4.3.3. Replace the Dependent Variable

In 2017, President Xi Jinping proposed five development goals of "innovation, coordination, greenness, openness, and contribution" while considering economic development, achieving resource conservation, and environmental protection. This paper adopts the method of replacing the explained variables to verify further the impact of forest resource abundance on economic development. This section replaces GDP per capita with green total factor productivity (GTFP), which is essentially one of the ways to measure green economic growth [58]. Meanwhile, to examine the linear and non-linear effects of forest resource abundance on green total factor productivity, the core explanatory variable CORE first considers the forest coverage rate FT alone. It then simultaneously considers the forest coverage rate FT and its square. In order to stabilize the results and alleviate the endogeneity problem, the static and dynamic SAR and SDM are constructed as follows. The results are shown in Table 8.

$$GTFP\_{it} = \rho \sum\_{j=1}^{n} W\_{ij}'GTFP\_{jt} + \beta''CORE\_{it} + \beta''CONT\_{it} + \theta \sum\_{j=1}^{n} W\_{ij}'CONT\_{ijt} + \mu\_i + \xi\_t + \varepsilon\_{it} \tag{15}$$

$$GFFP\_{\rm if} = \rho \sum\_{j=1}^{n} W\_{\rm ij}^{\prime} GFFP\_{\rm jt} + \beta^{\prime\prime} CORE\_{\rm if} + \theta \sum\_{j=1}^{n} W\_{\rm ij}^{\prime} CORE\_{\rm ijt} + \beta^{\prime\prime} CONT\_{\rm if} + \theta \sum\_{j=1}^{n} W\_{\rm ij}^{\prime} CONT\_{\rm ijt} + \mu\_{\rm i} + \xi\_{\rm i} + \varepsilon\_{\rm it} \tag{16}$$

$$\begin{aligned} \text{GTFP}\_{it} &= \pi G T F P\_{i, t-1} + \rho' \sum\_{j=1}^{n} W\_{ij} G T F P\_{j, t-1} + \rho \sum\_{j=1}^{n} W'\_{ij} G T F P\_{jt} + \beta'' \text{CORE}\_{it} + \theta \sum\_{j=1}^{n} W'\_{ij} \text{CORE}\_{ijt} \\ &+ \beta'' \text{CONT}\_{it} + \theta \sum\_{j=1}^{n} W'\_{ij} \text{CONT}\_{ijt} + \mu\_i + \tilde{\xi}\_t + \varepsilon\_{it} \end{aligned} \tag{17}$$

where Equations (15)–(17) represent the static SAR, the static SDM, and the dynamic SDM, respectively. The explained variable GTFP is green total factor productivity. The core explanatory variables, control variables, other variables, and symbols are consistent with the benchmark model, and the spatial weight matrix adopts the Gaussian kernel function distance weight matrix.

As shown in Tables 7 and 8, the signs of all variables are basically unchanged. The phenomenon of economic convergence exists at the city level in the YRD region, and there is a lot of room for improvement in areas with relatively backward economies. The impact of forest resource abundance on the level of economic development has always maintained a U-shaped trend. Forest resources at the urban level in the YRD will play different roles in different stages of economic development, especially under the goal of low-carbon economic development. Resources play a pivotal role in carbon sequestration. This proves that the results of the model argument are robust. After the optimization and upgrading of the industrial structure, the energy utilization rate is high, and the investment in urbanization and science and technology education can effectively promote the economic level of the YRD region.

Table 9 shows the results of the influence of forest resource abundance on the green total factor productivity. SAR (1) and SAR (2) examine the linear and non-linear relationship between the abundance of forest resources and the level of green economic development. SDM (1) and SDM (2) represent the static and dynamic models, respectively. The results of the models show inertia in developing a green economy in the YRD region. The Ushaped characteristics of the impact of forest resource abundance on green economic development have not been verified. However, the impact of forest resource abundance on the development of the green economy is always positive. It shows that increasing the abundance of forest resources in the YRD will help improve its green economic development level and contribute to sustainable economic development. Greening is conducive to solving the trade-off dilemma of "economic growth, environmental friendliness, and resource conservation" in economic development. This is also in line with China's current "14th Five-Year Plan" and the strategic need for sustainable economic development. All results prove that the influence of forest resource abundance on economic development in the YRD region presents a U-shaped feature, and forest resource abundance is conducive to improving green total factor productivity.


**Table 9.** Estimation results of effects of forest resource abundance on GTFP.

Note: \*\*\* significant at the 1% level; \*\* significant at the 5% level; \* significant at the 10% level; standard errors are in parentheses.

#### *4.4. Reverse Causation*

In order to verify the impact of economic development on forest resource abundance, this paper constructs the dynamic SDM model, where the dependent variable is the Forest Resource Abundance (FT), and the core explanatory variables are the economic development, namely GDP per capita (GDPPC) and its square to adopt the environmental Kuznets curve. Other control variables are kept consistent with the baseline model. The spatial weight matrix adopts the queen adjacent weight matrix. Furthermore, this paper also examines the linear relationship between economic development and forest resource abundance. All variables are standardized. The estimation results of the SAR model are also shown to check the robustness of the results.

As shown in Table 10, although there is a U-shaped curve between the level of economic development and the abundance of forest resources, the results are not stable. However, the level of economic development inhibits the abundance of forest resources. There are two main reasons. First, the cost of land-use is high in YRD. This region contributes a lot to the national economy and is one of the important urban agglomerations in China. The total economic volume accounts for 25% of the country's total. In addition, the two most powerful harbors in the world, Shanghai and Zhoushan, are located here. Among the top 20 cities in terms of GDP, cities in the YRD region account for one-third at least. However, the area of such an economically developed region is only 358,000 km2, accounting for about 4% of the country's total. Second, forest resources are limited by land and depend largely on territorial planning, which is difficult to change in the short term. The growth of forest resources has always been a key concern of forestry. However, from the perspective of economic development and forest land use, large, continuous forests have the potential to be transformed into smaller, isolated fragmentation processes, which exacerbate the degree of forest fragmentation [59].


**Table 10.** Estimation results of effects of economic development on forest resource abundance.

Note: \*\*\* significant at the 1% level; \*\* significant at the 5% level; \* significant at the 10% level; standard errors are in parentheses.
