*2.2. Analysis of Measurement Results*

According to the specific calculation method introduced above, the provincial level of energy consumption in China from 1999 to 2020 is calculated firstly. To visually describe the characteristics of the temporal–spatial evolution in the provincial level of energy consumption, the measurement results are reported in the form of topographic maps. Figure 1 shows the concrete results. Due to space constraints, the results from 1999, 2007, 2013, and 2020 are only displayed. It can be clearly seen in Figure 1 that the dark blue areas represent the highest level of energy consumption in 30 provincial administrative units. Obviously, the

number of dark blue regions increased from zero in 1999 to ten in 2020. For the convenience of discussions, all provincial administrative units are called "province". Therefore, we can conclude from these results that the provincial level of energy consumption in China is on the rise over the study period [43,49]. And in 2020, the provincial level of energy consumption demonstrated visible spatial differentiation. Among the provinces with the highest level of energy consumption, four provinces lie in the western region: Xinjiang, Qinghai, Gansu, and Inner Mongolia. Four provinces, Beijing, Shanghai, Hebei, and Jiangsu, are in the eastern region. Only one province, Shaanxi, lies in the central region, and only one province, Liaoning, lies in the northeastern region. Generally, the western region has the highest level of energy consumption among the four types of regions, followed by the eastern region. And the rest of the two regions have comparatively lower levels of energy consumption. The provincial level of energy consumption from the eastern to the central, northeastern, and western regions is similar to the U-shaped curve [49]. The spatial characteristics of the provincial level of energy consumption imply preliminarily that it is necessary to give thought to regional heterogeneity on this subject.

**Figure 1.** Chinese provincial level of energy consumption for some years (million tons).

Then, the comprehensive urbanization index is calculated by the widely used entropy method. To visually describe the characteristics of the temporal–spatial evolution of this index, the measurement results are reported similarly in the same way. Figure 2 shows the concrete results. Due to space constraints, the results from 1999, 2007, 2013, and 2020 are only displayed. It can be clearly seen in Figure 2 that the dark blue areas represent the highest level of new-type urbanization in 30 provincial administrative units. Obviously, the number of dark blue regions is zero in 1999, yet nearly all provinces in the eastern and central regions belonged to the highest group in 2020. Therefore, we can conclude from these results that the provincial level of new-type urbanization has also

increased greatly over the study period, especially after the year 2013 [2,4,49]. On the one hand, the main reason for these time series characteristics can be probably ascribed to the "National New-type Urbanization Plan", which makes the economy, society, and ecology balanced in the process of urbanization. On the other hand, the increase in the level of energy consumption may also be conducive to speeding up the process of urbanization. And in 2020, the provincial level of new-type urbanization has also displayed significant spatial differentiation. Specifically, the majority of provinces have achieved the highest status in 2020. And only a few provinces have not reached the highest level, these are Xinjiang, Jilin, Shanxi, Gansu, Guizhou, Qinghai, Yunnan, and Guangxi. Additionally, these provinces mainly lie in the western and central regions. The average comprehensive urbanization index of the provinces in the eastern, central, northeastern, and western regions is calculated to be 0.299, 0.169, 0.188, and 0.159, respectively, which verifies that the provincial level of new-type urbanization looks like the inverted S-shaped curve, which is slightly different with the provincial level of energy consumption [46,47]. Therefore, the spatial characteristics of the provincial level of new-type urbanization further imply that it is necessary to give thought to regional heterogeneity on this subject.

**Figure 2.** Chinese provincial level of new-type urbanization for some years.

To display the tendency intuitively, the growth rate of the above measured variables is depicted with a scatter diagram in Figure 3 for the four groups, eastern (Figure 3a), central (Figure 3b), northeastern (Figure 3c), and western regions (Figure 3d), respectively. In Figure 3, the relationship between comprehensive urbanization index growth and energy consumption growth is significantly different in the four groups. Specifically, there seems to be a strong relationship between the northeast and western regions, yet no such relationship emerges in the eastern and central regions. These results further prove that we cannot neglect the "one size for all" homogeneity issue among provinces when investigating the causal relationship between the two, and it is reasonable to classify the data into four groups: east, northeast, central, and west. Secondly, the weak relationship between the two in the eastern and central regions denotes there may be a nonlinear relationship between comprehensive urbanization index growth and energy consumption growth, and the traditional linear model is not suitable for this subject. Nevertheless, it should be noted that these kinds of scatter charts roughly reflect the possible correlation relationship between the two, and the exact relationship between the two remains to be confirmed rigorously by employing reasonable econometric models.

**Figure 3.** Comprehensive urbanization index growth vs. energy consumption growth.

### **3. PVAR Model Regression Results and Analysis**
