Forestry Resource Efficiency, Total Factor Productivity Change, and Regional Technological Heterogeneity in China

Abstract

The efficient and sustainable management of forestry resources is crucial in ensuring economic and societal sustainability. The Chinese government has invested significantly in regulations, afforestation, and technology to enhance the forest resource efficiency, reduce technological disparities, and boost productivity growth. However, the success level of this undertaking is unclear and worth exploring. To this end, this study applied DEA-SBM, meta-frontier analysis, and the Malmquist productivity index to gauge the forest resource efficiency (FRE), regional technology heterogeneity (TGR), and total factor productivity growth (MI) in 31 Chinese provinces for a study period of 2001–2020. Results revealed that the average FRE was 0.5430, with potential growth of 45.70%, to enhance the efficiency level in forestry resource utilization. Anhui, Tibet, Fujian, Shanghai, and Hainan were found to be the top performers in forestry utilization during the study period. The southern forest region was ranked highest, with the highest TGR of 0.915, indicating advanced production technologies. The average MI score was 0.9644, signifying a 3.56% decline in forestry resource productivity. This deterioration is primarily attributed to technological change (TC), which decreased by 5.2%, while efficiency change (EC) witnessed 1.74% growth over the study period. The Southern Chinese forest region, indicating an average 3.06% increase in total factor productivity, ranked highest in all four regions. Guangxi, Tianjin, Shandong, Chongqing, and Jiangxi were the top performers, with prominent growth in MI. Finally, the Kruskal–Wallis test found a significant statistical difference among all four regions for FRE and TGR.

1. Introduction

Forests provide livelihoods, protect watersheds, sustain biodiversity, and play an essential role in carbon sequestration and oxygen production. Forests are interdependent elements of the global ecosystem, influencing the climate, weather, and environmental health [1]. Preserving and managing forests sustainably is essential in addressing local and global challenges, such as ensuring clean water and biodiversity and mitigating climate change [2]. Industrial growth’s positive and negative effects on the ecosystem and forests are diverse and far-reaching. The technological advancements brought about by industrialization have enhanced forest management practices. Due to industrial innovation, sustainable logging techniques, reforestation efforts, and eco-friendly materials have been developed [3]. However, industrial expansion has also posed a significant environmental challenge. Expanding factories, infrastructure, and urban areas frequently result in deforestation and habitat loss, which endangers forest biodiversity. Both plant and animal species within forest ecosystems are negatively impacted by industrial pollution, such as air and water pollution [4].

In addition, industrialization has contributed to climate change by emitting greenhouse gases, which contribute to destabilizing forest ecosystems by altering temperature and precipitation patterns. This can potentially affect forest species’ distribution and viability [5]. Furthermore, the demand for basic materials from industrial sectors can lead to the overexploitation of forest resources, resulting in the degradation and loss of vital ecosystems. Striking a balance between industrial development and environmental preservation is a difficult task that requires careful planning, sustainable practices, and a commitment to protecting our forests and the larger ecosystem [6]. The UN has promoted afforestation and reforestation for 20 years to combat climate change and preserve biodiversity, increasing global forest cover (see Figure 1). Programs like CDM and REDD+ in developing countries promote tree cultivation and sustainable forest management [7]. International agreements like the Bonn Challenge and the New York Declaration on Forests set ambitious global goals for the restoration of degraded landscapes and cessation of deforestation, emphasizing the crucial role of afforestation and reforestation in reducing greenhouse gas emissions and safeguarding ecosystems [8,9].

With the fifth largest forest area globally, China’s forests have experienced complex challenges in recent years. Significant efforts have been made to increase the forest cover through afforestation and reforestation initiatives, resulting in substantial gains in forested areas [10]. However, many of these new forests are monoculture plantations that lack the biodiversity and resilience of natural forest ecosystems. China also grapples with illegal forestry, urbanization, and infrastructure development, contributing to deforestation and habitat loss. Due to industrial activities, air pollution, water contamination, and soil degradation further strain forest ecosystems. In addition, the effects of climate change, such as increased temperatures and altered precipitation patterns, threaten the health of forests [11,12].

The Chinese government has taken several measures to address these issues, such as enforcing stricter logging restrictions, promoting sustainable forest management, and making significant investments in projects focused on ecological restoration. China faces a significant challenge in balancing economic growth with protecting and restoring its various forest ecosystems [13]. For numerous important reasons, forestry efficiency is of paramount importance. First, effective forestry practices contribute to the sustainable management of forest resources, thereby preventing overexploitation and environmental destruction. Second, it maximizes economic benefits by optimizing timber production, improving the livelihoods of forest-dependent communities, and bolstering forest-related industries. Thirdly, increased efficiency reduces the carbon footprint of forestry operations, thereby contributing to climate change mitigation. Furthermore, effective forestry practices play a role in biodiversity conservation, safeguarding clean water sources, and maintaining crucial ecosystems. In essence, the efficiency of forestry operations is vital in achieving a harmonious balance between environmental preservation, economic growth, and social well-being [14,15].

Moreover, optimizing the forest efficiency relies on considering regional variations in production technology. Climate, ecosystems, and infrastructure influence the choice of methods. The diversity in forest cover and resources across various regions of the country should be considered. Modifying technology to the local conditions maximizes resource utilization and supports sustainable forest management in different regions [16]. Dynamic growth in forestry resource efficiency is also a powerful indicator of consistent growth in resource-optimized utilization. The Chinese government has significantly improved the forestry efficiency and dynamic productivity in recent years. These initiatives acknowledge the importance of forests for ecological preservation, carbon sequestration, and sustainable resource management [17]. China has implemented modern forest monitoring and management technologies, including remote sensing and geographic information systems (GIS). To combat deforestation and enhance forest health, the government has promoted sustainable logging practices, reforestation, and afforestation projects [18]. In addition, stringent regulations have been implemented to combat illegal harvesting and encourage responsible forest management. The government’s focus on green finance and investments in ecological restoration initiatives further underscore its commitment to enhancing forestry efficiency and sustainability in China. These extensive efforts are consistent with global environmental objectives and contribute to China’s role as a custodian of its forest resources [19].

Nevertheless, despite these commendable efforts by the Chinese government, the extent of its success in achieving forestry efficiency, assessing the total factor productivity change and understanding the regional technological disparities in different regions of China, remains largely unexplored. It represents a crucial research gap and is worth investigating. To this end, our study contributes to the existing literature in several ways. In the first stage, we employ DEA-SBM on the data of China’s 31 mainland provinces and municipalities from 2001 to 2020 to gauge the forestry efficiency, providing insights into how effectively these provinces utilize their forest resources. We evaluate the forestry efficiency over the study period and investigate the growth or decline in forestry efficiency in China. We also explore the level of success in the Chinese government’s mission of forestry efficiency growth over an extensive period of 20 years. In the second stage, the study includes a meta-frontier analysis to assess production technology discrepancies across four distinct Chinese regions, namely Northeast, Northern, Southern, and Southwest China, exploring the regional heterogeneity in technological advancements within the forestry sector. Thirdly, the study uses the Malmquist productivity index to measure changes in total factor productivity (TFP) over time, allowing us to determine whether efficiency improvements or technological changes primarily drive shifts in productivity growth. We also explore the dynamic change (year to year) in forestry efficiency. This can help policymakers and forest management authorities on regional and national levels to identify potential opportunities to enhance forest resource utilization. Productivity growth could be achieved by either strengthening the technical efficiency in forest resource utilization or acquiring the advanced technological capabilities used in forest resource conversion. Finally, to strengthen the study’s results, the Kruskal–Wallis test is used to estimate the significant statistical differences among the forestry efficiency, total factor productivity change, and technology gap ratio in four different regions of China. The rest of the study is distributed as follows. Section 2 presents the comprehensive literature review. Section 3 and Section 4 contain the Materials and Methods and Results and Discussion sections. Section 5 presents the conclusions and policy implications of the study.

2. Literature Review

Data envelopment analysis (DEA) is a powerful tool to assess the efficiency of forestry operations in different countries and regions. Chen et al. [20] stress the necessity of evaluating the forestry eco-efficiency, considering ecological preservation and economic output conflicts. They employ a three-stage DEA model with the DEA-Malmquist approach, finding strong influences from internal practices and external factors. Environmental variables impact the input parameters differentially. The study identifies absolute and conditional convergence in Chinese forestry eco-efficiency, implying catch-up effects in less developed provinces. Młynarski et al. [21] assess the Southern Polish Forest district’s efficiency using conventional data envelopment analysis and the Tobit econometric model. They cover 113 forest districts in four Regional Directorates of State Forests from 2008 to 2012. The economic and financial resource efficiency vary across districts, with higher efficiency in lowland districts. External factors have significant impacts, such as the population density affecting financial efficiency in lowlands and the population density and forest complexes negatively affecting the economic resource efficiency. Financial effectiveness correlates positively with forest complexes in the highlands but adversely with nature reserve areas. These characteristics have limited effects on the highland economic resource efficiency. Forest-based enterprises embrace bio-based economic strategies for renewable resource utilization. Bont et al. [22] stress the need for efficient forest management for biodiversity and ecosystem services, including timber production and carbon sequestration. Modern harvesting technologies could make the forestry business in Switzerland more cost-effective, especially in difficult terrain, resulting in economic gains and enhanced harvesting feasibility. The spatial decision support system estimates the best harvesting methods based on productivity models and expert-defined decision trees. It shows a 12% increase in the economically viable harvested forest area if applied nationwide. Further, numerous studies have used DEA to evaluate the forestry efficiency in different regions and countries [23,24,25,26,27,28,29].

Similarly, Li et al. [30] argue that climate change mitigation becomes more important as forests are valued for their non-timber benefits and biomass carbon sequestration. In Changbai Mountain, Northeast China, the forest type, age class, and establishment mode considerably affect the net primary productivity (NPP), evapotranspiration (ET), and water use efficiency (WUE). Coniferous woods have higher NPP and WUE but lower ET than broadleaved forests, indicating the importance of planning and management for temperate carbon sequestration and water conservation. Moreover, Jin et al. [31] examine the changes in China’s forestry industry, stressing integrated development and productivity for sustainable growth. Research using many analytical methodologies shows that the integration of the forestry industry has increased, with most provinces at moderate to medium–high levels. The total factor productivity in the forestry industry has improved, primarily due to technology. According to a study, forestry industry integration boosts total factor production by improving the pure technical efficiency and spatial spillover effects on nearby regions. These findings emphasize the importance of integrated forestry industry development, regional collaboration, and successful regional forestry growth policies. Lin and Ge [32] emphasize forestry’s role in climate change mitigation and regional economic output. They analyze the static efficiency and dynamic changes in forestry productivity in thirty Chinese regions using the slacks-based measure (SBM) technique and the Malmquist–Luenberger index. Ecological and economic development through forest carbon sinks and production value are highlighted in this study. The research shows that environmental efficiency and productivity estimators outperform economic progress, with the southwest region of China leading the way. Zhang and Xu [33] use a novel hybrid approach, combining LCA and time-series DEA, to evaluate an enterprise’s eco-efficiency in the forestry industry. Efficient years experience 45.25% of the environmental impact of inefficient years, with fiber and energy consumption further reducing these percentages to 65.53% and 77.66%. The study offers practical recommendations for forestry firms to minimize their environmental impacts and optimize resource use in the industrial chain. Obi and Visser [34] report 1.7% annual productivity growth in New Zealand’s forest harvesting sector from 2009 to 2018, driven by technology and efficiency improvements. Their research emphasizes that optimizing the technology efficiency and reducing inputs can sustain productivity growth in the forestry industry. Yin et al. [35] argue that the carbon sequestration efficiency, ecological afforestation, temperature, GDP per capita, urbanization, population, and total imports and exports have spillover effects. Many studies use the Malmquist productivity index to gauge the dynamic change (TFP) in forest resource efficiency globally [36]. Further, DEA has been extensively used in forestry and environmental efficiency estimation [37,38,39,40].

Studies have proven that the forestry efficiency and sustainability depend on technological advancement. GIS, remote sensing, and drones provide real-time data on forest health, growth, and environmental variables, enabling more precise forest management. Mechanized harvesters and GPS-guided tractors speed up logging, minimizing waste and environmental effects. Analytics and predictive modeling improve forestry decision-making, maximizing resource allocation and minimizing ecological impacts. Technology boosts productivity, resource use, and carbon reduction in the forestry business, making it more efficient and sustainable [41,42]. Obi and Visser [43] evaluated the efficiency of New Zealand’s forest harvesting sector using categorical data envelopment analysis (DEA) to resolve the increasing competition in the global forestry market. They demonstrated the suitability of categorical DEA for this application and evaluated the impact of log extraction techniques and processing locations on efficiency scores. The analysis revealed that timber extraction techniques significantly affected the overall performance, with grapple skidders exhibiting the highest average efficiency of 58%. However, processing at the stump is associated with the most significant average efficiency score. To remain competitive and profitable in the ever-changing timber market, selecting efficient log extraction technologies is crucial. Ke et al. [44] also demonstrated the importance of technology in enhancing forestry efficiency. This regional heterogeneity of production technology utilizing forest resources is essential in efficiency and productivity evaluation [45,46]. This study evaluates the forestry efficiency and total factor productivity growth in different regions and provinces of China to gauge the level of success and further understand the production technology heterogeneity in four other regions.

3. Materials and Methods

The synergy between forecasting and optimization is crucial in decision-making. Forecasting, utilizing methodologies like time-series analysis and machine learning, predicts future trends based on past data, providing insights for informed decision-making, risk management, and strategic planning. Methods for optimization, such as linear and integer programming, focus on finding the most efficient solutions to maximize or minimize objectives. It aids in the improvement of resource allocation and processes. Precise forecasts serve as valuable inputs for optimization models, guiding production planning and inventory management decisions. Optimization models like DEA, with dynamic capabilities, adjust plans in response to real-time or updated forecasts, enhancing the adaptability in volatile contexts. Data envelopment analysis (DEA) is a commonly used mathematical method that utilizes linear programming to evaluate the efficiency of similar decision-making units (DMUs) [47,48]. The selection of data envelopment analysis (DEA) over stochastic frontier analysis (SFA) in our research was driven by the unique characteristics of our dataset. Our data did not conform to the normality assumption, making DEA a more suitable choice. DEA is a non-parametric method that does not rely on distributional assumptions, making it robust when dealing with data that may not follow a normal distribution. This flexibility allowed us to effectively assess the efficiency and productivity in the forestry sector, considering the specific nature of our dataset. The traditional DEA model, pioneered by Charnes et al. [49], assumes a constant return to scale (CSR). Based upon this, Banker et al. [50] modified the model to incorporate a variable return to scale (VSR). A preliminary [51] investigation by Tone introduced the slacks-based measure (SBM) model. Subsequently, Tone [52] devised a method of ranking the most effective DMUs. The selection of DEA-SBM to estimate the forestry resource efficiency is attributed to its capacity to evaluate systems with multiple inputs and outputs. Specifically, it compares forestry units (provinces) based on inputs (forest area, investment) and outputs (forest output value, timber output, forest stock volume). DEA-SBM uses the efficiency frontier as a measure to evaluate unit performance. Importantly, DEA-SBM is suitable when it is difficult to establish a specific functional form for the production frontier, distinguishing it from stochastic frontier analysis (SFA). Unlike SFA, which assesses inefficiency by comparing it to a particular form, DEA-SBM’s non-parametric approach offers flexibility in evaluating the efficiency of China’s varied forest resources. The study used Max-DEA for estimation.

3.1. DEA-SBM Model

The slacks-based measure (SBM) represents a non-radial approach to assessing the data envelopment analysis (DEA) efficiency. Its primary strength lies in its capacity to directly evaluate excess inputs and insufficient outputs. When determining efficiency, it considers the slack, which represents the difference between inputs and outputs at the production frontier. This method operates based on the following principles. Suppose that we have a study with n decision-making units (DMUs) referred to as “Provinces”. M input indicators and s output indicators characterize each DMU. Let ${\mathit{B}}_{\mathit{j}}$, represent the $\mathit{j}$-th DMU, where j ranges within $\mathit{j}=\mathrm{1,2},\dots ..,\mathit{n};\left[{\mathit{x}}_{\mathit{i}\mathit{j}}\right]$, represents the $\mathit{m}\times 1$ input indicators of DMU ${\mathit{B}}_{\mathit{j}}$, with $\mathit{i}$ ranging from 1 to $\mathit{m};\left[{\mathit{y}}_{\mathit{r}\mathit{j}}\right]$ represents the $\mathit{s}\times 1$ output indicators of DMU ${\mathit{B}}_{\mathit{j}}$, with r ranging from 1 to $\mathit{s}$. The relative efficiency value of the j_0-th DMU is denoted as ${\mathit{h}}_{\mathit{j}0}$. Now, let us discuss how the output-focused SBM-DEA model with variable returns to scale operates:

$$\begin{array}{c}\mathbf{M}\mathbf{i}\mathbf{n}{\mathit{h}}_{\mathit{j}0}=\mathit{\theta }\\ \mathrm{s}.\mathrm{t} \left\{\begin{array}{cc}{\displaystyle \sum _{\mathit{j}=1}^{\mathit{n}}}{\mathit{\lambda }}_{\mathit{j}}{\mathit{x}}_{\mathit{i}\mathit{j}}\le \mathit{\theta }{\mathit{x}}_{\mathit{i}\mathit{j}0},\mathit{i}=1,\dots ,\mathit{m}& \hspace{1em}\hspace{1em}\hspace{1em}\left(1\right)\\ {\displaystyle \sum _{\mathit{j}=1}^{\mathit{n}}}{\mathit{\lambda }}_{\mathit{j}}{\mathit{y}}_{\mathit{r}\mathit{j}}\ge {\mathit{y}}_{\mathit{i}\mathit{j}},\mathit{r}=1,\dots ,\mathit{s}& \hspace{1em}\hspace{1em}\hspace{1em}\left(2\right)\\ {\displaystyle \sum _{\mathit{j}=1}^{\mathit{n}}}{\mathit{\lambda }}_{\mathit{j}}=1,{\mathit{\lambda }}_{\mathit{j}}\ge 0,\mathit{j}=1,\dots .,\mathit{n}& \hspace{1em}\hspace{1em}\hspace{1em}\left(3\right)\end{array}\right.\end{array}$$

The efficiency value at the j-th position is represented as $\mathit{\theta }$, where ${\mathit{\lambda }}_{\mathit{j}}$ is a nonnegative vector. A DMU is considered efficient if and only if $\mathit{\theta }$ equals 1, indicating that it operates at maximum efficiency. If $\mathit{\theta }$ is not equal to 1, the DMU is inefficient and has room for improvement.

3.2. DEA-Meta-Frontier Model

The utilization of the meta-frontier model yields enhanced precision in the evaluation of DMU efficiency across distinct groups. Consequently, given that all DMUs within a given group operate under equivalent technological conditions, the most prudent approach is to undertake efficiency comparisons within the confines of the same group. The technical gap ratio (TGR) metric proves instrumental in gauging the extent of divergence in technological progress between specific groups [53].

$$\mathit{T}\mathit{G}\mathit{R}=\frac{\mathit{M}\mathit{F}\mathit{R}\mathit{E}}{\mathit{G}\mathit{F}\mathit{R}\mathit{E}\mathit{i}}$$

(4)

The forestry resource efficiency (FRE) is computed for the complete set of decision-making units (DMUs) under consideration. In this context, GFREi represents the FRE applicable to DMUs categorized within a specific group. In contrast, MFRE pertains to the Meta-FRE associated with DMUs spanning the entire population, including all distinct groups. By juxtaposing the divergence between a meta-frontier technology and the frontier technology specific to a given group, the technology gap ratio (TGR) functions to numerically characterize the discrepancy between these two cohorts of DMUs [54]. An equivalent of TGR denotes the absence of a technological rift between the comprehensive population and the group frontier, thus establishing TGR as a prevalent instrument in evaluating territorial disparities.

3.3. DEA-Malmquist Productivity Index

The Malmquist productivity index (MPI) is a comprehensive analytical tool to assess temporal productivity variations in decision-making units (DMUs), such as organizations and businesses. Derived from production economics and building upon data envelopment analysis (DEA), the MPI comprises two vital elements for nuanced evaluation. Firstly, the technical efficiency change (TEC) measures shifts in a DMU’s distance to the production frontier, indicating changes in efficiency relative to top performers. A positive TEC signifies increased efficiency, while negative values denote a decrease. Secondly, the technological change (TC) assesses transformations in the production frontier, reflecting technological and managerial advances. The MPI, obtained by multiplying TEC and TC, represents the aggregate productivity change between two periods. A score above one indicates heightened productivity due to technological advances and improved efficiency, while a score below one signifies a decline. The MPI is valuable for decision-makers seeking insights into productivity shifts, offering applicability across industries and guiding strategic decisions based on a comprehensive understanding of technological dynamics and efficiency. Malmquist productivity indices provide a valuable tool for a decision-making unit (DMU) to track improvements in efficiency over time. To effectively utilize this index, it is assumed that a production function that accurately represents the current technological environment exists. DEA models precisely pinpoint the location of this production function’s threshold. The difference in output between periods t and t + 1 defines a specific DMU, referred to as ($\mathit{D}\mathit{M}{\mathit{U}}_0$) [55].

$${\mathit{M}}_0=\frac{{\mathit{D}}_0^{\mathit{t}+1}\left({\mathit{x}}_0^{\mathit{t}+1},{\mathit{y}}_0^{\mathit{t}+1}\right)}{{\mathit{D}}_0^{\mathit{t}}\left({\mathit{x}}_0^{\mathit{t}}{\mathit{y}}_0^{\mathit{t}}\right)}{\left[\frac{{\mathit{D}}_0^{\mathit{t}}\left({\mathit{x}}_0^{\mathit{t}+1},{\mathit{y}}_0^{\mathit{t}+1}\right)}{{\mathit{D}}_0^{\mathit{t}}\left({\mathit{x}}_0^{\mathit{t}},{\mathit{y}}_0^{\mathit{t}}\right)}\frac{{\mathit{D}}_0^{\mathit{t}+1}\left({\mathit{x}}_0^{\mathit{t}+1},{\mathit{y}}_0^{\mathit{t}+1}\right)}{{\mathit{D}}_0^{\mathit{t}+1}\left({\mathit{x}}_0^{\mathit{t}}{\mathit{y}}_0^{\mathit{t}}\right)}\right]}^{1/2}$$

(5)

where:

• ${\mathit{D}}_0^{\mathit{t}}\left({\mathit{x}}_0^{\mathit{t}},{\mathit{y}}_0^{\mathit{t}}\right)$ shows the TE estimation of the $\mathit{D}\mathit{M}{\mathit{U}}_0$ for period t;
• ${\mathit{D}}_0^{\mathit{t}+1}\left({\mathit{x}}_0^{\mathit{t}+1},{\mathit{y}}_0^{\mathit{t}+1}\right)$ illustrates the TE estimation for period t + 1;
• ${\mathit{D}}_0^{\mathit{t}}\left({\mathit{x}}_0^{\mathit{t}+1},{\mathit{y}}_0^{\mathit{t}+1}\right)$ specifies the variation in TE from time t to t + 1;
• ${\mathit{D}}_0^{\mathit{t}+1}\left({\mathit{x}}_0^{\mathit{t}},{\mathit{y}}_0^{\mathit{t}}\right)$ represents the technical efficiency of a specific $\mathit{D}\mathit{M}{\mathit{U}}_0$. This efficiency is computed by replacing its data from period t with the corresponding data from period t + 1.

The initial segment of Equation (5) without parentheses denotes the variation in the technical efficiency of $\mathit{D}\mathit{M}{\mathit{U}}_0$ between time t and t + 1. The timeframe enclosed within the square brackets illustrates the advancement in technology for the same DMU. If the index value exceeds 1, it signifies that $\mathit{D}\mathit{M}{\mathit{U}}_0$ achieved a greater output during the second period compared to the first. Two hypotheses can be put forth to elucidate this substantial rise in output. Firstly, it is plausible that $\mathit{D}\mathit{M}{\mathit{U}}_0$ embraced state-of-the-art methodologies, enhancing its efficiency.

3.4. Kruskal–Wallis Test

The Kruskal–Wallis test is a non-parametric statistical method used to assess the presence of statistically significant differences between three or more independent groups or treatments [56]. The Kruskal–Wallis test differs from the Mann–Whitney U test, designed to compare two independent groups by not assuming a normal data distribution. Instead, it determines whether the medians of the groups are comparable by rating all values collectively and then determining whether these ranked values exhibit significant disparities across groups. When the Kruskal–Wallis test identifies a statistically significant difference, it indicates that at least one group differs concerning the investigated variable. This test proves invaluable when ordinal or continuous data fail to satisfy the requirements of parametric tests such as the analysis of variance (ANOVA). The Kruskal–Wallis test is frequently used in social sciences, healthcare, and environmental studies to compare groups with non-normally distributed data. In this study, the different regions’ forestry efficiency, productivity growth, and technology gap ratios should be heterogeneous. However, are these values statistically significantly different for other regions of China? To prove this, the Kruskal–Wallis test identifies the significant statistical differences among the four Chinese regions for the average FRE, MI, and TGR. The hypotheses are as follows:

H₁: 

The FRE is the same in the four different Chinese regions.

H₂: 

The TGR change is the same in the four different Chinese regions.

H₃: 

The MI is the same in the four different Chinese regions.

4. Variable Selection and Data Collection

The selection of inputs and outputs in DEA efficiency estimation carries specific significance, as the inappropriate choice of variables can lead to inaccurate and biased estimation outcomes [57,58]. Numerous research studies have employed diverse combinations of inputs and outputs to assess the efficiency of forest resources [59]. Consequently, in line with previous research, we have utilized six distinct inputs and outputs for the estimation of the forest resource efficiency (FRE), as presented in

Table 1. Inputs and outputs used to estimate the forestry efficiency.

No.	Inputs	Unit
1	Forest area	10,000 hectares
2	Investment	10 thousand yuan
3	Employees	10 thousand persons
	Outputs
4	Forestry output value	100 million yuan
5	Timber output	10,000 cubic meters
6	Forest stock volume	10,000 cubic meters

. The forest area pertains to the comprehensive land area designated for forestry production, reflecting investments in natural resource ecology. Investment elucidates the accumulation of capital investments since the beginning of the year, contributing to the inputs involved in reforestation. Employees represent the personnel employed at the end of the year, serving as an indicator of investments in human resources. The forestry output value signifies the resultant economic benefits.

Timber output embodies the generation of social benefits. The forest stock volume refers to the precise measurement, typically expressed in solid cubic meters, of the complete volume occupied by the stems of all living trees within a specified forest area. This measurement includes the entirety of the tree stem, starting from either the ground level or stump height up to a designated top diameter. It focuses explicitly on living trees, excluding any non-living biomass, such as dead trees or fallen branches. The primary aim is to provide a precise quantitative assessment of the amount of wood and biomass present in the defined forest area. Beyond its economic implications in assessing timber and forest products, this metric is crucial in evaluating sustainable forest management practices and determining the forest’s capacity to regenerate and sustainably provide resources. Moreover, the “forest stock volume” contributes to economic decision-making and ecological understanding, offering insights into carbon sequestration, biodiversity, and overall ecosystem health. Its frequent use in academic and scientific studies underscores its significance in analyzing forest growth trends, studying the impact of environmental factors, and assessing the effectiveness of various forestry management strategies. The specific variable names and units are specified in Table 1. The data were collected from China’s Forestry and Grassland statical yearbook.

5. Results and Discussion

The following three sections present the forestry efficiency, productivity change, and production technology heterogeneity in the 4 regions and 31 provinces of China. Section 3.1 explains how efficiently the decision-making units (provinces) utilize the forest resources to convert them into economic output. Further, it evaluates the impact of production technology differences in different regions on FRE. In Section 5.2, the Malmquist productivity index is used to gauge the dynamic change in forest resource efficiency in different Chinese areas and provinces. It also estimates the primary determinant of the total factor productivity change. Finally, Section 5.3 gauges the statistically significant differences in the forestry efficiency, productivity change, and technology gap ratio in the four different regions of China.

5.1. Forest Resource Efficiency and Technological Heterogeneity in China

The application of DEA-SBM is used to evaluate the forestry efficiency and meta-frontier analysis is used to gauge the meta-frontier, group frontier, and technology gap ratio in 31 Chinese provinces from 2001 to 2020. The results in Figure 2 and

Table 2. Forestry efficiency in Chinese provinces (2001–2020).

Year	Meta-Frontier	Group Frontier	TGR
2001	0.5264	0.8299	0.6343
2002	0.4901	0.8249	0.5941
2003	0.5664	0.8229	0.6883
2004	0.5156	0.8314	0.6202
2005	0.4899	0.7688	0.6372
2006	0.4705	0.756	0.6224
2007	0.4924	0.7928	0.6211
2008	0.4612	0.7411	0.6223
2009	0.5579	0.7994	0.6979
2010	0.5520	0.8409	0.6564
2011	0.5160	0.7976	0.6469
2012	0.4906	0.8305	0.5907
2013	0.5830	0.8331	0.6998
2014	0.5943	0.8626	0.6890
2015	0.6140	0.8772	0.7000
2016	0.6000	0.8827	0.6797
2017	0.5753	0.8591	0.6697
2018	0.5899	0.8406	0.7018
2019	0.5890	0.8343	0.706
2020	0.5856	0.8543	0.6855
Avg.	0.5430	0.8240	0.6582

illustrate that the average forestry efficiency is 0.5430 in China for the study period. It indicates that Chinese provinces, on average, still have potential of 45.70% to enhance their forest resource utilization efficiency. The efficiency of forest resources could be increased by reducing the input resources or increasing the output growth in forestry. This emphasis on efficiency ensures the sustainability of vital ecosystems, contributes to economic growth, conserves natural resources, aligns with international commitments, and fosters innovation. As the global demand for forest products rises, optimizing resource utilization to meet these demands while minimizing environmental impacts and promoting responsible forest management becomes essential [60].

The forest resource efficiency (FRE) trend exhibited notable fluctuations throughout the study period. Initially, there was a gradual increase in FRE from 2001 to 2003, indicating an improvement in the efficient utilization of forest resources. However, this positive trend was reversed, and the FRE started to decline, reaching a low point of 0.4612 in 2008. Notably, in 2009, there was rapid and substantial growth in FRE, with a value of 0.5579. Following this spike, the FRE had a consistent and gradual upward trajectory. Particularly noteworthy are 2015, 2016, and 2018, when the FRE reached its peak levels. These years signify that the Chinese provinces successfully implemented highly effective operational strategies to optimize the utilization of their forest resources. These periods of peak efficiency serve as examples of the best practices in sustainable forestry management, highlighting the potential for improved resource utilization in other years. Studies have proven that the policies of the Chinese central government have led to this growth in forest resource efficiency. The increase in forest resource efficiency (FRE) in 2009 was attributed to various factors, such as potential policy changes, technological advancements, heightened environmental awareness, shifting market demands, capacity development, or external influences [61,62]. The results of the FRE in the group frontier are relatively better than in the meta-frontier. This demonstrates that provinces perform better in their group frontier compared to the meta-analysis. The average FRE in the group frontier is 0.8240, which indicates growth in forest resource utilization; however, there is still potential of 17.60%. TGR represents the technological gap ratio between the meta-frontier and group frontier. The value of 0.6582 for TGR indicates a massive technological gap over the study period. A TGR value of 1 indicates no gap between the technology utilized by the group and the meta-frontier.

Table 3. Forestry efficiency (meta-frontier, group frontier) and TGR in 4 regions of China.

Forest Region	Province	Meta-Frontier	Group Frontier	TGR
Northeast	Heilongjiang	0.3299	0.8939	0.3691
	Inner Mongolia	0.2271	1	0.2271
	Jilin	0.794	1	0.794
	Liaoning	0.4068	1	0.4068
Avg. Northeast		0.4394	0.9735	0.4492
Northern	Beijing	0.4676	0.9322	0.5016
	Gansu	0.1528	0.6389	0.2392
	Hebei	0.2946	0.7077	0.4163
	Henan	0.5097	1	0.5097
	Ningxia	0.1555	0.3763	0.4132
	Qinghai	0.0673	0.2907	0.2315
	Shaanxi	0.2108	0.9552	0.2207
	Shandong	0.6888	0.9536	0.7223
	Shanxi	0.2449	0.599	0.4088
	Tianjin	0.6489	0.8854	0.7329
	Xinjiang	0.2206	1	0.2206
Avg. Northern		0.3329	0.7581	0.4197
Southern	Anhui	1	1	1
	Fujian	0.9705	1	0.9705
	Guangdong	0.8441	0.8751	0.9646
	Guangxi	0.8255	0.8291	0.9957
	Guizhou	0.3591	0.6107	0.588
	Hainan	0.9462	0.979	0.9665
	Hubei	0.3618	0.4649	0.7782
	Hunan	0.5387	0.5671	0.9499
	Jiangsu	0.8131	0.8131	1
	Jiangxi	0.5321	0.6328	0.8409
	Shanghai	0.9635	0.9636	0.9999
	Zhejiang	0.8519	0.9201	0.9259
Avg. Southern		0.7505	0.8046	0.915
Southwest	Chongqing	0.401	0.9383	0.4274
	Sichuan	0.434	0.7173	0.605
	Tibet	1	1	1
	Yunnan	0.5725	1	0.5725
Avg. Southwest		0.6019	0.9139	0.6512

and Figure 3 elaborate on the study results and show the efficiency of forest resource use in different provinces and forest regions of China. The study divides the 31 Chinese provinces into four forest regions: the northeast, north, south, and southwest. Anhui (1), Tibet (1), Fujian (0.9705), Shanghai (0.9635), and Hainan (0.9462) are found to be the top performers in forestry utilization during the study period. In contrast, Qinghai (0.0673), Gansu (0.1528), Ningxia (0.1555), Shaanxi (0.2108), and Xinjiang (0.2206) are the poorest performers in terms of forestry efficiency. Except for these, Zhejiang, Guangdong, Guangxi, Jiangsu, Jilin, Shandong, Tianjin, Yunnan, Hunan, Jiangxi, and Henan show scores between 0.50 and 0.86. Meanwhile, Beijing, Sichuan, Liaoning, Chongqing, Hubei, Guizhou, Heilongjiang, Hebei Shanxi, and Inner Mongolia have scores between 0.22 and 0.48. These scores indicate the success levels of all mainland Chinese provinces in efficiently utilizing their forest resources. To narrow the efficiency gap, less efficient Chinese provinces can learn from the exemplary practices of proficient ones, as identified in Table 3 and Figure 3, where the provinces are classified into four forest regions. Less efficient provinces should adopt the best practices, invest in capacity building, consider policy reforms, leverage modern technology, raise environmental awareness, and promote regional collaboration. These steps can pave the way for improved, sustainable forest resource management, benefiting the environment and the economy. Further elaborating on the regional results, the study found that the average FRE in the southern forest region of China is at an optimum level, with an efficiency score of 0.7505. Similarly, the southwest (0.6019) ranked second, and the northeast (0.4394) and northern (0.3329) regions ranked third and fourth, respectively. This indicates that, on average, provinces in the southern region are more efficient in utilizing their forest resources.

Explaining the group frontier’s results, we found that the northeastern region secured the highest efficiency score. The group frontier’s efficiency score demonstrates a particular province’s performance in its respective region. Inner Mongolia, Jilin, and Liaoning are all efficient in their groups. In Northern China, Henan and Xinjiang are the most efficient in their group, with an efficiency score of 1. In Southern China, Anhui and Fujian are the top performers, with a group frontier score of 1. Finally, in the southwestern region, Tibet is the most efficient among all group members. Inefficient provinces can improve their efficiency by comparing themselves to their more efficient counterparts. The group frontier results emphasize provinces with strong performance within their respective regions. For instance, the northeastern region has the maximum score for efficiency, with provinces such as Inner Mongolia, Jilin, and Liaoning demonstrating efficiency as well. Inefficient provinces should identify these standards, evaluate successful techniques, set concrete objectives, undertake reforms, cooperate with efficient provinces, track their progress, adjust their plans, involve stakeholders, raise public awareness, and celebrate their successes. Underperforming regions can improve their efficiency and development by implementing these strategies and learning from more effective provinces [63,64].

Table 3 provides comprehensive data on the forestry efficiency within the four distinct regions of China in conjunction with a pertinent metric called the technology gap ratio (TGR). The TGR is a quantitative measure to assess the technological disparity between the meta-frontier, representing the pinnacle of potential performance, and the group frontier, which denotes the actual performance of a given province or region. A TGR value of 1 signifies that a region is operating at its technological zenith or the highest level of efficacy attainable. The TGR values for each province within the various regions are meticulously delineated, serving as indicators of the extent to which each province approximates its maximum technological frontier. As an illustrative example, Jilin exhibits a TGR of 0.794 in the northeast region, signifying that it is relatively close to achieving maximum efficiency. At the same time, Inner Mongolia lags, with a TGR of 0.2271, denoting significant room for improvement.

Conversely, provinces with a TGR of 1 have already achieved the zenith of forestry efficiency, having reached their technological frontiers. On average, the TGR value for the northeast region stands at 0.4492, a substantial departure from the optimal value of 1. This finding underscores the growth potential in production technology utilization within this region. In comparison, the northern part records the lowest TGR of all four regions, averaging 0.4197. However, Tianjin stands out as a noteworthy exception within the northern region, maintaining superior technology with a TGR value of 0.7329. At the same time, Xinjiang lags, with a TGR of 0.2206, marking it as the least efficient performer among the northern provinces.

In contrast, the southern forest region of China demonstrates an impressive technological advantage, with an average TGR value of 0.915, approaching the ideal value of 1. This finding suggests that the southern provinces excel in utilizing modern technology to harness forest resources. The TGR values of Anhui and Jiangsu reach 1, underscoring the contemporary production technologies employed within these provinces. Conversely, Guizhou emerges as the least efficient in terms of production technology within the southern forest region of China, with a TGR value of 0.588. Lastly, the southwest forest region of China is characterized by moderate production technology, with an average TGR of 0.6512, ranking second after the southern region. Notably, Tibet is the top performer in this region, achieving a TGR value of 1. The study result, which demonstrates the presence of production technology heterogeneity in different Chinese areas, is aligned with numerous studies on the subject [59,65]. Studies have found that several crucial measures can be implemented in regions with lower technology gap ratio (TGR) values in forest resource utilization to improve the production technology. It is imperative to reinforce investments in research and development (R&D) to remain abreast of contemporary methodologies. Investing in training and capacity development for forest workers and encouraging technology transfer from regions with higher TGR values is essential. Priority should be given to installing more modern and efficient equipment; consequently, comprehensive resource management strategies should be formulated as per technological developments. Incentives from the government, collaboration, and knowledge exchange can all contribute to accelerating technology adoption. Community engagement, environmental sustainability, and monitoring and evaluation are also essential components of this process [66,67,68].

5.2. Malmquist Total Factor Productivity Index Results

This section examines the Malmquist total factor productivity index by employing panel data that cover China’s municipalities and inland provinces from 2001 to 2020. As a result, the Malmquist total factor productivity and its constituent values are derived. By employing this analytical methodology, one can examine the dynamic efficiency of forestry resources in China. The results are displayed in

Table 4. Total factor productivity, efficiency, and technology change in forest resources in China (2001–2020).

Year	MI	EC	TC
2001–2002	1.0965	0.9736	1.1262
2002–2003	0.9818	1.2958	0.7577
2003–2004	0.897	0.8873	1.0109
2004–2005	1.037	0.9538	1.0872
2005–2006	1.144	0.9332	1.2259
2006–2007	1.1178	1.0468	1.0678
2007–2008	0.9091	0.9648	0.9423
2008–2009	1.136	1.2941	0.8778
2009–2010	0.9032	1.0029	0.9006
2010–2011	1.0118	0.9291	1.089
2011–2012	1.1259	1.0148	1.1095
2012–2013	1.0145	1.4035	0.7228
2013–2014	0.7841	1.0408	0.7534
2014–2015	0.7619	1.0834	0.7032
2015–2016	1.0216	0.9689	1.0544
2016–2017	1.0395	0.9583	1.0847
2017–2018	1.119	1.0312	1.0851
2018–2019	1.152	1.0199	1.1295
2019–2020	0.9802	0.9959	0.9842
Avg.	0.9644	1.0174	0.948

. It offers valuable insights into the broader patterns and changes in the efficiency of forest resources at the provincial and municipal levels. MI presents the total factor productivity change over the study period. EC and TC are decomposing factors of total factor productivity change. EC presents the technical efficiency change from t to t + 1.

TC is a decision-making unit’s technological change in forestry resources from one year to the next. The average value (2001–2020) of MI is 0.9644. It demonstrates that the productivity change of forestry resources in Chinese provinces declined by 3.56% over the study period. Elaborating the determinant of productivity change, a study found that TC was the main factor responsible for productivity deterioration as the EC was higher than the TC; on average, technological change declined by 5.2 percent. In contrast, the EC grew by 1.74 percent over the study period. In the 2018–2019 period, Chinese provinces witnessed the highest total factor productivity growth, with an average value of 1.152. EC was at its optimum level in 2012–2013, with an average value of 1.4035. The average highest TC = 1.2259 was secured in the period 2005–2006. As the results indicate, technological advancement is a primary factor in total factor productivity decline; therefore, the Chinese government should develop policies that enhance the technological advancement in forestry resource utilization. In order to improve the Malmquist total factor productivity (TFP) index in China’s forestry sector, it is necessary to consider the allocation of resources towards technological advancement and the adoption of modern forestry technologies; encourage efficiency improvements via sustainable practices; conduct periodic evaluations of regional performance; facilitate the exchange of knowledge and collaboration; focus investments on addressing specific challenges; establish enabling policies; and promote research. The measures mentioned above are designed to enhance the forestry sector’s overall productivity, efficiency, and technological progress, thereby positively contributing to sustainable resource management [69,70,71].

To elucidate the comprehensive changes in total factor productivity related to forest resources within various regions and to assess the determinants underlying the productivity growth or decline, this study partitioned the 31 mainland Chinese provinces into four distinct forest regions. The Malmquist productivity index (MI) results for these four regions are delineated in

Table A1. MI, EC, and TC of forest resources in 4 different regions of China.

Forest Region	Province	MI	EC	TC
Northeast	Heilongjiang	0.9359	1.025	0.9131
Northeast	Inner Mongolia	0.9423	0.9956	0.9465
Northeast	Jilin	0.9723	1.0312	0.9429
Northeast	Liaoning	1.007	1.0177	0.9895
Avg. Northeast		0.9644	1.0174	0.948
Northern	Beijing	1.0659	1.1491	0.9276
Northern	Gansu	0.9467	1.0197	0.9284
Northern	Hebei	1.0377	1.0487	0.9895
Northern	Henan	0.9692	0.9693	0.9999
Northern	Ningxia	0.9518	0.9839	0.9674
Northern	Qinghai	0.9545	0.986	0.9681
Northern	Shaanxi	0.9857	1.0482	0.9404
Northern	Shandong	1.1199	1.0419	1.0749
Northern	Shanxi	1.0649	1.0998	0.9683
Northern	Tianjin	1.1554	1.1303	1.0222
Northern	Xinjiang	0.9533	1.0071	0.9466
Avg. Northern		1.0186	1.044	0.9757
Southern	Anhui	1.0152	1	1.0152
Southern	Fujian	1.007	1.0012	1.0058
Southern	Guangdong	1.0481	1.0308	1.0168
Southern	Guangxi	1.1693	1.0426	1.1215
Southern	Guizhou	1.0313	1.0821	0.9531
Southern	Hainan	0.9798	0.9805	0.9993
Southern	Hubei	1.0081	1.0361	0.973
Southern	Hunan	0.9776	1.0018	0.9758
Southern	Jiangsu	1.0095	1.0302	0.9799
Southern	Jiangxi	1.0781	1.0734	1.0044
Southern	Shanghai	1.0088	1.1046	0.9133
Southern	Zhejiang	1.0342	1.0224	1.0115
Avg. Southern		1.0306	1.0338	0.9975
Southwest	Chongqing	1.0847	1.0949	0.9907
Southwest	Sichuan	0.9712	1.1888	0.817
Southwest	Tibet	0.9174	1	0.9174
Southwest	Yunnan	0.9769	1.0589	0.9226
Avg. Southwest		0.9875	1.0856	0.9119

. These findings demonstrate that the MI score, with a value of 1.0306, is highest in the Southern Chinese forest region, indicating an average 3.06% increase in total factor productivity. The analysis of the total factor productivity change, decomposed into the efficiency change (EC) and technological change (TC), reveals that EC is the primary driver of productivity growth, with a value of 1.0338, surpassing TC’s value of 0.9975. Technical efficiency experienced a notable increase of 3.38%, whereas technological change displayed a modest decline of 0.25%. The northern forest region secured the second position, with an average MI score of 1.0186, indicating a 1.86% increase in total factor productivity. Analogous to the southern part, EC plays the predominant role in driving productivity growth, exhibiting substantial growth of 4.4%, with an EC value of 1.044. TC, conversely, deteriorated by 2.43%, with an average value of 0.9757. The southwestern forest region of China observed a decline of 1.25% in total factor productivity over the study period, primarily attributed to technological deterioration. The average TC value, at 0.9119, underscores an 8.81% decline, while technical efficiency change registered growth of 0.85%, with an EC value of 1.0856. Finally, the northeastern forest region of China emerged as the poorest performer, with a mean MI score of 0.9644, signifying a 3.56% decline in total factor productivity. Technological change is the principal factor driving this total productivity deterioration, with a TC value of 0.948. Like the southwest region, this area also experienced 1.74% growth in technical efficiency.

In the context of provincial total factor productivity changes related to forest resource utilization, the study identified Guangxi, Tianjin, Shandong, Chongqing, and Jiangxi as the top performers, with notable growth in MI. Conversely, Tibet, Heilongjiang, Inner Mongolia, Gansu, and Ningxia were identified as the poorest performers, experiencing a decline in MI. Further elaboration of the results reveals that Liaoning, Fujian, Hubei, Shanghai, Jiangsu, Anhui, Guizhou, Zhejiang, Hebei, Guangdong, Shanxi, and Beijing all exhibited growth in MI. At the same time, Xinjiang, Qinghai, Henan, Sichuan, Jilin, Yunnan, Hunan, Hainan, and Shaanxi showed a deterioration in their total factor productivity. The determinants of growth and the factors contributing to the decline in MI vary among each province and are detailed in Table A1. To enhance the total factor productivity (TFP) in forest resource management, regions and provinces can draw inspiration from their more efficient counterparts. In particular, emulating southern and northern forest regions, where the TFP grew due to enhanced technical efficiency, could involve investing in modernization, process optimization, and training.

Likewise, regions such as Guangxi, Tianjin, Shandong, Chongqing, and Jiangxi, which excelled in TFP, can serve as models for other areas through sustainable forest management practices, conservation policies, public–private partnerships, and incentives for sustainable practices. Strengthening the monitoring and reporting systems and promoting knowledge exchange between regions can further improve the TFP. The advancement of technology in forest resource management is critical to attaining sustainability and enhancing efficiency. By cultivating an environment that promotes innovation and adopts state-of-the-art technologies, regions and provinces can more effectively confront the obstacles and prospects associated with forest utilization and conservation [72,73]. Studies have found that to facilitate technological advancements in the administration of forest resources, it is recommended that regions and provinces allocate resources towards research and development, capitalizing on tools including blockchain, artificial intelligence, satellite imagery, drones, and the Internet of Things. It is imperative to prioritize collaboration with technology firms, implement training programs for personnel, and establish regulatory support. It is also important to initiate technical advancements through experimental efforts while ensuring ongoing monitoring and evaluation. By utilizing these approaches, regions may foster innovation, enhance their data-based decision-making, and bolster their sustainable forest management practices [74,75].

5.3. Kruskal–Wallis Test Results

The results in Section 5.1 and Section 5.2 indicate that the forest resource efficiency, technology gap ratio, and total factor productivity change are heterogeneous in the four different regions of China. However, whether these results are statistically significant is a question of great concern. To this end, the results of the Kruskal-Wallis test, a non-parametric statistical technique utilized to ascertain whether significant differences exist among multiple groups or conditions, are displayed in

Table 5. Kruskal–Wallis test results.

Hypothesis Test Summary
	Null Hypothesis	Test	Sig.	Decision
1	The average FRE scores are the same across four different Chinese regions	Independent-Samples Kruskal–Wallis Test	0.005	Reject the null hypothesis
2	The average TGR scores are the same across four different Chinese regions.		0.001	Reject the null hypothesis
3	The average MI scores are the same across four different Chinese regions.		0.096	Retain the null hypothesis

Asymptotic significances are displayed. The significance level is 0.050.

and Figure A1, Figure A2 and Figure A3. The test assesses three null hypotheses about diverse factors across four discrete regions in China. To begin, the null hypothesis positing that the mean forest resource efficiency (FRE) scores are consistent throughout these regions is rejected. This conclusion is reached using the Kruskal–Wallis test, which produces a p-value of 0.005, indicating substantial variations in the average FRE scores across the regions.

Additionally, the null hypothesis concerning the mean total growth rate (TGR) scores is denied, with a p-value of 0.001, suggesting that there are significant variations in the mean TGR scores. Nevertheless, the third null hypothesis pertaining to the mean Malmquist index (MI) scores among the regions remains valid due to the lack of statistical significance inferred from the p-value of 0.096 obtained from the Kruskal–Wallis test. Fundamentally, these findings indicate substantial discrepancies in the FRE and TGR scores across the regions, whereas no such variations are apparent in the MI scores. The insights that these results provide regarding the regional variations in these metrics are invaluable and will aid in subsequent analyses and decision-making. It is critical to initiate education and training initiatives in less technologically developed regions to advance forest resource management technology. These initiatives can potentially provide laborers and local communities with an understanding of contemporary forest management technologies and techniques [76].

Furthermore, integrating remote sensing and GIS technologies, such as drones and satellite imagery, can potentially augment the capacities for monitoring and mapping. Utilizing mobile devices and GPS-based inventory systems to collect digital data can facilitate more precise and effective data gathering, thereby enhancing the quality of decision-making. Community engagement, energy-efficient cooking appliances, and fire prevention measures can all promote sustainable practices [77]. Essential actions include implementing government policies, forming partnerships with organizations, and allocating funding to afforestation and reforestation projects. Communities can benefit from their forest resources and protect the environment by supporting fair trade practices and enhancing market access [78]. To optimize the utilization of forest resources in less efficient regions of China, it is imperative to implement modern forestry equipment, develop comprehensive forest management plans, and establish processing facilities on the local level. Critical stages include obtaining forest certification, integrating monitoring and inventory management technology, and implementing agroforestry practices [79].

Furthermore, community engagement, reforestation efforts, ecotourism development, and efficient forest fire management can all contribute to increased revenue and environmental protection. In addition to access to markets and wildlife protection, government support, and research and development, these regions will experience sustainable forest resource management and economic expansion by implementing fair trade practices [80]. Our research can be situated within the broader context of the Environmental Kuznets Curve (EKC) theory. This study enhances our comprehension of the applicability of the Environmental Kuznets Curve (EKC) framework to distinct environmental dimensions, with a specific focus on sustainable forest management. Increasing evidence indicates a transition towards the more effective and sustainable exploitation of forest resources as economies advance, which is consistent with the Environmental Kuznets Curve (EKC) hypothesis. The EKC hypothesis posits that, during economic development, there is an initial exacerbation of environmental deterioration as a result of economic expansion. However, beyond a particular income threshold, the environmental conditions begin to improve. Regarding the efficiency of forest resources, this theory posits that as economies progress and ecological consciousness grows, there could be a shift towards more effective and sustainable techniques for the management of forests [81]. This shift has the potential to enhance the efficiency of forest resources. The transformation is primarily propelled by the expansion of the economy, which facilitates the advancement of research and development in sustainable forestry technologies and enforces rigorous environmental rules. Consequently, this leads to a reduction in waste and an improvement in overall efficiency. In conclusion, this study offers significant contributions to our understanding of the impact of economic growth on the efficiency of forest resource utilization throughout the progression of economies [82].

6. Conclusions and Policy Implications

The optimization of forestry resource usage has been a longstanding and prominent issue, as the establishment of sustainable resource management is a fundamental requirement in achieving economic and societal sustainability. Further, assessing technological diversity across different regions to maximize the utilization of forest resources is of great importance for giant economies like China. It allows tailored strategies, ensuring efficient, sustainable management and equitable development while considering regional differences. Moreover, estimating the determinant of total factor productivity in forest resource utilization in China is important in evaluating the success of dynamic productivity growth over time. Using regulations, afforestation, and investments in technology, the Chinese government has made considerable effort to improve the efficiency of forest resources, decrease technological disparities, and boost productivity growth. These measures aim to reduce regional disparities, foster sustainability, and standardize procedures. The success of these government efforts in enhancing the forest resource efficiency, reducing the technological heterogeneity, and increasing productivity growth in forest resource utilization is a promising yet undiscovered area worth exploring.

This study assesses the forestry resource efficiency, technology gap ratios in different regions, and total factor productivity growth in China’s 31 provinces/municipalities (2001–2020). SBM-DEA is used to evaluate resource efficiency; meta-frontier analysis is used to gauge the TGR by forest region; and the Malmquist index estimates dynamic efficiency changes. Input indicators include the forestry investment, workforce, and forested area, while output indicators are the forestry output value, timber production, and forest reserves. Chinese provinces have 45.70% potential to enhance their forest resource utilization efficiency (average FRE: 0.5430). The FRE fluctuated over time, with peak levels in 2015, 2016, and 2018. Anhui, Tibet, Fujian, Shanghai, and Hainan excelled in forestry utilization. In contrast, Qinghai, Gansu, Ningxia, Shaanxi, and Xinjiang were the poorest performers in terms of forestry efficiency. Further elaborating on the regional results, the study found that the average FRE in the southern forest region of China was at an optimum level, with an efficiency score of 0.7505. Similarly, the southwest (0.6019) ranked second, and the northeastern (0.4394) and northern (0.3329) regions ranked third and fourth, respectively. This indicates that, on average, provinces in the southern region are more efficient in utilizing their forest resources.

In a frontier analysis of the technological efficiency in China’s forest regions, the southern forest region emerges as the leader in modern forest resource utilization, with an impressive average technology gap ratio (TGR) of 0.915, approaching the optimal value of 1. This indicates that the southern provinces excel in harnessing forest resources through contemporary production technologies. Within this region, Anhui and Jiangsu have already reached the pinnacle of forestry efficiency, each achieving a TGR of 1, underscoring their advanced production technologies. In the northern region, which records the lowest TGR among the four regions, with an average of 0.4197, Tianjin stands out as a notable exception, maintaining superior technology with a TGR value of 0.7329. In contrast, Xinjiang lags with a TGR of 0.2206, making it the least efficient performer among the northern provinces. The northeastern region displays significant growth potential, with an average TGR of 0.4492, signifying room for improvement. Jilin, in particular, exhibits promise, boasting a TGR of 0.794, indicating its proximity to maximum efficiency. However, Inner Mongolia lags, with a TGR of 0.2271, denoting substantial room for advancement. Lastly, the southwestern forest region of China demonstrates moderate production technology, with an average TGR of 0.6512, ranking second after the southern region. Notably, Tibet stands out as the top performer in this region, achieving a TGR value of 1, signifying its prowess in maximizing its technological frontiers.

The Malmquist productivity index (MI) averaged 0.9644 (2001–2020), reflecting a 3.56% decline in forestry resource productivity in Chinese provinces. Technological change (TC) was the primary factor behind the productivity decrease, reaching 5.2%, while the efficiency change (EC) increased by 1.74%. The highest MI score was in the Southern Chinese forest region (1.0306), indicating a 3.06% average total factor productivity increase, driven by 3.38% EC growth. The northern forest region had the second-highest MI (1.0186), with EC growing by 4.4%, offsetting a 2.43% TC decrease. In contrast, the southwestern forest region saw a 1.25% total factor productivity decline, mainly due to an 8.81% TC decrease. Conversely, technical efficiency grew by 0.85%. Lastly, the northeastern forest region of China exhibited the lowest performance with a mean MI score of 0.9644, signifying a 3.56% total factor productivity decline. Technological change is the primary factor behind this decline, with a TC value of 0.948. Similar to the southwestern region, there was 1.74% growth in technical efficiency. Finally, the results of the Kruskal–Wallis test proved the significant statistical differences among the four forest regions of China regarding the FRE and TFP changes.

The government’s commendable efforts through regulations, afforestation, and technology investments should continue to reduce regional disparities, foster sustainability, and standardize practices. The varying efficiency and productivity highlight the need for targeted interventions, particularly in less efficient provinces, promoting technological investments. High-efficiency provinces can serve as models for sustainable practices, with capacity-building and technology transfer programs bridging gaps in less efficient regions.

Moreover, encouraging research and development in forest management technologies is essential to complement China’s efficient and sustainable resource utilization efforts. In conclusion, this study’s findings offer valuable insights for policymakers to refine their strategies, promoting sustainable resource management and fostering economic and environmental sustainability in this vital sector. While some regions in China have excelled in their technological utilization of forest resources, others lag, indicating the need for technology transfer and knowledge sharing from more advanced areas. Provinces that would benefit from acquiring technology and expertise from their more proficient counterparts include Inner Mongolia and Xinjiang in the northern region, as they have shown lower efficiency and technology gaps. Similarly, the northeast region, with provinces like Jilin and Inner Mongolia, could benefit from adopting best practices from the southern and southwestern regions. These provinces could leverage the experiences of Anhui, Jiangsu, and Tibet, which have demonstrated advanced efficiency and technology utilization, to bridge the technological gap and improve their overall forest resource productivity. Effective technology transfer programs and capacity-building initiatives can facilitate this process, fostering the more equitable and efficient utilization of forest resources across all regions in China.