8682 *** (0.040)
lnsigma2	2.0582 *** (0.040)
Constant	7004.3232 ***	24.1986 *
	(1913.384)	(12.433)
Observations	310	310
ProvinceFE	YES	YES
YearFE	YES	YES

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

. The instrumental variable (I1) is found to be highly correlated with the potential endogenous variable at the 1% significance level, and the p-values of the AR and Wald tests are both less than 0.005, effectively ruling out the possibility of weak instrumental variables. The results of the endogeneity discussion indicate that there is a significant endogeneity issue between the digital economy and the coordinated development of the NTFBE and the ecological environment. Following the application of the IV-Tobit method to address the endogeneity issue, the digital economy continues to exert a positive influence on the coordinated development of the NTFBE and the ecological environment, with a statistical significance level of 1%. However, the regression coefficient (0.0006) is significantly larger than the regression coefficient (0.0003) in the baseline regression, indicating that the baseline regression model may have underestimated the positive impact of the digital economy on the coordinated development of the NTFBE and the ecological environment.

5. Further Analysis

5.1. Analysis of the Spatial Effects of the Digital Economy Affecting the Coordinated Development of the Non-Timber Forest-Based Economy and the Ecological Environment

To understand how the digital economy affects nearby areas and the environment, this paper added digital economy, ecological environment coordination, and control variables to a baseline regression model. This resulted in a spatial Durbin model (SDM), shown below.

$$\left\{\begin{array}{l}{\mathit{CD}}_{it}={\alpha }_0+\rho W\times D_{it}+{\varphi }_{1}W\times {\mathit{DE}}_{it}+{\alpha }_1{\mathit{DE}}_{it}+{\varphi }_{i}W\times X_{it}+{\alpha }_i{X}_{it}+u_i+v_t+{\epsilon }_{it},{\mathit{CD}}_{it}>0\hfill \\ 0,{\mathit{CD}}_{it}⩽0\hfill \end{array}\right.$$

(6)

In Equation (6), the variable ρ represents the spatial autoregressive coefficient, W is the spatial weight matrix, λ and θ are the coefficients of the core explanatory variable and control variable spatial cross-product terms, respectively. The remaining symbols are consistent with those used in the baseline regression. This study employed geographic and economic distance matrices to reflect the disparities between provinces. Based on this, the study evaluated the spatial autocorrelation of the digital economy and the coordination of the NTFBE and the ecological environment using the global Moran’s I method.

Table 10. Results of the Moran’s I index test for spatial autocorrelation.

Variables	W1 (Geographic Weighting Matrix)					W2 (Economic Weighting Matrix)
	I	E(I)	sd(I)	z	p-Value *	I	E(I)	sd(I)	z	p-Value *
y2011	0.113	−0.033	0.039	3.709	0.000 ***	0.113	−0.033	0.092	1.580	0.057 *
y2012	0.106	−0.033	0.039	3.58	0.000 ***	0.121	−0.033	0.091	1.689	0.046 **
y2013	0.087	−0.033	0.039	3.078	0.001 ***	0.164	−0.033	0.092	2.141	0.016 **
y2014	0.125	−0.033	0.039	4.082	0.000 ***	0.190	−0.033	0.091	2.455	0.007 ***
y2015	0.107	−0.033	0.039	3.591	0.000 ***	0.177	−0.033	0.092	2.293	0.011 **
y2016	0.086	−0.033	0.039	3.074	0.001 ***	0.201	−0.033	0.091	2.576	0.005 ***
y2017	0.085	−0.033	0.039	3.039	0.001 ***	0.217	−0.033	0.091	2.748	0.003 ***
y2018	0.082	−0.033	0.039	2.992	0.001 ***	0.169	−0.033	0.090	2.239	0.013 **
y2019	0.082	−0.033	0.039	2.976	0.001 ***	0.166	−0.033	0.091	2.191	0.014 **
y2020	0.070	−0.033	0.039	2.649	0.004 ***	0.159	−0.033	0.091	2.098	0.018 **

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

shows that the Moran’s I of China’s NTFBE coupling coordination degree with the ecological environment fluctuates between 0.070 and 0.217 and is significant. Both the digital economy and the coordination of the NTFBE and the ecological environment exhibit significant spatial autocorrelation, implying a clustering effect in their spatial distribution.

Considering the economic nature of the digital economy and NTFBE [61], this paper selected W2 (economic geographic matrix) as the main matrix for spatial econometric analysis. The Wald and LR tests on the SDM model showed it did not degenerate into a SEM or SAR. The Hausman test showed that a spatial and time-period fixed-effects model should be used. The spatial and time-period fixed-effects SDM model was selected as the primary model for spatial effect regression, with results in Model (41).

Table 11. Regression results for the spatial model of the coordinated development of NTFBE and ecological environment quality affected by digital economy.

Variables	(41)
	SDM
	CD
IA	0.002 ***
	(0.000)
PD	0.005
	(0.008)
UL	−0.276 ***
	(0.044)
GP	0.000 ***
	(0.000)
FR	13.845 ***
	(2.454)
GB	−2357.537 ***
	(260.091)
AT	−0.009 ***
	(0.001)
AP	33.494 ***
	(3.657)
W2 × IA	0.006 ***
	(0.001)
W2 × PD	0.085 ***
	(0.031)
W2 × UL	−0.198
	(0.165)
W2 × GP	0.000 ***
	(0.000)
W2 × FR	87.008 ***
	(15.988)
W2 × GB	2.002
	(739.044)
W2 × AT	−0.001
	(0.004)
W2 × AP	−47.998 ***
	(13.075)
ρ	−0.258 ***
	(0.112)
Log- L	493.038
Observations	310
Number of ID	31
R-squared	0.825
ProvinceFE	YES
YearFE	YES

Standard errors in parentheses *** p < 0.01.

demonstrates that, under the SDM model specification, the coefficient of the digital economy remains significantly positive, as does the coefficient of the interaction term between the digital economy and the spatial economic distance matrix. This indicates that, after accounting for the potential errors introduced by spatial positive correlation, the digital economy continues to exert a positive influence on the coordinated advancement of the NTFBE and the ecological environment, with a statistical significance level of 1%. This further substantiates the reliability of the preceding regression outcomes. However, the spatial autocorrelation coefficient ρ is significantly negative, indicating the potential for negative interactions in the coordinated development of the NTFBE and the ecological environment across regions. This suggests that, after controlling for other variables, the coordinated development of the local NTFBE and ecological environment may have an inhibitory effect on the coordinated development of the surrounding areas’ NTFBEs and ecological environments, resulting in a “siphoning effect”. This may be attributable to the heterogeneity in resource endowments, geographic features, and climate change across regions, which is consistent with the conclusions drawn from the regional heterogeneity analysis.

The SDM model has limitations in capturing the digital economy’s impact on the NTFBE and the environment. This paper used the partial derivative method to decompose the spatial effects into direct, indirect, and total, with the results displayed in

Table 12. Effects decomposition of the coordinated development of NTFBE and ecological environment quality affected by digital economy.

Variables	Direct	Indirect	Total
IA	0.001 ***	0.004 ***	0.006 ***
	(0.000)	(0.001)	(0.001)
PD	0.001	0.068 ***	0.069 ***
	(0.007)	(0.025)	(0.026)
UL	−0.267 ***	−0.098	−0.365 **
	(0.044)	(0.134)	(0.151)
GP	0.000 ***	0.000 **	0.000 ***
	(0.000)	(0.000)	(0.000)
FR	10.401 ***	70.163 ***	80.564 ***
	(2.948)	(15.432)	(16.678)
GB	−2364.553 ***	571.793	−1792.760 ***
	(267.780)	(550.833)	(574.343)
AT	−0.009 ***	0.001	−0.008 **
	(0.001)	(0.003)	(0.003)
AP	35.439 ***	−47.613 ***	−12.173
	(35.439)	(10.390)	(11.598)

Standard errors in parentheses *** p < 0.01, ** p < 0.05.

. The direct effect shows how the digital economy and control variables affect the local province’s NTFBE and environment. The indirect effect shows how the surrounding provinces’ digital economy and control variables affect the local province’s NTFBE and ecological environment. The total effect shows how the digital economy and control variables affect the development of the NTFBE and the ecological environment across all sample provinces. The results show that the digital economy has a positive effect on the local and surrounding provinces’ NTFBEs and ecological environments. This may be attributed to the pioneering demonstration effect of the digital economy, which transcends the spatial and temporal constraints of conventional economic models, creating a conducive environment for regional collaborative development and the formation of economies of scale, thereby propelling the coordinated advancement of the surrounding areas’ NTFBEs and ecological environments. Furthermore, the forestry industry exerts considerable externalities, and the digital economy has accelerated the flow of green, technological, and digital factors across regions, thereby activating the indirect effects on the coordinated development of surrounding areas.

5.2. Analysis of the Threshold Effects of the Digital Economy Affecting the Coordinated Development of the Non-Timber Forest-Based Economy and the Ecological Environment

The results of the regional heterogeneity and spatial effect analyses indicate that the positive impact of the digital economy on the coordinated development of the NTFBE and the ecological environment is contingent upon the local digital ecosystem. The principal factors influencing the regional digital ecosystem are the level of human capital (DI) and the extent of digital infrastructure development (HC). In light of the aforementioned studies, this study constructed a panel threshold model to capture the varying characteristics of the digital economy that may exist in promoting the coordinated development of the NTFBE and the ecosystem, due to the disparate levels of human capital and digital infrastructure construction. To further test the hypothesis that there are threshold effects based on the level of human capital and digital infrastructure construction in the relationship between digitalization and the coordinated development of the NTFBE and the ecological environment, this study constructed the following threshold effect model:

$$\left\{\begin{array}{c}{\mathit{CD}}_{it}={\theta }_i+{\beta }_1{\mathit{DE}}_{it}\times I\left(q_{it}⩽\gamma \right)+{\beta }_2{\mathit{DE}}_{it}\times I\left(q_{it}⩾\gamma \right)+{\beta }_i{X}_{it}+{\mu }_i+v_t+{\epsilon }_{it},{\mathit{CD}}_{it}>0\\ 0,{\mathit{CD}}_{it}⩽0\end{array}\right.$$

(7)

In Equation (7), $q_{it}$ represents the threshold variable, i.e., the level of human capital or digital infrastructure construction, $\gamma$ is a specific threshold value, and $I\left(·\right)$ is an indicator function segmented based on different threshold values. The other symbols are consistent with the baseline regression.

Firstly, this study tested whether there is a threshold effect between digitalization and the coordinated development of the NTFBE and the ecological environment and determined the number of threshold values. The test results are shown in

Table 13. Results of the threshold effect test for the impact of digital economy on the coordinated development of NTFBE and ecological environment quality.

Threshold	HC							DI
	RSS	MSE	Fstat	Prob	Crit10	Crit5	Crit1	RSS	MSE	Fstat	Prob	Crit10	Crit5	Crit1
Single	0.0834	0.0005	15.65	0.0500	13.5613	15.3215	19.9252	0.0638	0.0004	19.66	0.0133	14.0739	16.3644	21.3417
Double	0.0796	0.0004	8.68	0.2900	13.682	16.5668	20.8984	0.0613	0.0004	6.05	0.5767	24.9910	30.9251	40.0539
Triple	0.0745	0.0004	12.43	0.2933	19.1493	21.9486	26.6939	0.0585	0.0004	7.21	0.4567	18.1251	25.0079	42.5717

, indicating that the level of human capital and the level of digitalization infrastructure construction both pass the single-threshold test.

Table 14. Single-threshold regression results for the impact of digital economy on the coordinated development of NTFBE and ecological environment.

Variables	(42)	Variables	(43)
	HC		DI
	CD		CD
0_IA	0.0002 **	0_IA	0.0001
(HC ≤ 0.0196)	(0.0001)	(DI ≤ 0.0543)	(0.0001)
1_IA	0.0003 ***	1_IA	0.0002 **
(HC > 0.0196)	(0.0001)	(DI > 0.0543)	(0.0001)
PD	−0.0360	PD	−0.0354
	(0.0382)		(0.0363)
UL	0.2577	UL	0.1335
	(0.1567)		(0.1518)
GP	−0.0000	GP	−0.0000
	(0.0000)		(0.0000)
FR	29.0180	FR	57.0241
	(41.4932)		(39.3815)
GB	777.9704	GB	−868.9390
	(2428.8002)		(2324.2681)
AP	0.0033	AP	−0.0000
	(0.0093)		(0.0089)
AT	11.1648	AT	5.5536
	(13.0307)		(12.5105)
Constant	10.5308 ***	Constant	−0.0354
	(2.3008)		(0.0363)
Observations	310	Observations	310
Number of ID	31	Number of ID	31
R-squared	0.121	R-squared	0.194

Standard errors in parentheses *** p < 0.01, ** p < 0.05.

shows that human capital and digital infrastructure have a single-threshold effect. From model (42), it can be observed that when human capital exceeds the threshold value, the impact of digital economy on NTFBE and the ecological environment increases, with a positive impact at the 5% or 1% significance level. One potential explanation is that the positive effects of the digital economy can only be sustained as human capital accumulates and agglomerates. This is because both the transformation of the forestry industry structure and forestry technological innovation require the support of high-quality, knowledgeable, and technically skilled forestry talents. Consequently, the presence of a greater number of skilled professionals tends to facilitate the advancement of the NTFBE towards a more sophisticated level of development. Furthermore, educational and training programs can effectively cultivate an enhanced ecological awareness among operators. The dissemination and establishment of sustainable development concepts are inextricably linked to the accumulation of human capital, thereby contributing to the coordinated development of the NTFBE and the ecological environment. Additionally, from Model (43), it can be observed that when the level of digital infrastructure is below the threshold value (0.0543), the promoting effect of the digital economy on the coordinated development of the NTFBE and the ecological environment is not significant. Only when the threshold is exceeded does the digital economy have a positive effect on the NTFBE and ecology, at the 5% level. One potential explanation is that increased Internet penetration may facilitate the adoption of digital technologies in non-timber forestry production and operations. This could enable foresters and other forestry professionals to access information and resources online, including standardized non-timber forestry operational procedures, green non-timber forestry techniques, and insights into ecological civilization. Furthermore, the promotion of eco-friendly forestry and online financial services would facilitate the accumulation of start-up capital and enhance efficiency in production, and it would also lead to a reduction in pollution. Conversely, the level of digital infrastructure is insufficient to provide the basic protection necessary to safeguard the effects of the digital economy [62].

6. Conclusions and Implications

6.1. Conclusions

In the context of achieving the Sustainable Development Goals, a critical issue is how to reconcile economic growth with environmental sustainability. Following the implementation of a series of forest protection measures, such as the Natural Forest Protection Programme, economic entities that previously relied on timber harvesting for income may no longer benefit from the traditional forestry model or may receive only meager ecological compensation funds. The development of the NTFBE can facilitate the transformation of the economic model in forest areas. However, irrational production structures, planting methods and overexploitation, such as the introduction of high-value exotic species without considering the impact on biodiversity, or the use of non-green technologies without preserving the spatial integrity of forests, can seriously threaten the ecological environment, which would limit the harmonious development of the NTFBE and the ecological environment. The digital economy can effectively improve the drawbacks of the traditional economy, break through the limitations of time and space, help new forestry economic entities develop green forest management technologies, and promote the continuous transformation of the forestry industry structure, thereby achieving a green, advanced and technological forest economy model and forming a situation of the coordinated development of the NTFBE and the ecological environment.

At the theoretical level, this paper elucidated the direct and indirect mechanisms through which the digital economy influences the harmonious development of the NTFBE and the ecological environment. First, based on provincial-level balanced panel data in China from 2011 to 2020, this paper measured the development levels of the NTFBE system and the ecological environment system for China as a whole and for each province during the sample period; applied the coupling coordination degree model to calculate the development level of coordination between the NTFBE and the ecological environment; and used ArcGIS software to depict the temporal and spatial development across provinces. Second, the paper constructed a two-way fixed-effects model to estimate the impact of the digital economy on the coordinated development of the NTFBE and the ecological environment, as well as its mechanisms. Finally, the paper empirically investigated the spatial spillover and threshold effects of the influence of the digital economy on the coordinated development of the NTFBE and the ecological environment through spatial econometric models and panel threshold models. The NTFBE system, the ecological environment system, and the coupling coordination between the NTFBE and the ecological system all showed a trend of rapid increase followed by a slow decline. This suggests that the damage caused by the NTFBE to the ecological environment was not apparent in the early period of the study. However, in the later period of the sample, due to the influence of many factors such as the lack of formulation of relevant standards of the NTFBE, the long cycle of forestry activities, insufficient stamina for forestry scientific and technological innovation, and excessive start-up costs for forestry industry transformation, unreasonable, inappropriate and unscientific forest economic management models have emerged in large numbers. Some forest farmers lost enthusiasm, abandoning the management of the NTFBE. Others adopted traditional, low-cost and quick-return non-standard management, damaging the ecological environment and decoupling the growth of the NTFBE from the environment. Achieving the sustainable development of the NTFBE is difficult. The digital economy has a positive effect on the NTFBE and the environment. Green technology and the forest industry are key to this. The impact varies by region, with the greatest effect seen in the east and west. These findings are due to regional differences in labor, digital infrastructure and industry. The results show that the digital economy has a significant positive impact on the NTFBE and the ecological environment. The digital economy has a positive effect at the local level and on neighboring provinces. The results of the threshold effect model indicate that both human capital and digital infrastructure have a single-threshold effect on the digital economy’s impact on the NTFBE and the ecological environment.

6.2. Implications

In light of the aforementioned research conclusions, this study puts forth the following policy recommendations:

Firstly, developing and improving a standardized management system for NTFBE business activities is critical. The current level of coupled and coordinated development of the NTFBE and the ecological environment has gradually decreased, indicating that the two have not yet reached a benign relationship of mutual promotion, which has also constrained the independent development of both. The key problem lies in the lack of effective standardized guidance, which has led to the failure of business entities around the world to adopt standardized and green technical regulations for understory planting, breeding, and collection. Kenya’s agroforestry development project combines digital infrastructure to promote economic development in smallholder farming communities. The project uses mobile technology to provide market access, real-time weather data and mobile payment systems to help farmers increase the value of non-timber forest products such as honey, herbs and medicinal plants. Therefore, it may be beneficial to comprehensively assess and dynamically manage the ecological and economic benefits of existing NTFBE models. Based on the characteristics of different regions, models, and stakeholders, a series of standardized production systems could be formulated, including requirements such as standards for product raw materials, composite products, production and processing equipment, and ecological environment standards.

Secondly, promoting the rational application of digital technology in the NTFBE and environmental management in forest areas appears to be essential. The utilization of novel technological tools, such as blockchain, big data, and cloud computing, could optimize the market layout of the NTFBE on Internet platforms. This may help overcome time and space constraints, reduce wasteful resource utilization and environmental pollution, and facilitate the integration of the digital economy with the NTFBE and its green development. The Food and Agriculture Organization’s Forest Observation and Management System (FOMS) monitors forest resources and forest products through big data and cloud computing, which can optimize forest management and reduce forest degradation and pollution. Furthermore, encouraging social capital to invest in forest areas could support the creation of leading enterprises in the NTFBE sector, promote the reallocation of resources to high-value-added industries, and contribute to the establishment of industrial bases for forest planting, forest recreation and tourism, and forest research and study. By enhancing the digital literacy and green consumption concepts of those involved in forestry management, large-scale, high-efficiency, high-technology, and environmentally friendly NTFBE enterprises and cooperative organizations may emerge, ultimately forming a sustainable forest-based economy industry cluster. The process of digital transformation could enhance the competitiveness and sustainability of the global NTFBE sector.

Thirdly, improving digital infrastructure and cultivating human capital could be important. A differentiated policy design at the regional level may be needed. Policies on the NTFBE should take into account regional development, consumer needs, and forest resources. This approach could foster the development of local, high-value forest-based economic activities and help create local industries and brands. The talent pool in forest areas could be enhanced to attract individuals with dual competence in ecological protection and digital technology. A new service model combining inclusive and traditional finance could be established to provide financing, employment, and entrepreneurial opportunities for forest farmers and new forestry economic entities. This model could also encourage returnees to participate in NTFBE activities, providing essential human resources for industrial development. Furthermore, enhancing digital infrastructure remains crucial. For example, the research of Chen et al. shows that Internet technology upgrades have a positive impact on forestry green total factor productivity, which indicates that the application of digital technology can optimize the utilization efficiency of forestry resources and environmental management [63]. Integrating broadband and service assurance systems could facilitate an innovative marketing and operational model, such as “Internet+NTFBE”, potentially propelling the development of industrial network platforms and fostering a conducive environment for the intelligent advancement of the NTFBE.

6.3. Research Limitations

Despite the fact that our study has evaluated the spatiotemporal evolution trend of the current coordinated development of NTFBE–ecological system coupling in China and various provinces through the analysis of panel data from the provincial level in China; obtained the role path of the digital economy in promoting the coordinated development of the NTFBE and the ecological environment; and put forward corresponding policy recommendations that can be promoted, there is still some room for further research. Firstly, the granularity is large. It is difficult to obtain data below the provincial level. Therefore, this study only analyses at the provincial level. It is difficult to more accurately capture the driving role of the digital economy in the coordinated development of the forest economy and ecological environment in a smaller dimension. Further studies could obtain data with smaller granularity in the future, testing the existing results. Secondly, due to the limitation of data availability, the indicator system constructed in this paper still has room for improvement. Especially in the forest economy, this study can only construct relevant indicator systems from the two dimensions of input and output. Relevant departments can strengthen the data collection and collation of the entire process of forest economy management activities, supplement relevant data, and increase statistical indicators to support future research on the forest economy. Finally, this study is more concerned with the inference of causal effects between the digital economy and the coupled coordination of the systems, and more tests and analyses other than the level of significance can be considered to be added when conducting future analyses on prediction.