Urban Shrinkage from the Perspective of Economic Resilience and Population Change: A Case Study of the Shanxi-Shaanxi-Inner Mongolia Region

Abstract

With the increasing uncertainty of urban development, urban shrinkage in the rapid urbanization process in China has become increasingly serious. While many studies have explored urban shrinkage from the economic and population perspectives, they often ignore the essence of the phased evolution of economic and population factors. Thus, this study introduces the theory of economic resilience into the field of urban shrinkage and constructs a theoretical method for identifying urban shrinkage by integrating economic resilience and population change to reveal the evolutionary trajectory of regional urban growth and shrinkage. The results show that urban economic resilience and population change in the Shanxi-Shaanxi-Inner Mongolia region (SSIMR) exhibit strong volatility, highlighting the importance of conducting urban shrinkage studies within specific crisis disturbance scenarios. In the context of the “new normal of the economy”, the economic resilience of cities in the SSIMR has significantly declined, and the problem of economic recession is gradually intensifying. The population change trend of cities in the SSIMR is relatively stable, with population loss being a common problem in urban development in the area and its scope and intensity increasing daily. Urban development in the SSIMR is evolving from global growth to widespread shrinkage, with 56.67% of the cities experiencing relative shrinkage, showing a spatial pattern of “western growth–eastern shrinkage”. Factors such as the agglomeration effect, industrial structure, and policy system collectively shape the evolution of urban growth and shrinkage.

1. Introduction

Urbanization represents a significant trend in society’s development in the 21st century, particularly in developing countries. However, influenced by factors such as deindustrialization, resource depletion, and political shifts, the countervailing phenomenon of urban shrinkage, which is marked by population loss and economic recession, is constantly emerging in various corners of the world [1,2,3], posing severe challenges to the conventional notion of urban growth. Investigating the evolutionary trajectory and driving mechanism of urban growth and shrinkage is the scientific cornerstone for understanding the changes in complex urban systems and is also the frontier area for the “Future Earth” and the United Nations 2030 Sustainable Development Goals [2,4]. Since the reform and opening-up, China’s urbanization construction and urban development have achieved remarkable results, realizing the prosperity and development of the urban economy, society, and culture. However, there it also faces problems such as disproportionate scales, structural imbalances, and extensive inefficiency [5,6]. In recent years, the domestic and international environments have undergone profound and complex changes, intensifying the risks and uncertainties encountered in China’s urban development. The phenomenon of urban shrinkage is also increasing daily, so enhancing resilience has become a common consensus for urban systems to navigate shrinkage challenges, foster successful transitions, and achieve sustainable development [7,8].

As the core attribute of urban systems, economic resilience serves as the “guide indicator” and “quality inspector” of urban development, providing a new perspective from which to analyze the evolutionary differentiation of urban prosperity and decline [9,10]. At present, economic resilience research mainly contains two levels of theoretical discussion and empirical research, involving concept definition, resilience measurement, and analysis of influencing factors [11]. In terms of conceptual connotation, economic resilience has gradually shifted from the equilibrium theory of engineering resilience and ecological resilience to the evolutionary theory of adaptive resilience, with greater emphasis on the dynamic process of adaptation, adjustment, and learning within the economic system [12,13]. This transition has given rise to perspectives that highlight dynamic evolution, multi-agent interaction, and multi-scale integration [14]. At the level of empirical research, existing studies mainly use the core variable method and the comprehensive indicator method to measure urban economic resilience, and the studies have found that factors such as industrial structure, innovation capacity, and government regulation have strong explanatory power [15,16,17]. In recent years, with the integration of urban systems theory and resilience theory, economic resilience has emerged as a new concept for measuring the transformation benefits and development capability of cities, which is widely applied in the fields of urban development path creation, industrial transformation, and economic growth [17,18,19]. Studies have shown that cities with stronger economic resilience can effectively resist crises, maintain stability, and even achieve transformation and upgrading. In contrast, cities with weaker economic resilience are more vulnerable and prone to falling into recession crises [11,20]. Therefore, the theory of economic resilience helps elucidate the differentiation of urban development triggered by regional structural crises, thereby providing theoretical support for the dynamic evolution of urban growth and shrinkage [9].

The study of urban shrinkage originated in the late 20th century, when the concept was first proposed by Häußermann and Siebel [21], and has since been applied and promoted. At present, relevant research involves the fields of conceptual connotations [22], identification methods [23], spatiotemporal patterns [23], type models [24], influencing factors [25], and planning responses [26], forming a relatively complete theoretical framework and set of coping strategies to address urban shrinkage. However, owing to the complexities and particularities of urban shrinkage, the regional differences in its connotation, identification criteria, and formation mechanism are evident [3]. An important topic for urban geography research in the new era is how to scientifically recognize the problem of urban shrinkage and construct a theoretical framework for urban shrinkage with regional characteristics [3,4]. Currently, China is undergoing a distinctive phase of rapid urbanization accompanied by partial shrinkage, in which urban shrinkage is mainly caused by factors such as resource depletion, financial crisis, and economic transformation, manifesting as population loss and economic slowdown within a specific period [27,28]. Its essence is the outflow of urban production factors and the weakening of development potential triggered by internal and external disturbance factors [3,6], which is significantly different from the essence of urban “decline” in European and American countries. Therefore, Chinese scholars have utilized “localization” thinking [28,29] and analyzed urban shrinkage issues in resource-based cities [30], old industrial bases [31], export-oriented urbanization areas [32], underdeveloped areas [33], and other locales, contributing “Chinese wisdom” to resolve the problem of regional urban shrinkage [34,35]. However, there is still significant controversy regarding the diagnostic criteria for urban shrinkage, and such research has usually taken 5 years, 10 years, or a certain time frame as the research cycle, focusing on the appearance of the increase or decrease in digital scale but ignoring the structural crises that are the essence of urban shrinkage, due to which the integrity of urban evolutionary cycle becomes fragmented [36,37]. Furthermore, while there is an increasing focus on the issue of relative urban shrinkage in the Chinese context, research findings are still limited [38,39].

Given this, this paper introduces the theory of economic resilience into the field of urban shrinkage and analyzes the evolutionary trajectory and influencing mechanism of regional urban growth and shrinkage from the perspective of relative shrinkage to enrich the perspective of urban shrinkage research, deepen the theoretical understanding of urban shrinkage, and then provide a scientific reference for the urban transformation and sustainable development in the SSIMR.

2. Theoretical Framework

In the globalized world, with flow elements as the main feature and basic operation logic, urban development paths are becoming increasingly diversified, leading to the spatial differentiation phenomenon of growth and shrinkage [30]. Rational cognition of urban shrinkage and a scientific response to the risk of urban shrinkage have become important issues that need to be resolved in urban planning and urban governance [2,3]. According to the theory of urban growth, urban development is a spatial agglomeration process of production factors such as labor and capital, and population increase and economic development are the core manifestations of urban growth [40]. Consequently, urban shrinkage can be understood as a process of urban population loss and economic recession triggered by internal and external disturbance factors [2,8]. Currently, China is stepping into the stage of high-quality development, and urban development is facing a systematic reconstruction of its elements, structures, and functions. Under the interactive influence of factors such as production factors agglomeration, industrial structure transformation, and policy system reform, the risks and challenges faced by Chinese urban development are intensifying, and the spatial flow paths of urban production factors have changed significantly, triggering a new spatial phenomenon of urban growth and shrinkage merging [32]. There is an urgent need for new theoretical methods to analyze the differentiation of regional urban development. Economic resilience, as the economic system’s capacity to resist shocks, maintain stability, adjust adaption, and recover development when suffering from crises [10], is an intuitive manifestation of urban transformation benefits and development capabilities [17,19]. It provides a new basis for dividing the evolutionary cycle of urban systems and offers a new theoretical perspective for explaining the differentiation of urban growth and shrinkage triggered by regional structural crises [9]. In this regard, this paper argues that comprehensively considering the complex relationship between population change and economic resilience can help to accurately understand the evolution of urban growth and shrinkage under crisis disturbance scenarios.

Population and economic factors play a pivotal role in urban development, as they are interdependent and mutually constrained, collectively fostering the evolution of urban systems. Among them, people are the core of urban development, and population loss, as a result of the comprehensive effects of factors such as urban industrial recession, increased unemployment, and environmental degradation, is the most intuitive manifestation of urban shrinkage [37,41]. However, there are also obvious limitations in measuring urban shrinkage only through changes in population size. For example, it fails to explain the urban development brought about by the improvement of labor efficiency, and economic restructuring may also promote the transformation and development of cities experiencing population loss. Moreover, the organized migration of individuals in labor-rich areas itself contributes to the efficient development of cities [32,42]. Economic development is the foundation of urban development, and economic resilience, as an intuitive manifestation of urban economic growth potential, mirrors the prosperity and decline of cities [9,19]. It is an important basis for understanding urban growth and shrinkage, but a single indicator of economic resilience is inadequate to explain the phenomenon of population loss in urban growth scenarios. In addition, there exists a complex interactive feedback mechanism between population and economic factors [38,43]. Population size and quality have a strong supporting effect on urban economic development, affecting the evolution of urban economic resilience. Economic resilience characterizes the potential of urban economic development, guiding the flow direction and rate of population factors [9,44], both of which influence each other and evolve circularly, promoting the synergistic evolution of the flow trend of production factors and development potential among cities and then resulting in the relative shrinkage of low-efficiency cities [39].

Given this, this paper starts from the essence of regional structural crises and constructs a theoretical framework for analyzing urban growth and shrinkage under crisis disturbance scenarios (Figure 1). Subsequently, it divides the evolutionary cycle of regional urban systems from the dimensions of economic resilience and population change and explores the evolutionary process and influencing mechanism of regional urban growth and shrinkage, aiming to provide theoretical support for the high-quality development of cities in the SSIMR.

3. Materials and Methods

3.1. Study Area

The SSIMR is located in the transitional zone of agriculture and animal husbandry in inland China. It is an important area for the integration of Chinese agricultural civilization and grassland nomadic civilization, including the three modern provinces of Shanxi, Shaanxi, and Inner Mongolia, with a total area of 154.53 × 10⁴ km² (Figure 2). By the end of 2019, the region’s total population reached 10,144.80 × 10⁴, the population density reached 65.65 people/km², the per capita GDP was 59,175.54 yuan/person, and the ratio of three industrial structures (primary, secondary, and tertiary industries) was 7.79:43.73:48.48. The region is rich in energy resources and intensive in energy and chemical industries, making it an important energy strategic guarantee base in China and one of the most densely populated areas of resource-based cities [45]. The large-scale development of energy resources has exerted a strong impetus on regional economic and social development. However, the long-term economic development model of “relying on energy and heavy industry” has led to obvious characteristics of extensive development, such as slowing economic growth and increasing pressure on industrial transformation [46]. As China’s economic development steps into the “new normal”, urban development in the SSIMR is facing a systematic reconstruction and structural transformation of the internal and external environment. The rate, structure, and kinetic energy of urban growth are continuously evolving, leading to notable disparities in development trajectories among cities [46,47,48].

3.2. Methods

3.2.1. Economic Resilience Measurement

The core variable method measures urban economic resilience by the difference between the actual and expected changes of GDP during the crises, which not only reflects the ability of the economic system to resist crises and adapt to the recession but also indicates the ability of the economic system to achieve transformation and recovery development. It is a comprehensive expression of the potential of urban economic development [49,50]. In this paper, it is assumed that urban GDP changes according to the national average level; then, the expected change of GDP in city i during a recession or recovery period in k years, given as follows:

$${\left({∆E}_i^{t+k}\right)}^e=E_i^t\times G^{t+k}$$

(1)

where ${\left({∆E}_i^{t+k}\right)}^e$ is the expected change of GDP in city i from year t to t + k. $E_i^t$ is the GDP of city i at the starting year t, and $G^{t+k}$ is the national GDP growth rate from year t to t + k. Then, the measure of urban economic resilience can be expressed as follows:

$${\mathit{Res}}_i=\frac{{\left({∆E}_i^{t+k}\right)}^r-{\left({∆E}_i^{t+k}\right)}^e}{\left|{\left({∆E}_i^{t+k}\right)}^e\right|}$$

(2)

where ${Res}_i$ is the economic resilience of city i. If ${Res}_i$ is a positive value, the economic resilience of city i is strong, and if ${Res}_i$ is a negative value, the economic resilience of city i is weak. ${\left({∆E}_i^{t+k}\right)}^r$ is the actual change of GDP in city i from year t to t + k.

3.2.2. Population Change Measurement

People are the core element of urban development. Combined with the actual situation of population loss in the SSIMR and the macro context of China’s population growth, this paper uses the annual permanent resident population to reflect urban population change to eliminate the influence of overall population growth, and the relative change rate of the annual permanent resident population is used to measure the intensity of urban population change, reflecting the spatial flow trend of urban population factors [36,51]. The formula is as follows:

$$D=\mathsf{\Delta }p-\mathsf{\Delta }P$$

(3)

$$I=\frac{D}{\left|\mathsf{\Delta }P\right|}$$

(4)

where D is the difference of the annual permanent resident population change rate between the cities and nation. $\mathsf{\Delta }p$ is the change rate of the annual permanent resident population in cities, and $\mathsf{\Delta }P$ is the annual change rate of the national population. I is the relative change intensity of the annual permanent resident population; when I > 0, it means that the urban population is growing relatively, and vice versa.

3.2.3. Identification of Urban Growth and Shrinkage

Urban development is a complex and systematic process. Drawing on relevant studies [32,35], according to the relationship between economic resilience and population change, all case cities were divided into four types: significant growth, smart growth, slowing growth, and significant shrinkage (

Table 1. Criteria and characteristics of urban types.

Type	Economic Resilience (Resi)	Population Change (I)	Characteristics
Significant growth	Resi ≥ 0	I ≥ 0	Strong economic resilience and agglomeration of population factors
Smart growth	Resi ≥ 0	I < 0	Strong economic resilience and loss of population factors
Slowing growth	Resi < 0	I ≥ 0	Weak economic resilience and agglomeration of population factors
Significant shrinkage	Resi < 0	I < 0	Weak economic resilience and loss of population factors

).

(1)

Significant growth: strong economic resilience and agglomeration of population factors. Cities of this type have a good industrial foundation and strong economic growth momentum. Even under the influence of internal and external disturbance factors, the urban system still shows strong resilience, maintains the trend of economic development, and constantly attracts the inflow of production factors such as population and capital;

(2)

Smart growth: strong economic resilience but loss of population factors. This type of city is usually an area with abundant labor resources or that has undergone successful industrial transformation. With the changes in the internal and external environment of regional development, the elements, organization, and structure of the urban economic system are constantly reconstructed, and the transformation of traditional industries and the cultivation of new development paths are accelerating, resulting in the enhancement of the urban economy resilience and efficient allocation of labor resources. Consequently, the migration of surplus labor force is beneficial for urban economic and social development;

(3)

Slowing growth: weak economic resilience but agglomeration of population factors. These cities typically possess a solid economic foundation, but under the interactive influence of internal and external disturbance factors, the economic resilience of the city is weakened, and the economic development is seeing a short-term decline. However, it still has a strong ability to gather factors. If the economic decline persists, it may induce shrinkage problems such as population loss;

(4)

Significant shrinkage: weak economic resilience and loss of population factors. Cities of this type have a poor industrial foundation and insufficient economic growth momentum. Affected by the regional structural crises, the urban system has been severely impacted, presenting problems such as economic recession and population loss. It is urgent to achieve sustainable urban development through transformation.

3.2.4. Selection of Factors Influencing Urban Growth and Shrinkage

The evolution of urban growth and shrinkage is the result of the interaction of multiple factors, influenced by factors such as agglomeration effect, industrial structure, and government regulation [32,34]. Therefore, based on considering the regional economic and social development situation and the accessibility of data and drawing on relevant research results [8,9,34], this paper starts from the three dimensions of factor agglomeration, industrial structure, and policy system and then selects nine indicators, namely population agglomeration, capital agglomeration, industrialization level, service industry development, resource industry dependence, opening-up, government regulation, innovation guarantee, and institutional locking, to analyze the influencing mechanism of urban growth and shrinkage in the SSIMR.

Table 2. Definition and descriptive statistics of the independent variables.

Dimensions	Variables	Description and Measurement Method	Mean	Std.
Factor agglomeration	Population agglomeration (Lpop)	Initial population size of the municipal district (10,000 people)	97.93	102.79
	Capital agglomeration (Inves)	Fixed asset investment per capita (10,000 yuan)	2.89	2.19
Industrial structure	Industrialization level (Indus)	Industrial added value/GDP (%)	45.45	12.77
	Service industry development (Servi)	Service industry added value/GDP (%)	36.15	9.63
	Resource industry dependence (Resou)	Proportion of mining industry employees (%)	13.92	12.80
Policy system	Opening-up (FDI)	Actual utilization of foreign investment/GDP (%)	1.24	1.28
	Government regulation (Gover)	Government fiscal expenditure/GDP (%)	17.85	5.95
	Innovation guarantee (Techo)	Science and technology expenditure/government fiscal expenditure (%)	0.91	0.61
	Institutional locking (Uempl)	Proportion of employees in state-owned and collective units (%)	38.42	14.42

presents the detailed information for each variable. Among them, population agglomeration and capital agglomeration are represented by the initial population size of the municipal district and the fixed asset investment per capita, respectively, reflecting the population agglomeration capacity and capital agglomeration capacity of the city, which can help explain the impact of factor agglomeration on urban development. Three indicators, namely the proportion of industrial-added value in the GDP, the proportion of service-industry-added value in the GDP, and the proportion of mining industry employees, respectively, characterize the level of industrialization, service industry development, and resource industry dependence, which can explain the impact of industrial structure on urban development. Four indicators, namely the proportion of actual utilization of foreign investment in the GDP, the proportion of government fiscal expenditure in the GDP, the proportion of science and technology expenditure to government fiscal expenditure, and the proportion of employees in state-owned and collective units, respectively, characterize the level of urban opening-up, government regulation and control ability, innovation environment guarantee, and market system locking, which can reveal the influencing mechanism of policy system in urban development.

3.3. Data Sources

To scientifically reveal the impact of crisis disturbances on urban shrinkage and avoid the problem of the incomparability of urban shrinkage intensity in different shrinkage periods, the study period was finally determined to be 2008–2019. The relevant data were mainly collected from the China Statistical Yearbook, China City Statistical Yearbook, Shanxi Statistical Yearbook, Shaanxi Statistical Yearbook, and Inner Mongolia Statistical Yearbook from 2009 to 2020. Missing data were supplemented by interpolation. To eliminate the impact of inflation, the price data were flattened using 2008 as the base year. The vector data of administrative divisions were obtained from the 1:1 million database of the National Basic Geographic Information Center (http://www.ngcc.cn.ngcc/, accessed on 27 March 2023).

4. Results

4.1. Spatiotemporal Evolution of Urban Growth and Shrinkage in the SSIMR

Regional urban growth and shrinkage exhibit significant spatiotemporal heterogeneity. Accurately identifying the development cycle of cities is the scientific basis for understanding the differentiation of urban systems evolution [8,28]. Therefore, based on the logical idea of interactive feedback between “regional environmental evolution and changes in urban economic resilience and population factors”, this paper identifies the evolutionary cycle of urban growth and shrinkage in the SSIMR, explores the spatiotemporal characteristics of urban economic resilience and population change during different disturbance cycles, and then reveals the evolutionary trajectory of urban growth and shrinkage in this region.

4.1.1. Temporal Characteristics of Urban Economic Resilience and Population Change

(1)

Economic resilience. From the perspective of urban economic development potential characterized by economic resilience, the economic recession of cities in the SSIMR is the product of the interaction of internal and external disturbance factors, with obvious phases, volatility, and periodicity (Figure 3). From 2008 to 2009, the number of cities experiencing economic recession significantly increased, indicating that the problem of urban economic recession in the SSIMR was gradually worsening, which is closely related to the impact of the global financial crisis and the insufficiency of local economic response capacity. From 2010 to 2011, governments at all levels formulated large-scale economic stimulus plans to cope with the risk of economic recession caused by the financial crisis, which promoted economic recovery of the SSIMR, and the urban economic development showed a comprehensive growth trend. Subsequently, the effect of the economic stimulus policies gradually weakened, and the speed of urban economic development slowed down, with the phenomenon of economic recession occurring in individual cities. From 2014 to 2019, China’s economic development stepped into the new normal, facing great challenges of speed shifting, structural transformation, and kinetic energy conversion. The energy and heavy chemical industry bases of the SSIMR suffered particularly severe impacts, with a significant increase in cities experiencing economic recession, reaching its peak in 2019, accounting for 53.33%;

(2)

Population change. From the perspective of the flow trend of urban population factors reflected by the relative change rate of annual permanent resident population, the problem of urban population loss in the SSIMR is relatively severe and highly volatile. From 2008 to 2010, the number of cities experiencing population loss in the SSIMR significantly decreased, mainly due to the outbreak of the financial crisis, which caused a huge impact on the export-oriented economy in the eastern coastal areas of China, resulting in an increase in enterprise closures and a decrease in employment opportunities, which triggered a short-term population return to underdeveloped areas [32,52]. From 2011 to 2013, the number of cities experiencing population loss in the SSIMR notably increased, which may be related to the rapid recovery of the economy and the enhancement of employment transfer in the eastern coastal areas. From 2014 to 2019, the number of cities experiencing population loss in the SSIMR significantly increased, the main reason being that the regional energy and heavy chemical industry base was in a period of industrial transformation, with sluggish economic development and a sharp increase in unemployment. In addition, the “siphoning effect” of regional central cities such as Xi’an, Taiyuan, and Hohhot continues to strengthen, leading to the intensification of the spatially uneven flow of population factors [47];

(3)

Evolutionary cycle. The economic development potential and population factors flow trend of cities in the SSIMR exhibit obvious periodicity, with strong fluctuations in urban economic resilience and population change in different disturbance cycles. According to the actual development of regional cities and the changes in China’s macro environment, the process of urban growth and shrinkage in the SSIMR can be divided into two cycles [50]: the period of financial crisis disturbance (2008–2013) and the period of industrial transformation pain (2014–2019). During the period of financial crisis disturbance from 2008 to 2013, the economic resilience of cities in the SSIMR was generally strong, with only individual cities experiencing short-term economic recession. However, there were more cities experiencing population loss, and their scale showed an evolutionary trend of decreasing first and then increasing. During the period of industrial transformation pain from 2014 to 2019, the economic resilience of cities in the SSIMR significantly decreased, with a notable increase in cities experiencing economic recession and population loss.

4.1.2. Spatial Patterns of Urban Economic Resilience and Population Change

(1)

Economic resilience. The regional differentiation characteristics of urban economic resilience in the SSIMR are significant. The urban economic resilience in the southern Shaanxi and Guanzhong regions has been strong for a long time, while the urban economic resilience in the Shanxi region is relatively weak (Figure 4). From 2008 to 2013, the economic development potential of cities in the SSIMR was generally strong, with 96.67% of the cities having an economic resilience of greater than 0. Among them, Inner Mongolia and Shaanxi regions have strong economic development potential, with a regional economic resilience of 0.700 and 0.693, respectively. Cities such as Wuhai, Ordos, Hulun buir, Tongchuan, Ankang, and Hanzhong have economic resilience exceeding 0.573, ranking at the forefront of the SSIMR. Although the economic development potential of the Shanxi region is higher than the national average, the regional economic resilience is relatively weak (0.209), and the economic resilience of cities such as Taiyuan and Datong is less than or close to 0. From 2014 to 2019, the economic development potential of cities in the SSIMR significantly decreased, with 63.33% of the cities having an economic resilience of less than 0. Among them, the economic development potential of the Shanxi region is relatively weak, with a regional economic resilience of −0.266. Except for the provincial capital Taiyuan, the economic resilience of other cities is less than 0: Linfen, Shuozhou, and Lvliang have the weakest economic resilience, all less than −0.462, indicating that the economic system in this region is fragile, and the problem of structural recession is prominent. The economic development potential of the Inner Mongolia region has weakened significantly, with the economic resilience of most cities transitioning below 0, resulting in a regional economic resilience decrease to −0.098. This may be related to factors such as “squeezing water, reducing debt, and adjusting structure” in urban economic development. The Shaanxi region has the strongest economic development potential, with a regional economic resilience of 0.159, and the economic resilience of cities such as Ankang, Hanzhong, Shangluo, and Xi’an are all greater than 0, indicating that the economic growth capacity of the southern Shaanxi and Guanzhong regions is still strong;

(2)

Population change. The problem of population loss in the SSIMR is becoming increasingly severe, with a significant increase in the scope and intensity of cities losing population, roughly in line with the “core–periphery” structure centered on the provincial capital cities (Figure 5). From 2008 to 2013, 15 cities in the SSIMR experienced population loss, accounting for 50% of the total number of cities in the region, mainly clustered in Shaanxi (26.67%) and Inner Mongolia (16.67%). Among them, Hulun buir and Bayan Nur have the most severe population loss problems, with population change intensities of −3.512 and −2.566, respectively. The cities with population growth are mainly distributed near energy-rich areas such as the border areas of Shanxi, Shaanxi, and Inner Mongolia as well as the central and southern regions of Shanxi, which may be closely related to the development of energy resources in the SSIMR. From 2014 to 2019, the number of cities experiencing population loss in the SSIMR significantly increased, with 25 cities experiencing a population loss problem, accounting for 83.33% of the total number of cities in the region, and the intensity of population change showed a “core–periphery” structure centered on the provincial capital cities. Among them, the provincial capital cities of Xi’an, Taiyuan, and Hohhot have the most significant population growth, making up the core areas of population agglomeration in the SSIMR. Resource-based cities such as Baotou and Ordos, with their economic foundation and location advantages, have attracted the agglomeration of population factors and showed a clear trend of population growth. The scope of cities with population loss has significantly increased, widely distributed in non-provincial capital cities, which may be affected by the interaction between the economic transformation of resource-based cities and the imbalanced and insufficient of regional economic and social development.

4.1.3. Urban Growth and Shrinkage from the Perspective of Economic Resilience and Population Change

From the perspective of relative shrinkage, the recession of urban economic development potential and the intensification of population loss in the SSIMR are becoming increasingly severe. Urban development is transitioning from global growth to widespread shrinkage, with 56.67% of the cities in a state of relative shrinkage (Figure 6). From 2008 to 2013, the urban development of the SSIMR was in the stage of global growth. In total, 46.67% of the cities showed a significant growth trend with strong economic resilience and population factors agglomeration, 50% of the cities showed a smart growth trend with strong economic resilience but population factors loss, and 3.33% of the cities showed a slow development trend with weak economic resilience and population factors agglomeration. From 2014 to 2019, the urban development of the SSIMR entered a stage of widespread shrinkage, and 56.67% of the cities were facing the shrinkage challenges of economic recession and population loss. The proportion of growth-oriented cities decreased to 43.33%; within this group, 26.67% of the cities showed a smart growth trend, 10.00% of the cities showed a significant growth trend, and 6.67% of the cities showed a slow development trend. The phased evolution of urban growth and shrinkage in the SSIMR indicates that in underdeveloped areas undergoing rapid urbanization and industrialization, urban development is more volatile. Urban shrinkage is mainly manifested as a short-term phenomenon caused by internal and external disturbances, which is fundamentally different from the long-term decline phenomenon in highly urbanized areas in developed countries in Europe and America [32].

From the perspective of spatial distribution, the structure of urban types and their spatial layout in the SSIMR have undergone significant changes, especially in the densely populated areas of resource-based cities. From 2008 to 2013, urban development was dominated by significant growth and smart growth, showing a spatial pattern of “large agglomeration and small dispersion” (Figure 7). The significant growth cities are primarily resource-based cities, mainly concentrated near the border areas of Shanxi, Shaanxi, and Inner Mongolia, as well as the densely populated areas of resource-based cities such as the central and southern Shanxi region, with obvious “clump-shaped” layout characteristics. The smart growth cities are mainly concentrated in southern Shaanxi, Guanzhong, and eastern Inner Mongolia and scattered across various provinces and regions. From 2014 to 2019, urban development was dominated by significant shrinkage and smart growth, showing a spatial pattern of “western growth–eastern shrinkage”. The significant shrinkage cities are mainly distributed in the eastern region where resource-based cities are concentrated, exhibiting a clustered and contiguous distribution feature. The smart growth cities are still clustered in the southern Shaanxi and Guanzhong regions, with clear “cluster” layout characteristics. There are fewer significant growth cities, scattered in provincial capital cities such as Xi’an and Taiyuan, as well as areas with a good economic foundation, such as Baotou. The slowing growth cities are the fewest, mainly being distributed in areas with a good economic foundation and strong population aggregation capacity, such as Hohhot and Ordos.

4.2. Formation Mechanism of Urban Growth and Shrinkage in the SSIMR

The urban development in the SSIMR shows obvious differentiation and periodicity, and the development trend of cities is sharply transformed with the changes of regional environment. To this end, this section constructs a multiple regression model with urban economic resilience and population change intensity as dependent variables and nine indicators such as population agglomeration, capital agglomeration, industrialization level, and service industry development as independent variables to explain the influencing factors of urban growth and shrinkage evolution in different disturbance cycles (

Table 3. Results of stepwise regression models.

Variable	2008–2013		2014–2019
	Economic Resilience	Population Change	Economic Resilience	Population Change
Lpop	0.119	−0.100	0.204	0.600 ***
Inves	0.564 ***	0.488 ***	0.402 *	0.019
Indus	−0.747 **	0.485 *	−0.734 **	0.630 **
Servi	−0.609 **	0.570 **	−0.554 **	0.574 **
Resou	0.307	0.070	0.064	−0.229
FDI	−0.333 *	−0.096	−0.285	0.228
Gover	−0.124	−0.007	0.365 *	0.188
Techo	−0.336 *	0.218	0.435 **	−0.134
Uempl	−0.335 *	0.023	−0.328 *	0.057
N	30	30	30	30
Adj.R2	0.484	0.591	0.374	0.513
F	4.024	5.665	2.929	4.394

Note: *, **, and *** indicate significance at the 0.1, 0.05, and 0.01 levels, respectively.

). The results show that the variance inflation factor (VIF) of each influencing factor is less than 5, indicating that there is no collinearity among the independent variables; from 2008 to 2013 and 2014 to 2019, the determination coefficients R² of economic resilience and population change in the model are 0.484, 0.591, 0.374, and 0.513, respectively, and the F-statistic scores are 4.024, 5.665, 2.929, and 4.394, respectively. All of them reached the significance test, indicating that the model has a good fitting effect, and it can effectively reveal the evolutionary mechanism of urban growth and shrinkage in the SSIMR.

4.2.1. Factor Agglomeration Effect

The problem of imbalanced and insufficient development in the SSIMR is prominent, and the level of urban development is significantly differentiated, leading to the spatially uneven flow of production factors, resulting in the Matthew effect of “the strong balancing the strong and the weak balancing the weak”. As shown in Table 3, from 2014 to 2019, the initial population size of the municipal district had a significant positive impact on population growth, indicating that regional central cities can generate a strong agglomeration effect and attract the spatial agglomeration of population factors. The fixed assets investment per capita plays an important role in promoting urban economic resilience and population growth, indicating that capital agglomeration helps to promote urban development. However, in the period of industrial transformation, the strength of the impact on population growth significantly weakened and became insignificant, which may be because the key investment directions in this period include technological innovation, enterprise transformation, and environmental governance, which are more conducive to enhancing the resilience and sustainability of the urban economic system but are less attractive to population factors.

Central cities such as Xi’an, Taiyuan, and Hohhot have good infrastructure, employment opportunities, and business environments, which exert a strong attraction to factors such as population and capital and can effectively promote the utilization and transformation of production factors, presenting the benign development trend of economic prosperity and population growth [46,47]. In contrast, underdeveloped areas have weak urban economic foundation of cities, with poor regional conditions, resource endowments, and living environments. The ability to gather and transform factors is weak, and it is also affected by the siphoning effect of regional central cities, resulting in the continuous outflow of production factors such as population and capital, raising the recessionary risk of economic development slowdown and population loss [53]. As a result, the “core–periphery” structure of the urban system has been increasingly strengthened.

4.2.2. Industrial Structure Imbalance

The SSIMR is an important energy and heavy chemical industry base in China. Industry plays an important role in urban development, profoundly affecting the flow of urban production factors and the development of the service industry and driving the cyclical evolution of urban growth and shrinkage. The impact of the proportion of industrial-added value in the GDP and the proportion of service industry added value in the GDP on urban economic resilience was found to be significantly negative and weakened, while their impact on urban population growth was significantly positive and enhanced, indicating that the SSIMR has a high proportion of industry, while the development of the service industry is relatively lagging. The problem of industrial structure imbalance is prominent, and a benign interaction mechanism between industries has not yet been formed.

The large-scale exploitation of energy resources in the SSIMR has attracted the agglomeration of production factors such as capital, technology, and labor force. While promoting the rapid expansion of the industrial economy, this has squeezed the development space of the service industry. Along with the continuous development of the energy economy, the potential of the industrial economy to accumulate resources has reached its peak, and the factor agglomeration ability of the service industry has gradually increased. However, due to the path-locking effect of the energy economy, the path-creating ability of the service industry is weak, and its share growth is mainly caused by the slowdown of industrial growth rate, leading to the increasing fragility of the regional economic system and greater pressure on industrial transformation [11,19]. In the context of the new normal of China’s economic development, urban development in the SSIMR is affected by factors such as path dependence, structural contradictions, and institutional constraints, and problems such as industrial recession, increasing unemployment, and environmental degradation have emerged [54]. Therefore, governments at all levels have attempted to cultivate new drivers for urban development through factor reorganization, industrial restructuring, and other means, promoting the development of advanced manufacturing and modern service industries and alleviating the problem of industrial structure imbalance. However, as it is still in the throes of industrial structure transformation, the diversified industrial system has not yet been built [30,55], and the promoting effect of the service industry development on urban economic resilience has not yet been formed.

4.2.3. Policy System Reform

The spatiotemporal differentiation of urban development in the SSIMR is a product of the policy system changes, influenced by the interaction between national policies and local strategies. At the macro level, driven by China’s energy and heavy industrialization strategy, the energy and heavy industry in the SSIMR developed rapidly from 2008 to 2013, with energy production increasing from 1040.28 million tons of standard coal to 1753.26 million tons of standard coal, at an average annual growth rate of 11.00%. The industrial-added value increased from 930.49 billion yuan to 1709.44 billion yuan, with an average annual growth rate of 12.93%, injecting a strong impetus to regional economic and social development. With the implementation of policies such as supply-side structural reform and high-quality development strategy, the extensive development model of “relying on energy and heavy industry” in the SSIMR is facing enormous challenges, and the energy and heavy chemical industry has been severely impacted. The average annual growth rate of energy production decreased to 1.77% from 2014 to 2019, and the average annual growth rate of industrial-added value was only 4.46%, which was significantly lower than the national economic growth rate during the same period (8.92%), resulting in a series of problems such as the weakening of regional economic development potential, the increase of enterprise closures, and the intensification of population loss [11,54].

At the local level, policy system factors such as opening-up, government regulation, innovation guarantee, and institutional locking have a significant impact on urban development. During the period of financial crisis disturbance, the impact of the level of opening-up on urban economic resilience was significantly negative, indicating that cities with high levels of opening-up suffered more severe economic shocks. During the period of industrial transformation pain, the impact of government regulation on urban economic resilience was significantly positive, reflecting the promoting effect of government intervention on industrial transformation. The impact of innovation guarantee on urban economic resilience was significant and enhanced, indicating that strengthening government financial support for technological innovation is an important path to build a resilient urban economic system in the new era. The impact of institutional locking on urban economic resilience has been negative for a long time and is statistically significant, indicating that institutional locking has constrained the urban economic development of the SSIMR. It is thus urgent to enhance the market vitality of regional economic development.

5. Discussion

5.1. Innovation and Applicability

Population loss is the core indicator for identifying urban shrinkage in academia, but many studies have questioned the interdependent relationship between population change and economic development, and the development paradox of population loss and economic growth has also been confirmed [32,35,36]. For this reason, the complex relationship between population change and economic development has become an important aspect of identifying urban shrinkage. However, existing studies tend to ignore the essence of the phased evolution of population and economic factors, leading to significant controversy in the division of the evolutionary stages of urban systems [28,38]. In addition, existing studies focus on the issue of absolute urban shrinkage, with a greater emphasis on the short-term irreversibility of urban shrinkage [3,37], and less attention is paid to the issue of relative urban shrinkage that can be reversed in the short term. Therefore, starting from the perspective of relative shrinkage, this paper introduces the theory of economic resilience into the field of urban shrinkage and constructs a theoretical method for identifying urban shrinkage by integrating economic resilience and population change, making the division of the evolutionary stages of urban population and economic factors more scientific. Compared with similar studies [35,38], this paper promotes the integration of urban systems theory and resilience development theory and demonstrates the application value of economic resilience theory in the field of urban shrinkage. It provides a new perspective for urban shrinkage research and further supplements the theory of urban shrinkage. The research results can provide a scientific basis for the construction of resilient national land space and the transformation and sustainable development of shrinking cities.

5.2. Complexity and Particularity

Urban shrinkage is a complex and systematic process, embedded in globalization and rooted in localization, which determines the regional heterogeneity of its connotations, types, and formation mechanisms. Therefore, attention should be paid to “localization” thinking in studying urban shrinkage [3]. In China, where the population continues to grow, and the economy steadily develops, urban shrinkage is mainly manifested as the loss of urban population and the slowdown of economic growth in specific periods. Essentially, it is caused by the relative weakening or absolute degradation of urban development potential, resulting in the redistribution of urban production factors and the reconstruction of spatial structure [3,28]. This is significantly different from the essence of urban “decline” in European and American countries. Therefore, the study of urban shrinkage in the Chinese context should emphasize relativity and reversibility. The evolutionary relationship between urban economic resilience and population change in the SSIMR is complex, showing phenomena of population loss with economic growth and economic recession with population increase. This is not only a clear manifestation of the complexity and particularity of urban shrinkage in China but also an important sample of global urban shrinkage. Unlike the “chronic burning” process of urban shrinkage in European and American countries [1,56], urban shrinkage in the SSIMR exhibits more volatility, which is a short-term reversible phenomenon caused by factors such as the factor agglomeration effect, industrial structure imbalance, and policy system reform.

5.3. Policy Recommendations

The urban development in the SSIMR is evolving from global growth to widespread shrinkage, with 56.67% of the cities in a state of relative shrinkage. Among them, resource-based cities account for 82.35%, indicating that the development of resource-based cities is more volatile and fragile. This conclusion is similar to earlier findings that cities with a single industrial structure are prone to drastic shifts from prosperity to depression [11,57]. Therefore, the key to the transformation and development of cities in the SSIMR lies in enhancing the resilience of resource-based cities. We should attach importance to the synergistic development of energy and non-energy industries, undertake orderly promotion of the transformation of traditional industries and the cultivation of emerging industries, and build a green and diversified modern industrial system to avoid the risks of economic development instability and population loss caused by a single industrial structure. In addition, the significant differentiation of urban development in the SSIMR has intensified the spatially uneven flow of production factors. We should be alert to the siphoning effect of the regional central cities, scientifically plan the industrial layout, promote the orderly flow of production factors, and build a benign development mechanism of win-win cooperation and mutual benefit among regions.

5.4. Limitations and Future Prospects

Urban shrinkage is a complex system process that integrates multiple factors, multiple dimensions, and multiple scenarios [3,4]. Based on the perspective of relative shrinkage, this paper explores the spatiotemporal heterogeneity of urban growth and shrinkage in the SSIMR from the dimensions of economic resilience and population change, which helps to elucidate the evolutionary laws of urban systems under the crisis disturbance scenario and provides a scientific reference for avoiding the risk of urban shrinkage. However, due to the limitation of research data, urban landscape factors such as idle land and housing vacancies were been included in the measurement range, and the issue of absolute urban “decline” needs to be further analyzed. In addition, the dynamic simulation and sustainable regulation of urban shrinkage under multiple-scenarios integration were not addressed. In the future, with the help of big data mining techniques and methods, we should systematically depict the evolutionary process of regional urban shrinkage through data such as nighttime lighting brightness and urban vitality indices, simulate and predict the development trend of urban shrinkage, and further deepen the study of the regulatory mechanism and response strategies of urban shrinkage.

6. Conclusions

To evaluate the evolutionary process and influencing mechanism of urban growth and shrinkage in the SSIMR, this study introduces the theory of economic resilience into the field of urban shrinkage and proposes a theoretical method for identifying urban shrinkage from the dimensions of economic resilience and population change. The main findings are as follows.

First, the economic resilience and population change of cities in the SSIMR exhibit strong volatility, and the study of urban shrinkage should be conducted under specific crisis disturbance scenarios to ensure the integrity of the urban evolutionary cycle. Based on the realities of regional development and macro environmental changes in the SSIMR, its urban development cycle can be divided into two phases: the period of financial crisis disturbance and the period of industrial transformation pain.

Second, the spatiotemporal heterogeneity of urban economic resilience and population change in the SSIMR is significant. From the economic perspective, the economic resilience of cities in the SSIMR has significantly declined, and the problem of economic recession is gradually intensifying in the context of the “new normal of the economy”. The economic resilience of the southern Shaanxi and Guanzhong regions has been strong for a long time, while the economic resilience of the Shanxi region is relatively weak. From the population perspective, the population change trend of cities in the SSIMR is relatively stable, and the problem of population loss is relatively severe, whose scope and intensity are increasing daily, showing a “core–periphery” structure centered on provincial capital cities.

Third, the SSIMR mainly includes three types of cities: significant growth, significant shrinkage, and smart growth. Urban development is evolving from global growth to widespread shrinkage, with 56.67% of the cities experiencing relative shrinkage, showing a spatial pattern of “western growth–eastern shrinkage”.

Finally, the evolution of urban systems in the SSIMR is the result of the interaction between internal and external environmental changes. Agglomeration effect, industrial structure, policy system, and other factors constitute the environmental basis of regional urban development, affecting the spatiotemporal heterogeneity of urban growth and shrinkage evolution.