The Impacts of Urban Population Growth and Shrinkage on the Urban Land Use Efficiency: A Case Study of the Northeastern Region of China

Abstract

In the context of rapid urbanization, urban population differentiation has become increasingly pronounced. Regional development strategies based on growth scenarios often lead to continuous expansion, regardless of urban population status. Such “one-size-fits-all” models exacerbate resource waste and negatively impact urban land use efficiency (ULUE). This study aims to explore the mechanisms by which urban population growth and shrinkage (UPGS) affect ULUE, with the goal of enhancing ULUE and promoting sustainable urban development. We analyzed 34 prefecture-level cities in China’s three northeastern provinces. First, we identified UPGS using population data. We then employed a three-stage SBM-DEA model to measure ULUE from 2000 to 2020. Spatial analysis methods were used to examine the spatiotemporal characteristics and correlations between UPGS and ULUE. Additionally, mediating effect models and spatial Durbin models were utilized to empirically test the impact processes, mechanisms, and spatial heterogeneity. Our findings reveal that: (1) Over the past 20 years, urban population shrinkage in northeastern China has intensified, and significant regional disparities in urban development are evident. (2) Population growth positively influences ULUE, while population shrinkage inhibits its improvement. (3) Economic development, technological innovation, and industrial structure upgrading are key factors in enhancing ULUE in this region, while the impact of public services on ULUE varies significantly at different stages of urban development. (4) Economic development, technological innovation, and industrial structure upgrading exhibit spatial spillover effects, whereas public services are constrained by regional limitations, resulting in minimal spatial spillover effects. To foster coordinated regional development, this study proposes policy recommendations, including strengthening support for resource-dependent cities, optimizing the allocation of public resources, and promoting technological innovation and industrial diversification.

1. Introduction

In a complex global economic landscape characterized by slowing population growth and an aging population, migration trends increasingly favor large cities and metropolitan areas [1]. Moreover, non-core cities and regions are exhibiting population decline [2]. The economy of China occurs at a critical transformation stage, with both resource and environmental constraints increasing. The Matthew effect of factor agglomeration has amplified the advantages of economically developed regions, exacerbating regional development imbalances. Consequently, the spatial differentiation of population distributions is increasing, with urban populations demonstrating both growth and shrinkage trends [3]. At the middle and late stages of urbanization, a significant influx of people and economic factors occurs in megacities and metropolitan areas [4]. However, most regional development strategies and plans in China, under the pressure of rapid urbanization, are based on growth scenarios. This has led to continuous development and construction in cities, regardless of population trends. As a result, areas exhibiting population decline face economic and social issues, such as housing vacancies, inefficient land use, and spatial disorder [5,6]. These problems exacerbate land resource waste and negatively impact the urban land use efficiency [7]. These issues are particularly severe in the northeastern region of China, where the population loss proportion is significantly greater than the national average, although there are cities that exhibit rapid population growth.

The urban land use efficiency is not only an important indicator of the urban development and transformation potential but also a key measure of rational resource utilization and sustainable production and living methods [8,9,10,11,12]. As population dynamics shift, fluctuations in land and labor inputs impact urban planning, land use structure, and market supply and demand, ultimately affecting land use efficiency. Therefore, urban population growth and shrinkage are closely linked to urban land use efficiency.

How does urban population growth and shrinkage affect land use efficiency? What forms do these impacts take? What reference value do these impacts have for policy making? These questions have become focal points and challenges in current research on urban population dynamics. Therefore, it is particularly important to explore population growth and shrinkage and their effects on urban land use efficiency in the northeastern region. Although recent research has touched on this issue [7,13,14], there is still a lack of in-depth and systematic analysis of the specific relationship and underlying mechanisms between the two. This study systematically analyzes the population growth and shrinkage in Northeast China’s cities and their impact on urban land use efficiency, revealing existing problems and further exploring the mechanisms at play between the two. This research not only offers new insights for improving urban land use efficiency and fills gaps in existing studies, but also identifies the fundamental issues in Northeast China. It provides a scientific basis for the government of Northeast China to optimize population mobility and land use policies and serves as a useful reference for other regions facing similar challenges.

The structure of this paper is as follows: First, the core concepts of urban population growth, shrinkage, and land use efficiency are explained from a theoretical perspective, along with a discussion of the mechanisms of interaction between them. Next, based on empirical data, a comprehensive assessment is conducted of the current state of population changes and land use efficiency in Northeast China. Then, combining theoretical analysis with data processing results, the study explores the specific impact mechanisms for Northeast China. Finally, the research conclusions are summarized, and corresponding policy recommendations are presented.

2. Literature Review and Theoretical Framework

2.1. Understanding Urban Population Growth and Shrinkage

Population shrinkage, also referred to as depopulation or population decline, is a social phenomenon that has gained widespread attention within the context of population decrease and aging [15]. The issue of urban population shrinkage is a key aspect. Research on urban population shrinkage originated in Germany in the 1980s, when $H\ddot{a}u\beta ermann$ first introduced the concept of “shrinking cities” to describe the phenomenon of urban population decline or decrease. It is important to note that while urban population shrinkage is a significant component of shrinking cities, the two concepts are not entirely synonymous [15]. Urban population shrinkage directly reflects changes in the number of residents within cities, whereas shrinking cities encompass broader dimensions, such as economic, social, and spatial transformations. This review focuses on research related to urban population shrinkage, examining the characteristics, causes, and regional development impacts of this phenomenon. Since the 1990s, more than a quarter of cities worldwide with populations exceeding 100,000 residents have exhibited population shrinkage, rendering this issue a global phenomenon. China faces a similar dilemma, with the northeastern region exhibiting significant long-term population loss [16]. However, many scholars consider urban shrinkage a neutral term [17]. Urban population growth and shrinkage (UPGS) are not isolated phenomena but rather interconnected and opposing processes within regional urbanization development, dynamically transforming with the evolution of urbanization stages [18]. This process represents spatial restructuring for optimal resource allocation at the national level, aiming to maximize the intensive use and effects of land resources and human capital. Both domestic and international scholars have extensively researched the population shrinkage phenomenon, covering aspects such as its connotations [15,19,20], spatial differences [21], and regional impact coping strategies [22]. In regard to determining connotations and methods, the urban population in China can be categorized into permanent and registered residents, in contrast to the West. The activities of permanent residents more effectively reflect urban activities. Therefore, when studying urban population shrinkage in China, the use of the permanent resident population as the research subject more accurately reflects the actual population dynamics and scale changes in cities [23].

2.2. Understanding the Urban Land Use Efficiency

Research on the urban land use efficiency (ULUE) concept has evolved through several stages: from a dual interpretation of the structural efficiency and the marginal efficiency [24] to quantitative assessments of input-output ratios [25] and, finally, to comprehensive definitions that consider material inputs, ideal outputs, undesirable outputs, and various activities on land [26,27]. Despite continuous refinement and development of the definition, its core encompasses two basic principles: first, minimizing inputs for a fixed output or maximizing outputs for a fixed input; second, optimizing the allocation ratio of production factors to achieve the optimal combination of resources [28].

Regarding ULUE measurement methods, there has been ongoing optimization and innovation. Initially, research relied on qualitative descriptions and formal analyses [24], but it has gradually advanced to the use of various quantitative analysis methods. These methods include principal component analysis and weighting methods [29], coordination degree models [30], regression analysis methods [31], data envelopment analysis (DEA) models [32], slack-based measure (SBM) models [33], and stochastic frontier analysis (SFA) models [34]. Moreover, most domestic scholars have employed one of the aforementioned methods to measure efficiency, which often exhibits certain limitations in application. For example, these methods may not fully account for the impacts of the external environment and random error factors on efficiency values, leading to biased evaluation results. The three-stage DEA model [35], however, effectively eliminates the interference of environmental factors and random errors. Comparative tests between one- and three-stage DEA models have revealed that ignoring these interference factors can result in inaccurate and overestimated efficiency measurements. Consequently, the three-stage DEA model has been widely applied in various fields [36,37,38,39,40]. Therefore, to avoid the influence of external environmental variables and random disturbances and to explore true productivity levels in depth, the three-stage SBM-DEA method was adopted to evaluate the ULUE in this study.

The connotations of UPGS and ULUE are shown in Figure 1.

2.3. Understanding the Relationship between UPGS and ULUE

2.3.1. Direct Impacts of UPGS on the ULUE

The mechanisms of the direct impacts of UPGS on ULUE are primarily manifested in the allocation and utilization of production factors. Changes in the urban population affect land inputs [41], thereby modulating urban planning and land use structures and leading to fluctuations in the land input efficiency. Additionally, labor inputs are influenced by changes in the urban population: population growth stimulates employment and attracts talent, whereas population shrinkage can lead to talent loss, reduced job opportunities, and increased unemployment rates [42]. Moreover, population changes affect market size and supply-demand dynamics, thereby influencing capital inputs [43].

However, these processes, whether positive or negative, do not directly determine the ULUE. The relationship between inputs and outputs may follow an inverted U-shaped pattern, indicating that at certain urban development stages, reducing land inputs may help increase rather than decrease outputs, while increasing inputs may not necessarily improve outputs [44]. For example, in densely populated areas, reducing inputs might alleviate the pressure from land economic activities and pollution, reduce redundant costs, and thereby enhance efficiency levels. Conversely, increasing inputs may result in greater population agglomeration, leading to resource scarcity and environmental stress, ultimately reducing efficiency.

2.3.2. Mediating Effects of UPGS on the ULUE

The mechanisms of the mediating effects of UPGS on ULUE are manifested in economic development, technological innovation, public services, and industrial structure upgrading, forming a dynamic network.

The impact of urban population changes on economic development is multifaceted and depends on regional conditions. The existing research encompasses two main perspectives. One perspective argues that UPGS impose complex, nonlinear impacts on economic development, with both positive and negative effects [45,46,47]. The other perspective argues that there is no negative correlation between population growth and economic development [45], but population shrinkage can adversely affect economic development [48]. Moreover, the influence of economic development on ULUE exhibits duality. On the one hand, stable economic growth can enhance ULUE by increasing land use density, causing land use structure optimization, and enhancing land resource management levels [49]. On the other hand, unbalanced economic development, excessive resource exploitation, and environmental degradation can lead to ULUE decline. Therefore, under the mediating effect of economic development, the impact of urban growth or shrinkage on the ULUE is twofold.

Urban population changes significantly drive technological innovation. From a supply-side perspective, a large population suggests a greater reserve of talent for technological innovation. From a demand-side perspective, population increase promotes market demand diversification, making it easier to generate product and business model innovations that meet various needs. Conversely, while population shrinkage may limit the pace of technological innovation [41], continuous social progress and technological development, along with the spillover effects of knowledge, may still effectively promote the substitution of capital and technology. Technological innovation can significantly impact ULUE in positive ways. For example, innovation can enhance government methods for managing urban land, enabling flexible responses to various risks. Furthermore, technological innovation is linked to other areas, such as attracting foreign investment to promote economic diversification or enhancing urban intelligence to improve the living quality. Notably, scholars have also observed that technological innovation reduces pollution emissions and increases economic benefits, thereby improving the ecological efficiency. This indirectly indicates that land use efficiency will also increase.

Urban population changes profoundly impact the unique products provided by the government, namely, public services [50]. Population growth often promotes an increase in public service levels. As the urban population size increases, social service capacity and urban accessibility levels gradually increase, which not only helps reduce the circulation time of resource factors and transportation costs but also promotes industrial deepening [51]. In contrast, population shrinkage is usually detrimental to the comprehensiveness of public services [52,53]. With decreasing population, the government faces resource waste and surplus in public services, along with a decline in the supply scope and quality. This results in public service facilities and infrastructure that were originally designed for a larger population becoming idle or wasted. Although increasing the public service level can enhance ULUE, this positive effect is not absolute. When the scale of public services exceeds the optimal level, an overcapacity effect may occur. In such cases, fiscal expenditure on public service construction can become a burden, reducing the efficiency of the production and living activities of residents and thus negatively impacting ULUE. This overcapacity effect may result in the over-construction of public service facilities, increased maintenance costs, and heightened fiscal pressure [54].

The impact of urban population changes on industrial structure upgrading is multifaceted and complex. As urban populations increase, the population urbanization process is accelerated, promoting capital accumulation, knowledge spillover, and market competition, thus facilitating industrial structure upgrading [55]. Carlino et al. [56] indicated that urban development, through enhanced industrial agglomeration and innovation synergy, positively influences industrial structure upgrading. However, this upgrading process is not linear but is influenced by various factors, such as government policies and market demands. Conversely, the impact of urban population shrinkage is dual-faceted, potentially leading to industrial degradation or serving as a catalyst for industrial upgrading. First, the outflow of human capital reduces technological strength and market competitiveness, possibly resulting in slow or even regressive industrial structure upgrading. Second, population shrinkage accompanied by aging and low birth rates might cause a change in the labor structure, with the service industry becoming dominant. If the government implements effective interventions, this shift could present an opportunity for industrial structure upgrading. While industrial structure upgrading has not yet universally driven ULUE improvements [57], theoretically, it could improve ULUE through resource optimization, increased outputs, enhanced intensification, and increased sustainability. Therefore, it is necessary to conduct empirical analysis studies that account for specific regional circumstances.

2.3.3. Spatial Effects of UPGS on ULUE

The first law of geography reveals the interaction patterns of geographical elements, indicating that the flow and aggregation of elements in various regions are not isolated but influenced by other regions, with this influence increasing with decreasing distance. On the basis of this theory, UPGS can impact ULUE through key factors such as economic development, technological innovation, public services, and industrial structure upgrading in the form of spatial effects. This phenomenon not only affects the internal region but may also cross regional boundaries, producing ripple effects in other areas. However, from an international perspective, the spatial impact of population changes on land use efficiency varies significantly.

On one hand, the pathways through which population changes exert influence may serve as a demonstration for others. For example, central cities, through demonstration, radiation, and sharing effects, can promote population concentration and economic development in surrounding cities, creating a spatial spillover effect on population growth in nearby cities. This, in turn, encourages these cities to adopt similar strategies to drive population growth. On the other hand, population change also exhibits a “siphon effect”, where populations and productive factors migrate from less economically developed areas to more advanced regions, further widening the regional disparities.

Economic agglomeration and diffusion modulate land use patterns spatially, whereas the knowledge spillover effect of technological innovation promotes overall ULUE improvements. Industrial structure upgrading not only causes land resource allocation optimization but also triggers urban spatial restructuring, further affecting ULUE [58]. Additionally, the spatially balanced distribution of public service resources plays a crucial role in enhancing ULUE. In summary, these four factors are interwoven through their respective spatial effects, collectively influencing ULUE and generating its unique spatial characteristics in Northeast China.

The theoretical framework is shown in Figure 2.

3. Materials and Methods

3.1. Study Area

Northeast China is located between 118° E and 135° E longitude and 48° N and 55° N latitude, covering an area of 788,000 square kilometers. The region includes the provinces of Heilongjiang, Jilin, and Liaoning. As one of China’s four major economic zones, the Northeast has historically played a vital role in the country’s development. Administratively, it consists of 34 prefecture-level cities, one autonomous prefecture, and one region. Liaoning Province contains 14 prefecture-level cities, including the two sub-provincial cities of Shenyang and Dalian. Jilin Province comprises 8 prefecture-level cities and one autonomous prefecture, with Changchun as a sub-provincial city. Heilongjiang Province includes 12 prefecture-level cities and one region, with Harbin serving as a sub-provincial city (Figure 3).

Historically, industrialization in Northeast China began in the early 20th century, with infrastructure development focused primarily on heavy industries such as steel, coal, and machinery. By the late 1950s, the region’s GDP accounted for 19% of the national total, marking the peak of its economic power. However, as China’s economy transformed in subsequent decades and global economic patterns shifted, the region fell into industrial decline and economic stagnation. Since then, Northeast China has faced significant challenges. Despite recent efforts to upgrade industries and promote innovation, its share of the national GDP had fallen to just 4.9% by the end of 2021.

Geographically and culturally, Northeast China is one of the snowiest regions in the country and shares borders with Russia and North Korea, giving it significant geopolitical and economic importance. In recent years, this unique snow-covered landscape and border advantages have been increasingly leveraged to develop tourism resources. Particularly, cross-border tourism and cultural exchange activities with Russia and North Korea have injected new vitality into the local economy. These border cities, by promoting the tourism industry, have diversified their economies and effectively mitigated population outflows in certain areas [59]. Compared to other regions, these border areas, where tourism and cultural industries have gradually flourished, exhibit distinct land use patterns. The construction of tourism service facilities and the development of scenic areas have become central to land use in these regions.

In terms of population, Northeast China exhibits a relatively large total population. However, in recent years, due to the loss of geographical advantages and resource depletion, the region has exhibited significant population loss, referred to as the northeast phenomenon, the new northeast phenomenon and the post-northeast phenomenon. From 2000 to 2020, approximately 12.81 million permanent residents left the region. In regard to land use and efficiency issues, the northeastern region features various land use types, including industrial land, agricultural land, and residential land. However, with the advancement of urbanization and industrial restructuring, problems such as land idleness and inefficient use have increased.

Population shrinkage refers to the decline in the number of inhabitants within a region over a specific period of time [15]. This study examines the population growth and shrinkage across the 34 prefecture-level city districts of the three northeastern provinces during the 2000–2020 period, divided into five-year intervals (excluding the Da Hinggan Ling region and Yanbian Korean Autonomous Prefecture due to data limitations).

3.2. Methods

The research methodology of this study was designed from three main perspectives. First, quantitative analysis methods were employed to explore the spatiotemporal characteristics of population growth, population shrinkage, and ULUE. Specifically, population growth and shrinkage were calculated via the widely recognized measure of the permanent resident population. To comprehensively evaluate ULUE, an integrated evaluation index system was constructed. Second, considering the specific pathways through which UPGS impact ULUE, a mediating effect model was used to analyze the influence mechanism. In this model, economic development, technological innovation, public services, and industrial structure upgrading are adopted as mediating variables. Finally, considering the potential spatial effects of the mediating variables, spatial econometric models were employed to quantitatively analyze the relationship between population changes and ULUE. A flowchart of the research design and methodology is shown in Figure 4.

3.2.1. Measurement of Population Growth and Shrinkage

Population shrinkage refers to the phenomenon in which the population of a region decreases over a certain period [60]. In this study, which draws on existing research on population change rates, the five-year interval from 2000 to 2020 was used to measure population growth and shrinkage in cities. Notably, permanent resident population data for 2000, 2005, 2010, 2015, and 2020 served as statistical indicators for evaluating city population growth and shrinkage. The population change rate was calculated via data from each pair of consecutive years. The population change rate can be calculated as follows (Equation (1)):

$$S_{ipop}=(P_{iy}-P_{ix})/P_{ix}$$

(1)

where $S_{ipop}$ denotes the degree of population growth or shrinkage, with values > 0 indicating population growth and values < 0 indicating population shrinkage, and $P_{iy}$ and $P_{ix}$ are the permanent resident populations of city $i$ in the starting year ($x$) and the ending year ($y$), respectively. To analyze the status of population growth and shrinkage of cities in different regions, cities were categorized on the basis of their population growth–shrinkage status. Given the relativity of growth and shrinkage, the degree of shrinkage was used to measure all categories except for sustained growth over the entire period. Specifically, if the population shrinks during any phase of the study period, this phenomenon is classified as near-term shrinkage; if the population shrinks during two consecutive phases, it is categorized as short-term shrinkage; if the population shrinks during three consecutive phases, it is regarded as medium-term shrinkage; and if the population shrinks over the entire study period, it is defined as long-term shrinkage. Conversely, the absence of shrinkage during the study period is considered long-term growth.

3.2.2. Three-Stage SBM-DEA Model for Measuring ULUE

On the basis of the analysis in the Literature Review section, the three-stage SBM-DEA model was employed to measure ULUE. At the first stage, the traditional DEA model was used to preliminarily calculate the efficiency of each decision-making unit (DMU) and determine the slack values of the inputs, outputs, and undesirable outputs. At the second stage, a panel SFA model was applied to the slack variables to differentiate the effects of external environmental factors, random disturbances, and managerial inefficiency on the slack values. As such, all DMUs were adjusted to the same external environment, eliminating heterogeneity effects. Finally, at the third stage, the input indicators were adjusted according to the regression results for the environmental variables from the second stage, and the super-efficiency SBM-DEA model was then employed to calculate the efficiency. The advantage of the three-stage SBM-DEA model lies in its rigor and accuracy, as it provides the stepwise removal of external interference factors, thus more precisely reflecting the efficiency level of each DMU. Considering that this model is already well developed, the calculation process and equations are not elaborated herein and can be found in existing studies [61,62,63].

In terms of indicator selection, literature searches were conducted in the authoritative domestic databases of CNKI and Web of Science to screen indicators. The search keywords included “urban land use efficiency” and “land use efficiency”. The selection of evaluation indicators has gradually evolved from single indicators reflecting the economic benefits of urban land use to multiple indicators covering economic, social, and environmental aspects. Moreover, as research has progressed, most studies have begun to account for undesirable outputs. Therefore, on the basis of previous research results, the indicator system detailed in

Table 1. Input and output indicators of urban land efficiency.

Type	Indicators	Indicator Meaning
Input indicators	R&D cost input	Technical investment
	Number of employees	Labor input
	Built-up area	Land input
	Fixed asset investment	Capital investment
	Electricity consumption of the whole society	Energy input
Desirable output indicators	GDP per capital	Material output
	Total retail sales of consumer goods	Material output
Undesirable output indicators	CO2 emissions	Undesirable output

was selected. Inputs were considered from four perspectives, namely, technology, labor, land, and capital, with the selection of R&D expenditure, number of employees, built-up area, and fixed asset investment, respectively, as the four corresponding indicators. Additionally, considering energy as a normal production guarantee, the total electricity consumption of society was introduced as an energy indicator. The expected outputs were represented by the per capita GDP and the total retail sales of consumer goods, whereas the undesirable outputs were represented by CO₂ emissions.

3.2.3. Mediating Effect Model

Considering the impact of independent variable X on dependent variable Y, if X affects Y through variable M, then M is referred to as a mediating variable, and the effect through M is considered the mediating effect [64]. In this study, a mediating effect model was employed to explore the impacts of UPGS on ULUE. On the basis of a literature review and theoretical mechanism analysis, a model of the mediating effects of UPGS on ULUE was constructed, referencing the methods used by Liu et al. [65] and Wen et al. [66].

The following mediating factors were chosen:

(1) Economic development (ED): In this study, the per capita GDP was used as an indicator of economic development. (2) Technological innovation (TI): The research and development (R&D) expenditure effectively reflects the level of investment in technological innovation. Additionally, the number of patent applications and the number of grants are key indicators of the technological innovation level of a city, reflecting the quantity and quality of innovation outputs, respectively, over a given period [67]. Therefore, the R&D expenditure and the number of granted patents were chosen as indicators of technological innovation, with weights determined via the entropy weight method. (3) Industrial structure upgrading (ISU): Following the indicators used in studies on industrial structure upgrading [68,69,70], the proportion of the output of the tertiary industry in the GDP was employed for evaluation. (4) Public services (PSs): Public services encompass multiple domains, including education, healthcare, social security, and culture [71]. To measure public services, four aspects were considered on the basis of data availability: the education expenditure in the education dimension, the number of hospitals and health centers in the healthcare dimension, the total collection of public libraries in the culture dimension, and the number of urban employees enrolled in basic pension insurance in the social security dimension. The entropy weight method was adopted to comprehensively evaluate these indicators for measuring the level of public services.

The following control variables were selected:

(1) Ecological environment quality (EEQ): The green coverage rate in built-up areas (%). (2) Environmental governance level (EGL): The harmless treatment rate of domestic waste (%). (3) Resource utilization (RU): The total electricity consumption of society. (4) Foreign investment level (FD): The proportion of the total foreign investment to the GDP, which indicates the foreign capital utilization level [72].

On the basis of the selected variables, several theoretical models were constructed, as expressed in Equations (2)–(7).

$$ULU{E}_{it}={\beta }_{0}UGS{D}_{it}+{\delta }_0{X}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(2)

$$E{D}_{it}={\alpha }_0+{\beta }_{1}UGS{D}_{it}+{\delta }_1{X}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(3)

$$T{I}_{it}={\alpha }_1+{\beta }_{2}UGS{D}_{it}+{\delta }_2{X}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(4)

$$IS{U}_{it}={\alpha }_2+{\beta }_{3}UGS{D}_{it}+{\delta }_3{X}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(5)

$$P{S}_{it}={\alpha }_3+{\beta }_{4}UGS{D}_{it}+{\delta }_4{X}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(6)

$$ULU{E}_{it}={\eta }_0+{\eta }_{1}UGS{D}_{it}+{\eta }_{2}E{D}_{\mathrm{it}}+{\eta }_{3}T{I}_{\mathrm{it}}+{\eta }_{4}IS{U}_{\mathrm{it}}+{\eta }_{5}P{S}_{\mathrm{it}}+{\eta }_6{X}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(7)

where $ULU{E}_{it}$ is the dependent variable, indicating the ULUE of city $i$ in year $t$; $UGS{D}_{it}$ is the key explanatory variable, indicating the degree of urban growth or shrinkage in city $i$ in year $t$; ${\mu }_i$ denotes the individual fixed effects; ${\lambda }_t$ denotes the time-fixed effects; ${\epsilon }_{it}$ is the error term; $X$ is the set of control variables; and $\alpha ,\beta ,\delta ,\eta$ denote the parameters to be estimated.

3.2.4. Spatial Econometric Model

In the field of geography, spatial interactions and spatial spillover effects may occur between different regions. Considering that spatial correlation is neglected in the original ordinary least squares (OLSs) model, which can lead to biased estimation results, we constructed a spatial lag model (SLM), a spatial error model (SEM), and a spatial Durbin model (SDM) to analyze the relationships between urban growth–shrinkage and ULUE [73].

In the SLM, compared with the original model, there is an additional spatial weight matrix, which includes a spatial lag term of the dependent variable among the explanatory variables to account for the influence of the dependent variables from neighboring regions on the study area. In the SEM, spatial effects are incorporated into the error term (the spatial weight matrix is incorporated into the unobservable error term) to account for the impact of other omitted variables or unobservable factors from neighboring regions on the study area. The SDM includes spatial lag terms of both the dependent and explanatory variables relative to the original model, which aim to measure the influence of the dependent and explanatory variables from neighboring regions on the study area. Compared with the SLM and SEM, the SDM considers a broader range of factors. The SLM, SEM, and SDM are expressed as Equations (8), (9) and (10), respectively, and the ULUE can be obtained by Equation (11).

$$ULU{E}_{it}=C+\rho WULU{E}_{it}+{\beta }_{0}UGS{D}_{ij}+{\beta }_i{X}_{\mathrm{it}}+{\mu }_i+{\upsilon }_t+{\epsilon }_{it}$$

(8)

$$ULU{E}_{it}=C+{\beta }_{0}UGS{D}_{ij}+{\beta }_i{X}_{\mathrm{it}}+{\mu }_i+{\upsilon }_t+{\epsilon }_{it}$$

(9)

$${\epsilon }_{it}=\gamma W{\epsilon }_{it}+{\tau }_{\mathrm{it}}$$

(10)

$$ULU{E}_{it}=C+\rho WULU{E}_{it}+{\beta }_{0}UGS{D}_{ij}+{\delta }_{0}WUGS{D}_{ij}+{\beta }_i{X}_{\mathrm{it}}+{\delta }_{i}W{X}_{\mathrm{it}}+{\mu }_i+{\upsilon }_t+{\epsilon }_{it}$$

(11)

where $ULU{E}_{it}$ is the ULUE, $UGS{D}_{it}$ is the degree of urban growth and shrinkage, and $X_{it}$ denotes the control variables. Moreover, $C$ is a constant term, $p$ is the spatial autoregressive coefficient, and $W$ is the spatial weight matrix. In addition, $WULU{E}_{it}$, $WUGS{D}_{it}$ and $W{X}_{it}$ are the spatial lag terms of the ULUE, urban growth and shrinkage, and control variables, respectively; ${\beta }_i$ and ${\delta }_i$ are the coefficients to be estimated; ${\mu }_i$ and ${\upsilon }_i$ are the individual fixed effects and time-fixed effects, respectively; ${\epsilon }_{it}$ is the spatial error term; $\gamma$ is the spatial error coefficient; and ${\tau }_{\mathrm{it}}$ is the error term.

3.3. Data Sources and Processing

The data used in this study cover the following aspects: (1) The resident population data used to identify population growth and shrinkage were sourced from the Heilongjiang Statistical Yearbook, Jilin Statistical Yearbook, and Liaoning Statistical Yearbook from 2000 to 2020, as well as the main data bulletins of the fifth, sixth, and seventh population censuses of these three provinces. (2) The input-output indicators for the ULUE (Table 1) were obtained from the China Urban Statistical Yearbook from 2000 to 2020, with carbon dioxide emission data originating from the National Earth System Science Data Center (https://www.ceads.net.cn/, accessed on 28 09 2023). (3) Data for the mediating and control variables were sourced from the China Urban Statistical Yearbook from 2000 to 2020 and the official websites of various municipal statistics bureaus. (4) Urban administrative division vector data were acquired from the Resource and Environmental Science and Data Center (https://www.resdc.cn/Login.aspx, accessed on 10 10 2023). (5) The 30 m × 30 m digital elevation model (DEM) raster data of the administrative division map were obtained from the geospatial data cloud (https://www.gscloud.cn, accessed on 10 10 2023).

Given that the model involves multiple mediating variables and that there may be homogeneity among these variables, it is necessary to consider the potential issue of multicollinearity.

Table 2. Descriptive statistics of the variables.

Indicator Name	Abbreviation	MIN	MAX	AV	SD	CV	VIF	Tolerance
Urban Land Use Efficiency	ULUE	0.390	1.740	0.844	0.235	0.278	——	——
Urban Population Growth and Shrinkage	UPGS	−0.350	0.260	−0.024	0.084	−3.509	1.425	0.702
Economic Development	ED	4370	106,846	31,710.633	22,025.296	0.695	2.659	0.376
Technology Innovation	TI	0.040	462.540	28.639	67.409	2.354	2.957	0.338
Industrial Structure Upgrading	ISU	9.780	60.990	39.039	7.726	0.198	5.370	0.186
Public Services	PS	4.440	100	23.067	22.264	0.965	1.673	0.598
Ecological Environment Quality	EEQ	7.050	51.800	35.780	7.769	0.217	1.349	0.741
Environment Governance Level	EGL	27.000	76,381.250	1915.802	10,838.006	5.657	1.062	0.941
Resource Utility	RU	12,455	4,840,417	641,738.871	748,699.945	1.167	2.842	0.352
Foreign Investment	FI	0.040	462.540	28.639	67.409	2.354	2.939	0.340

Note: Min represents the minimum value, Max represents the maximum value, AV represents the average value, SD represents the standard deviation, CV represents the coefficient of variation, and VIF represents the variance inflation factor.

provides detailed descriptive statistics of the independent variables, dependent variables, and mediating variables. On the basis of these statistical data, the selected variables did not exhibit multicollinearity, ensuring the stability and reliability of the regression model.

4. Results

4.1. Spatiotemporal Characteristics of ULUE and UPGS

4.1.1. Spatiotemporal Characteristics of ULUE

The ULUE in Northeast China exhibited significant spatiotemporal characteristics. In the temporal dimension (Figure 5), ULUE generally demonstrated a fluctuating upward trend during the study period. The average ULUE values of the three provinces, namely, Heilongjiang, Jilin, and Liaoning, all indicated similar increasing trends. Notably, there was a notable downward trend in ULUE between 2015 and 2020. In the spatial dimension (Figure 6), ULUE clearly demonstrated regional differences. The number of cities with medium to high ULUE values increased, and these cities were mainly concentrated in Liaoning Province, which features a more diversified industrial structure and an advantageous geographical location. In contrast, ULUE in Jilin and Heilongjiang Provinces was relatively low, which is related to their reliance on traditional agriculture and heavy industry. Moreover, certain cities with distinctive industries and favorable urban planning, such as Shenyang and Daqing, consistently maintained high ULUE levels.

4.1.2. Spatiotemporal Characteristics of UPGS

In the temporal dimension, population shrinkage in Northeast China has progressively intensified over time (Figure 7a). The shrinkage rate significantly increased from 30% in 2000 to 79% in 2020, highlighting the widespread nature of population loss. From a regional perspective (Figure 7b), Heilongjiang province has shown a continuous population shrinkage trend, with the shrinkage rate sharply rising from 6% to 29%. Meanwhile, Liaoning and Jilin provinces have experienced fluctuating population changes and shrinkage trends. In the spatial dimension (Figure 7c), the shrinkage phenomenon showed considerable regional imbalances. Cities exhibiting long-term population decline were mainly resource-based cities such as Dandong and Jixi, which face challenges due to resource depletion and imbalanced population structures. Conversely, cities with high development levels and notable competitiveness, such as Shenyang and Dalian, maintained consistent growth. These cities leveraged their siphoning effect and regional collaborative development to drive progress across the entire region.

4.2. Mediating Effects of UPGS on ULUE

4.2.1. Analysis of the Mediating Effect

Since the effects of urban population growth and urban population shrinkage on ULUE differ, these two phenomena were separately analyzed to gain a greater understanding of their respective impacts on ULUE. Before the analysis, the UPS indicator was subjected to negative preprocessing. This preprocessing step ensured that an increase in the value reflects a greater degree of urban population shrinkage.

4.2.2. Analysis of the Mediating Effect of Urban Population Growth

The results of the mediating effect model for urban population growth are listed in Figure 8a,c. There was a significant positive correlation between population growth and ULUE, with a total effect of 0.0925, indicating that population growth positively affects ULUE improvement. Specific mediating paths such as population growth → economic development → urban land use efficiency, population growth → technological innovation → urban land use efficiency, population growth → industrial structure upgrading → urban land use efficiency, and population growth → public services → urban land use efficiency amount to 0.0885. The order of the indirect path coefficients was as follows: economic development > public services > industrial structure upgrading > technological innovation. Notably, the direct effect coefficient was significantly lower than the coefficients of the mediating variables, highlighting the importance of the indirect effect of population growth on improving ULUE. From the perspective of comparing mediating effects, the proportions of the various mediating factors in the total mediating effect clearly varied. Economic development dominated, accounting for 82.15% of the total effect, whereas the proportions of industrial structure upgrading, public services, and technological innovation were 8.47%, 8.81%, and 0.56%, respectively. This finding indicates that economic development plays a major mediating role in the process of population growth promoting ULUE.

4.2.3. Analysis of the Mediating Effect of Urban Population Shrinkage

The results of the mediating effect model for urban population shrinkage are shown in Figure 8b,c. There was a significant negative correlation between urban shrinkage and ULUE, with a total effect of −0.0121. This result clearly indicated that urban shrinkage adversely affects ULUE improvement. Similarly, examining the four specific mediating pathways, the total indirect effect value is 0.1157, and could be ranked as follows: industrial structure upgrading > economic growth > technological innovation > public services. In terms of the proportion of each mediating factor, industrial structure upgrading and economic development dominated, accounting for 40.93% and 22.09%, respectively, of the total effect, with a combined proportion of 63.02%. Moreover, the proportion of technological innovation was only 1.63%, whereas public services accounted for −35.34% of the total effect, indicating that in the population shrinkage process, the mediating effect of technological innovation on ULUE improvement was still insufficient; public services imposed a significant negative mediating effect, exacerbating the negative impact of population shrinkage on ULUE.

4.2.4. Comparative Analysis of the Direct and Mediating Effects of Urban Population Growth and Shrinkage

The comparative results of the UPGS models are shown in Figure 8c. In terms of direct effects, population growth directly promoted ULUE, whereas population shrinkage clearly inhibited ULUE. In terms of mediating effects, both population growth and shrinkage could significantly enhance ULUE through the three core elements of economic development, technological innovation, and industrial structure upgrading, which are considered the three key driving forces for their improvement. However, it was noteworthy that while technological innovation had shown its potential to promote ULUE, its driving effect still required further enhancement and strengthening compared to economic development and industrial structure upgrading. Additionally, the impact of public services on ULUE varied depending on the development stage of the city. At the growth stage, the demand for public services increases and is subsequently improved to meet the basic needs of residents, such as education and healthcare, maintain social stability, promote economic development, and improve the quality of life, thereby causing ULUE improvement. However, at the shrinkage stage, service facilities may become excessive, leading to increased maintenance costs, thereby causing ULUE reduction.

4.2.5. Robustness Tests

To ensure the robustness of the multiple mediating effects of UPGS on ULUE, regression analysis was again performed after replacing the mediating and control variables. First, the consumer price index (CPI) was used to replace the per capita GDP as an indicator of economic development. Second, the ratio of the GDP of the tertiary industry to the GDP of the secondary industry in each region was calculated to assess the overall upgrading of the industrial structure, replacing the original variable of industrial structure upgrading. Finally, in terms of the environmental control variables, the environmental pollution index was adopted, with the specific calculation method derived from the literature [70,74]. The results after replacing the above variables are shown in Figure 9. The results indicate that the total effect coefficients, direct effect coefficients, and characteristic path mediating effects and significance values for the four types of mediating variables of population growth and shrinkage on ULUE did not change significantly, with only minor coefficient variations. This demonstrated that the model estimation results are robust.

4.3. Spatial Effects of UPGS on ULUE

4.3.1. Spatial Correlation Test

Setting of the Spatial Weight Matrix

The spatial weight matrix is a core element of spatial econometric models, reflecting the spatial relationships among geographical elements. The main types of spatial weight matrices are the 0–1 adjacency matrix, spatial geographic distance weight matrix, spatial economic distance weight matrix, and spatial economic geographic nested matrix. In this empirical study, a 0–1 matrix was used on the basis of the first law of geography. The constructed matrix is expressed as Equation (12).

$$W_{ij}=\{\begin{array}{l}1,\begin{array}{cc}& i\begin{array}{c}\end{array}and\begin{array}{c}\end{array}j\begin{array}{c}\end{array}neighboring\end{array}\\ 0,\begin{array}{cc}& not-neighboring\end{array}\end{array}$$

(12)

Spatial Correlation Analysis

When exploring the potential relationship between ULUE and UPGS, ensuring consistency between the two variables in the temporal dimension is essential. Therefore, during data preprocessing, the time-averaging method was used, specifically by calculating the average value of ULUE data for two adjacent years to represent the value for a specific period. The spatial correlation between ULUE and urban growth and shrinkage characteristics in Northeast China was subsequently analyzed.

According to

Table 3. Spatial correlation analysis of UPGS and ULUE.

	ULUE			UPGS
	Global Moran’s I	Z Score	p Value	Global Moran’s I	Z Score	p Value
2000–2005	0.072	1.427	0.081	0.113	1.288	0.006
2005–2010	0.120	1.333	0.090	0.218	2.333	0.017
2010–2015	0.143	1.551	0.070	0.192	2.171	0.025
2015–2020	−0.132	−1.012	0.161	0.133	1.496	0.047

, there were significant differences in Moran’s I and its changes between ULUE and UPGS in Northeast China. Specifically, the global Moran’s I of the ULUE showed a significant positive correlation during the three consecutive periods of 2000–2005, 2005–2010, and 2010–2015, whereas it exhibited a negative correlation during the subsequent period of 2015–2020. In contrast, the global Moran’s I of UPGS remained significantly positively correlated throughout all periods. Regarding ULUE, although the values during the 2015–2020 period did not pass the significance test, the shift from a positive correlation to a negative correlation indicates that ULUE in Northeast China significantly changed during this period. Regarding UPGS, Moran’s I remained positive and showed a fluctuating upward trend, clearly revealing that the spatial clustering degree of urban growth and shrinkage has gradually increased over time.

4.3.2. Empirical Results and Analysis of the Spatial Econometric Model

Selection and Assessment of Spatial Econometric Models

The previous spatial correlation tests demonstrate that there is a positive spatial correlation between UPGS and ULUE in Northeast China. Therefore, when studying the impact of urban growth/shrinkage on ULUE, it is essential to consider geographic spatial factors and interaction terms. The SLM, SEM, and SDM can all be used for result estimation, but the appropriate model should be selected on the basis of the test results and decision rules, as detailed in

Table 4. Statistical test results for spatial econometric model selection.

Model	Test	Statistic	p Value
SLM and SEM selection	LM error	7.899	0.005
	R-LM error	17.952	0.000
	LM lag	4.344	0.037
	R-LM lag	14.396	0.000
SDM simplification	Wald lag	26.510	0.002
	LR lag	6.680	0.083
	Wald error	30.220	0.000
	LR error	182.07	0.000
Random and fixed effects tests	Hausman	80.720	0.000

:

(1) Selection of the SLM with the SEM: Both the LM and RLM test results for the SLM and SEM are significant, indicating that both models are preferable to the original model. Further testing is needed to determine the suitability of the SDM. (2) Simplification of the SDM: The Wald and LR statistics are significant, indicating the rejection of the hypothesis that the SDM can be simplified to the SLM or SEM. This suggests that the SDM provides more accurate estimates for analyzing the impact of urban growth/shrinkage on ULUE. (3) Fixed effects with random effects test: After the SDM model is selected, the Hausman test statistic is 80.72, which is significant, rejecting the random effects model. Therefore, a fixed effects SDM model should be used.

Analysis of the Spatial Durbin Model Estimation Results

The estimation results of the no fixed effects SDM, time-fixed effects SDM, spatial fixed effects SDM, and double-fixed effects SDM were analyzed. On the basis of the results obtained with the models with different interaction effects (

Table 5. Basic regression results of the spatial Durbin model.

Main	Mixed Effect	Time Fixed	Spatial Fixed	Double Fixed		Mixed Effect	Time Fixed	Spatial Fixed	Double Fixed
R2	0.3943	0.7863	0.4459	0.5988	Wx
ED	0.046	0.322 ***	0.368 *	0.153 **	ED	−0.326 ***	−0.182 ***	−0.195 *	0.339 ***
ISU	−0.193 *	0.575 **	0.306 **	0.221 **	ISU	−0.851 ***	−0.216 *	−0.211 ***	0.596 ***
PS	−0.190	0.154 ***	0.209 *	0.115 **	PS	−0.041	0.034	0.012	0.200 *
TI	−0.105	0.244 *	0.158 *	0.199 *	TI	−0.461 **	0.203 **	0.385 **	0.436 **

Note: *, **, and *** indicate significance levels of parameters at 10%, 5%, and 1%, respectively.

), the model with the highest R2 value was selected. Ultimately, the time-fixed effects SDM achieved the highest goodness-of-fit, so this model was chosen. The results in Table 5 indicate that ED, ISU, PS, and TI all significantly impact ULUE in Northeast China. Moreover, the coefficients W*ED, W*ISU, and W*TI are significant at the 10% level, indicating that these mediating factors impose spatial spillover effects on ULUE.

Owing to the presence of spatial spillover effects in the SDM, the impact of each variable on the dependent variable cannot be interpreted solely on the basis of the regression coefficients. Therefore, it is necessary to examine the direct effect and indirect effect of each variable on the dependent variable. The direct and indirect effects of the SDM on ULUE in Northeast China are shown in Figure 10.

Direct Effects of the SDM

Economic development, industrial structure upgrading, technological innovation, and public services all yield significant positive impacts. Among these factors, industrial structure upgrading most notably impacted ULUE, with an effect coefficient of 0.444. This factor was followed by economic development, technological innovation, and public services, with direct effects of 0.320, 0.251, and 0.145, respectively, indicating that these factors remain crucial for ULUE improvement. Notably, the overall analysis demonstrated that industrial structure upgrading is critical for enhancing ULUE.

Indirect Effects of the SDM

Notably, economic development and industrial structure upgrading in neighboring cities negatively impact ULUE in the local region, indicating that these factors in adjacent cities can reduce ULUE locally. In contrast to these factors, the indirect effect of technological innovation is positive, suggesting that there is a notable knowledge spillover effect between cities. The technological innovation capabilities of central cities impose positive spillover effects on surrounding cities. Moreover, although public services positively affect ULUE, this effect does not pass the significance test.

5. Discussion

This study utilizes panel data from 34 prefecture-level cities in Northeast China (excluding Daxinganling and Yanbian Korean Autonomous Prefecture) from 2000 to 2020 to explore the impact of urban population growth and shrinkage on urban land use efficiency from the perspectives of mediation pathways and spatial effects. The findings reveal significant differences in how population growth and shrinkage affect ULUE: population growth promotes ULUE, while population shrinkage significantly hinders it. This result aligns with existing research [7,75]. Furthermore, there is a notable regional development imbalance in Northeast China. In this context, urban population changes affect ULUE through different mediation pathways and spatial effects. Regardless of the specific forms of impact, the ultimate goal is to better utilize land and improve land use efficiency. Overall, there is still considerable room for improvement in regional coordinated development and ULUE enhancement.

5.1. Analysis of the Significant Regional Development Imbalance in Northeast China

Northeast China exhibits severe population outflow, with numerous resource-based cities facing economic stagnation and outdated industrial structures due to resource depletion, exacerbating population loss [76]. In response, the Chinese government introduced the Revitalize the Northeast strategy in 2003, and subsequent policies aimed at transforming and upgrading old industrial bases. The industrial structure in the region subsequently shifted from heavy industries to high-tech industries, modern services, and modernized agriculture, leading to diversified land use, improved urban planning, and significantly enhanced land use efficiency (ULUE). However, in 2015, China’s structural reform aggravated the challenges in Northeast China, which relies heavily on heavy industry and resources. Core cities with robust economic foundations and diverse industrial structures demonstrate greater resilience and leadership amid risks, thereby intensifying regional development differences [77]. Simultaneously, there exists a significant imbalance in economic factor allocation. Core cities attract more resources and capital, whereas resource-based cities face the dual challenges of resource depletion and industrial transformation. Similarly, innovation resources are mainly concentrated in a few large cities, further exacerbating regional development imbalances. This uneven allocation results in the concentration of core development elements in major cities, widening regional development gaps. Under this imbalance, factors such as population and capital flow from smaller cities to central cities, leading to spatial agglomeration of population and land use. While this concentration contributes to overall development and balance, it necessitates addressing deficiencies to promote regional coordinated development. Specific measures include enhancing policy support for resource-based cities, optimizing public resource allocation, promoting industrial diversification, and increasing innovation capabilities to achieve holistic regional coordination and sustainable development.

5.2. Analysis of the Impacts of UPGS on ULUE

In this study, it was revealed that UPGS exert contrasting impacts on ULUE. Population growth stimulates an increased land demand, reduces idle land use, and promotes orderly economic activities, attracting other factors to urban centers and thus increasing ULUE. In contrast, population shrinkage leads to economic downturns in various sectors, such as commerce and industry, slowing urbanization and causing outflows of talent and labor. These factors collectively result in land resource waste and decreased land use efficiency.

Furthermore, it explored the core driving forces for improving urban land use efficiency from multiple dimensions. Economic development, technological innovation, and industrial structure upgrading were identified as the three key elements, all of which exhibited spatial spillover effects, consistent with existing research findings [16,17,21]. Specifically, economic growth serves as the cornerstone of national stability and development, satisfying various macrolevel demands and indirectly promoting land use improvements and optimizations. Moreover, economic activities driven by profit motives generate spatial mobility, determining spatial economic patterns. Technological innovation is pivotal in enhancing efficiency, thereby driving improvements in information infrastructure and management efficiency, and positively impacting surrounding areas through knowledge spillover phenomena. Industrial structure upgrading is a critical strategy for high-quality regional development, thereby adjusting land resource usage and configuration to better align with current economic trends and demands, thus creating more resources and market opportunities.

Additionally, this study also examined how population dynamics impact public services, which vary with urban development stage. Population growth triggers economies of scale, increasing the overall efficiency of public service delivery. Conversely, population shrinkage leads to surplus resources and waste. However, regional constraints on public services and government-led service delivery models slightly limit effective dissemination between different regions, thereby reducing their spatial spillover effects. Addressing these issues will require attention and improvements in future policy formulation.

In particular, while technological innovation serves as a core driver, it remains notably deficient in the Northeast region. Compared to developed areas like the Yangtze River Delta and Beijing–Tianjin–Hebei, the Northeast lags significantly behind in funding, talent, and services for technological innovation. Additionally, the low efficiency in transforming technological achievements also restricts the direct contribution of technological innovation to regional economic development. Therefore, in the process of revitalization, the Northeast should place high importance on optimizing the allocation and enhancing the efficiency of technological innovation resources. This will enable technological innovation to lead and promote high-quality development of the regional economy.

5.3. Analysis of Coordinated Development to Improve ULUE

Economic development, technological innovation, and industrial structure upgrading are not only the three key factors for enhancing ULUE but also exhibit closely interconnected and interdependent relationships. In the process of economic development, the accumulation of capital and resources lays a solid foundation for technological innovation. Technological innovation, in turn, promotes continuous economic growth by increasing productivity, reducing production costs, and improving product quality. Simultaneously, economic development provides the material basis and broad market demand for industrial structure upgrading. Through industrial structure adjustment and optimization, more rational resource allocation can be achieved, thereby enhancing the overall economic efficiency. Moreover, technological innovation plays a crucial role in promoting industrial structure upgrading. By introducing new technologies, processes, and materials, technological innovation effectively drives the transformation of industries from low- to high-end development stages. The greater application scenarios and market demands resulting from industrial structure upgrading further depend on the continuous promotion of technological innovation. Therefore, the coordinated development of these three elements is vital for improving ULUE.

Within the context of global economic transformation and upgrading, pursuing high-quality economic development has become the consensus and a strategic goal for governments worldwide. In 2023, China proposed the concept of new quality productivity forces, which fully embodies the coordinated role of economic development, technological innovation, and industrial structure upgrading. It positioned economic development as the foundation of productivity, innovation as the driver of productivity, and industrial structure upgrading as the goal of productivity, aiming to overcome the constraints of traditional economic growth models and becoming a key driver and focus point for promoting high-quality development [78]. This concept emphasizes breakthroughs in green and low-carbon technological innovations and focuses on the green and low-carbon transformation and upgrading of industries, ensuring ecological security [79]. It notably reflects the core principle of innovation-lead development and green-driven progress in shaping the future. Similarly, other countries and regions worldwide are actively exploring similar development paths. For example, in response to the severe challenges of global climate change, Europe proposed the European Green Deal in 2020, reflecting the goals of eco-friendliness and sustainable development [80]. Additionally, both Japan’s Society 5.0 and Singapore’s Smart Nation 2025 emphasize the critical role of technological innovation in driving comprehensive economic and social upgrades. Among these global policy trends, a common theme emerges, namely, governments worldwide are carefully planning and implementing policies aimed at promoting sustainable social construction through coordinated development.

5.4. Limitation

This study has certain limitations, primarily in two aspects. First, the measurement of urban population changes and land use efficiency requires more refinement and spatial precision. Although traditional data were used to analyze the overall city conditions, future research should integrate big data (e.g., POI data, street view data) with survey data to reveal spatial issues more accurately. Integrating multiple data sources will help better analyze the impact of regional population growth and shrinkage on land use efficiency and provide new insights into the reasons for local population loss and future regional governance.

Second, while this study focuses on the impact of population growth and shrinkage on land use efficiency in Northeast China, the findings may not fully reflect dynamics in other regions. Future research should expand to include case studies from other typical regions or urban agglomerations in China to capture commonalities and differences and explore universal patterns applicable to different regions and nationwide.

5.5. Summary

In summary, regional coordinated development and ULUE enhancement are global challenges. The regional development imbalance in Northeast China has worsened due to population loss and resource depletion. While core cities have shown resilience, the uneven distribution of resources across regions has further intensified the disparities. In this context, promoting coordination among technological innovation, industrial structure upgrading, and economic development is key to enhancing ULUE. This trend is being advanced in China, and other countries worldwide are exploring similar development paths. Future policymaking will need to effectively coordinate economic, technological, and industrial structure aspects to drive balanced regional development and sustainable societal construction.

6. Conclusions and Policies Implications

6.1. Conclusions

This paper focused on the impacts of UPGS on ULUE. First, the core concepts of UPGS and ULUE were explained theoretically. Next, the interaction mechanisms between these two variables, including direct, mediation, and spatial effects, were analyzed in detail. The current status of UPGS and ULUE during the research period was comprehensively evaluated, providing a thorough analysis of the measurement results. Finally, with the use of mediating effect models and SDMs, the specific impact pathways and spatial effects of UGSD on ULUE were empirically revealed. The main research conclusions are as follows:

Significant Regional Development Imbalance in Northeast China: Northeast China exhibits a considerable regional development imbalance. Core cities in the region (e.g., Shenyang, Dalian) maintain relatively high ULUE levels, while resource-based cities (e.g., Hegang, Shuangyashan) show a significant decline in ULUE due to population shrinkage and industrial decline. Although policy guidance has somewhat promoted the transformation of resource-based cities, regional development imbalances remain prominent, necessitating more refined and differentiated policy support.

Dual Effects of Population Growth and Shrinkage on ULUE: Population growth has a clear positive effect on ULUE, particularly in core cities where population concentration leads to optimized resource allocation and enhanced economic vitality, resulting in more intensive and effective land use. Conversely, population shrinkage results in decreased ULUE, especially in resource-based and small-to-medium cities, where land idleness and waste are severe, further exacerbating regional development imbalances.

Key Factors for Enhancing ULUE: Economic development, technological innovation, and industrial structure upgrading are crucial for improving ULUE. Economic growth and technological innovation enhance ULUE by optimizing land use structure, increasing land use density, and improving resource management. Additionally, industrial structure upgrading further promotes the efficient allocation of land resources. However, these factors exhibit significant spatial heterogeneity among cities, indicating that policymaking should be tailored to local conditions.

Impact of Public Services on ULUE at Different Urban Development Stages: The impact of public services on ULUE varies significantly depending on the stage of urban development. In cities experiencing population growth, improvements in public services significantly promote the intensive use of land resources. In contrast, in cities experiencing population shrinkage, the oversupply or waste of public service resources leads to inefficient land use. Therefore, the provision of public services should be flexibly adjusted according to urban population dynamics to maximize land use efficiency.

6.2. Policy Implications

Based on the findings of this study, the following policy implications are proposed regarding the impact of urban population growth and shrinkage on land use efficiency:

Tailored Policies: Policies should be tailored to local conditions, developing differentiated land use policies based on population trends in different cities. For cities with sustained population growth, optimizing land resource allocation to prevent disorderly expansion and resource waste is essential, aiming to enhance land use efficiency. For cities with population shrinkage, the government should implement incentive measures to promote the integration and reuse of land resources, reducing land idleness and waste.

Role of Technological Innovation: Technological innovation plays a key role in enhancing land use efficiency. The government should increase investment in the development of land management technologies, promoting information and intelligent management models. At the same time, optimizing the layout of public service facilities to match urban population sizes is crucial to avoid resource waste and service inadequacies.

Dynamic Monitoring Mechanism: The government should establish a dynamic monitoring mechanism to track changes in population and land use in a timely manner, ensuring the flexibility and sustainability of policies. Accurate policy interventions will better promote the rational and efficient development of land resources and support sustainable urban development.