*4.4. Discussion*

In China, slightly growing and slightly shrinking cities account for the majority, while rapidly growing and rapidly shrinking ones account for a small proportion. Growing cities are mostly distributed in the eastern region and shrinking cities in the northeastern, central, and western regions. This is consistent with Yang et al. [36]. After China's "reform and opening up," the eastern region attracted an inflow of FDI (foreign direct investment) by virtue of geographical advantage. This created many jobs, and many laborers came there in search of work, prompting rapid urban expansion. The northeast, as the old industrial base of China, has long relied on mineral resources for development. In recent years, with the exhaustion of resources and the emergence of excess capacity, cities have been forced to adjust their industrial structures. However, such adjustment has been sluggish, and new economic growth points have not been fostered. As a result, many people lost their jobs and left the region, resulting in the emergence of shrinking cities. The central and western regions are relatively closed geographically and backward in their economic development. A large number of laborers thus migrated out to seek employment opportunities, and urban populations gradually shrunk.

Growing and shrinking cities have different carbon-emission characteristics. The emissions of RGCs show a trend of fluctuating growth, while RSCs present an inverted U-shaped trend. SGCs and SSCs show a trend of first rising and then developing steadily. This differs from Xiao et al. [30] but supports Qiang et al. [42], namely, that air pollution is reduced in shrinking cities. RGCs have developed economies and high urbanization levels, which all contribute to increases in carbon emissions. At the same time, the rapid development of transportation and tertiary industry also increase emissions. As for RSCs, most depend on heavy industry. With the exhaustion of resources and the emergence of overcapacity, cities have been forced to reduce the scale of secondary industry, causing urban economies to decline rapidly, which correspondingly reduces emissions. The economic growth of SGCs and SSCs is relatively flat, and technology is improving, causing emissions to tend to be stable.

The results indicate that economic development and population growth positively affect emissions in the four city groups, while technological progress has a negative effect. The proportion of tertiary industry positively affects emissions in growing cities and negatively affects them in shrinking cities. Supporting Zheng et al. [43], economic development and population growth both increase energy consumption and therefore play obvious roles in increasing emissions. Technological advancement can produce more output using less energy, which reduces carbon emissions. The reason the industrial structure of different groups has different effects on emissions might be related to tertiary industry development. If tertiary industry is fully developed, services such as urban transport will require more energy, contributing to higher emissions. However, if urban tertiary industry development is immature, tertiary industry will consume less energy than secondary industry, which will reduce emissions.

#### **5. Conclusions and Policy Implications**

Aiming to support emission-reduction policy making, this study investigates the characteristics of and factors affecting carbon emissions in growing and shrinking Chinese cities. Taking 280 cities as samples, a UDD index developed by the authors was used to divide the cities into four groups: RGCs, SGCs, RSCs, and SSCs. Emission characteristics are discussed in terms of economics, population, energy intensity, and industry structure. The main findings are summarized below.

For RGCs, their economies and populations grow rapidly, their industrial transformation takes place early, and their tertiary industry development is sufficient. Correspondingly, carbon emissions show a fluctuating growth trend. The regression results show that the proportion of tertiary industry has a significant positive effect on carbon emissions, while other factors have no significant effects. For RSCs, economic growth is declining, population loss occurs with the decline of secondary industry, and industrial structure adjustment is belated. Their carbon emissions show an inverted U shape. The regression results show that economics and population have significant positive effects on emissions, while technology and the proportion of tertiary industry have significant negative effects. For SGCs, their economies and populations have both grown steadily, and tertiary industry

has developed continuously with the optimization of industrial structure. Their carbon emissions first rise and then develop steadily. The regression results show that economics and population have significant positive effects on emissions, while technology has a significant negative effect. For SSCs, the level of economic development is low but continues to grow, and the population tends to decrease. Their carbon emissions are similar to those of SGCs, first rising and then developing steadily. The regression results show that economics and population have a significant positive effect on emissions, while technology and the proportion of tertiary industry have a significant negative effect.

The policy implications are as follows. For growing cities with sufficient human capital, they can vigorously develop industries with high-added value and low-carbon emissions. Meanwhile, the government should devote more financial funds to improving the level of low-carbon technology. For shrinking cities, future policies should continue to optimize the industrial structure and encourage the development of the tertiary industry. In addition, the government should increase subsidies for high-quality talents to provide sufficient human capital for technological upgrading.

This study has some limitations. First, in the accounting of urban carbon emissions, only energy-related carbon emissions are accounted, and the carbon emissions generated in cement production processes are not accounted. The accounting scopes will be further expanded to improve data quality in future work. Second, in terms of the influencing factors of carbon emissions, this study focuses on social factors, such as population, economy, and technology. In fact, urban growth and shrinkage, as a comprehensive reflection of population, economic, and social changes, have also been proven to have an impact on carbon emissions and environmental conditions [31,42]. In the future, we will further explore the impact of urban growth and shrinkage on carbon emissions.

**Author Contributions:** Conceptualization, methodology, formal analysis, writing—original draft preparation, S.G.; writing—review and editing, supervision, X.T. and H.D.; data curation, Z.D., C.G. and W.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by National Natural Science Foundation of China (Grant Numbers: 71773034 and 71704157).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

