How Does Population Aging Affect New Quality Productivity in Economic Sustainability? An Empirical Study Based on Mediating Mechanisms and Moderating Effects

Abstract

Population aging is increasingly recognized as a challenge to sustainable economic development, raising concerns about its impact on innovation and productivity. This study examines how population aging affects “new quality productivity” in China, using a balanced panel dataset of 30 provinces from 2011 to 2022. The analysis employs panel regression models with fixed effects and incorporates mediation and moderation approaches to explore underlying pathways. Land productivity is identified as a significant channel through which aging influences productivity, while the level of urbanization is examined as a moderating factor in this relationship. The results indicate that population aging significantly inhibits new quality productivity. Specifically, an aging population leads to lower land productivity, which hinders the growth of new quality productivity. However, higher urbanization is found to mitigate the adverse effect of aging on productivity. These findings are robust under various model specifications and statistical checks. In conclusion, the study underscores the necessity for proactive measures to mitigate the adverse effects of demographic aging. The findings provide policy insights, suggesting that boosting technological innovation, improving agricultural efficiency, and leveraging urbanization can help sustain high-quality development in the face of an aging population.

1. Introduction

With the acceleration of the global demographic transition, population aging has become a significant challenge to the socio-economic development of all countries [1]. Many countries are witnessing a decline in the share of young people within their populations, accompanied by falling birth rates and a sustained shift toward an aging demographic structure. According to United Nations projections, by the mid-twenty-first century, the global population aged 65 and above will surpass younger cohorts, leading to the gradual erosion of the demographic dividend [2]. Against this backdrop, the critical question is how to convert the traditional demographic dividend into a talent dividend through technological innovation, labor force restructuring, and institutional adaptation—a challenge that countries across the globe must urgently confront [3].

At the same time, both scholars and policymakers have increasingly highlighted the significance of productivity transformation. Within the framework of traditional Schumpeterian growth theory, technological innovation is regarded as the fundamental engine of sustained economic growth [4]. The extended model of total factor productivity (TFP) underscores the interdependence among humans, capital, institutional environments, and innovation ecosystems [5]. In recent years, driven by advances in digitalization, the green transition, and the implementation of the Sustainable Development Goals (SDGs), the notion of productivity has undergone a significant transformation. It is no longer confined to the efficiency of factor allocation but has expanded to include new dimensions such as innovation-driven development, digital infrastructure, and ecological sustainability [6]. Within this evolving framework, “new quality productivity” can be conceptualized as the synergistic force of digitalization, green transition, and innovation, designed to foster high-quality economic growth aligned with the Sustainable Development Goals.

First, at the level of labor force structure, population aging may undermine the accumulation of human capital and weaken technological absorptive capacity, thereby slowing the diffusion of new technologies and hindering industrial upgrading [7]. Second, at the fiscal level, aging heightens pressure on public expenditures related to pensions, healthcare, and social security, which may constrain investment in research and development, digital infrastructure, and green initiatives through a “fiscal crowding-out effect” [8]. Finally, from the perspective of capital markets, population aging may reduce societal tolerance for high-risk innovative activities, diminish financing flows to emerging industries, and thereby indirectly inhibit the enhancement of productivity quality [9].

Previous studies have investigated the effects of population aging on economic growth and labor markets [10,11]. Most of this literature highlights its indirect influence on traditional productivity or innovation. Yet, few have systematically examined its direct impact on new quality productivity, characterized by digitalization, green transformation, and innovation. While an aging workforce may bring extensive experience, research suggests that its overall learning capacity and willingness to adopt new technologies tend to decline, thereby impeding technological progress and sustainable innovation [2,12]. Moreover, existing evidence has been derived predominantly from developed countries [3], with limited attention to sustainable development pathways such as digital economic growth, industrial upgrading, and green transitions. This gap is particularly salient in developing economies like China, where a large population and pronounced regional disparities amplify the complexity of the relationship between aging and new quality productivity. To date, these linkages and their underlying mechanisms have not been systematically or empirically investigated. Consequently, there is an urgent need for more in-depth research situated within the framework of sustainable development, explicitly incorporating intermediary mechanisms and moderating effects.

This study addresses this gap by investigating how population aging influences new quality productivity, through the mechanisms by which this influence operates, and whether moderating factors can mitigate its effects. Drawing on panel data from 30 Chinese provinces and applying econometric regression techniques, the analysis systematically examines the relationship between population aging and new quality productivity, with multiple robustness checks—including outlier treatment, alternative fixed-effects specifications, and instrumental variable approaches—ensuring the reliability of the results.

The findings indicate that population aging is significantly negatively associated with both new quality productivity and land productivity. Further analysis reveals that land productivity acts as a mediating channel, while urbanization development moderates this relationship, particularly in central and western regions, as well as in high-tech, service-oriented provinces. These results contribute to the theoretical discourse on population economics and innovation-driven growth by situating aging within the framework of new quality productivity. The study also offers policy-relevant insights by highlighting how urbanization and structural upgrading can buffer the adverse effects of demographic aging, thereby providing strategies for developing economies to achieve the dual goals of actively addressing population aging and promoting high-quality development.

The remainder of this paper is structured as follows. Section 2 reviews the relevant literature and outlines the theoretical framework. Section 3 introduces the variables, model specification, and data sources. Section 4 presents the empirical results, including: (1) baseline analysis of the impact of aging on new quality productivity; (2) robustness tests; (3) analysis of the mediating mechanism; (4) analysis of the moderating mechanism; and (5) heterogeneity analysis. Section 5 discusses the empirical findings. Section 5 concludes the paper and offers policy recommendations.

2. Literature Review

2.1. Population Aging and Urbanization

A substantial body of research has examined the economic consequences of population aging [13,14]. At the macroeconomic level, international studies consistently show that population aging exerts downward pressure on growth and productivity by reducing labor participation and savings rates. For example, Bloom et al. [15] argue that aging undermines innovation and human-capital–driven growth trajectories. Using European panel data, it has been shown that workforce aging significantly reduces labor productivity growth, primarily through its negative effects on total factor productivity [16]. Mason et al. [17] further highlight that Asia’s sustainable development is at risk from demographic shifts, calling for a dual strategy that combines the transition from a “demographic dividend” to a “talent dividend” with green transformation.

At the sectoral level, the adverse effects of aging are particularly evident in agriculture. Feder [18] and Munnich et al. [19] find that aging increases healthcare and pension expenditures, reduces agricultural capital investment, and slows the adoption of advanced agricultural technologies, thereby constraining productivity growth. These findings are not limited to China [20] but are also applicable to other developing economies. In the Chinese context, recent studies indicate that population aging hinders economic growth by reducing effective labor supply, dampening household consumption—particularly in urban areas—and altering, as well as decelerating, the trajectory of urbanization [21]. Taken together, this evidence suggests that population aging has both common global impacts and context-specific transmission mechanisms in developing economies.

Meanwhile, urbanization is widely recognized as both a driver of economic development and a potential moderator of aging’s effects. From a general perspective, urbanization facilitates factor agglomeration, improves spatial structure, and strengthens governance capacity [22,23]. Within the framework of sustainable development, urbanization drives the transformation of green spaces and advances smart city governance, enhancing the efficiency of public service delivery and promoting environmental sustainability through the integration of digital infrastructure and institutional coordination [24]. These mechanisms align closely with the goals of Sustainable Development Goal (SDG) 11 on sustainable cities and communities, which emphasize inclusiveness, safety, resilience, and sustainability [25].

However, the effects of urbanization are conditional on factors such as human capital, institutional quality, and governance capacity [26]. Empirical evidence from China suggests heterogeneity: the adverse impact of population aging on labor productivity is mitigated in highly urbanized regions where demographic dividends remain, but becomes more pronounced in less urbanized areas [27]. Wang et al. [28] further reveal a nonlinear threshold effect in the interaction between aging, urbanization, and environmental outcomes, with notable differences across countries at varying income levels.

In addition, international urban studies highlight potential risks. Seto et al. [29,30] show that global urban expansion can either serve as a key lever for sustainable transformation or exacerbate ecological pressures, depending on governance and planning quality [31]. Hsieh and Moretti [32] introduce the concept of “spatial aging” in megacities, where high housing costs and rising living expenses displace younger populations to peripheral areas, resulting in long commutes and labor mismatches that diminish the potential productivity spillovers of urbanization.

2.2. New Quality Productivity

Currently, as China’s economy transitions toward high-quality development, the cultivation of new quality productivity, driven by innovation, technological advancement, and efficient synergy, is becoming a core engine for achieving sustainable growth [33]. In recent years, the concept of new quality productivity has been introduced in academic discourse as a productivity paradigm oriented toward future industries and emerging growth drivers [34]. At its core, it reflects a shift from quantity-based expansion to innovation-driven and quality-oriented development [35].

Scholars have initiated multidimensional discussions on new quality productivity, analyzing its implications from spatiotemporal [34] and structural perspectives, defining its conceptual scope [36], and offering theoretical interpretations [37]. Compared with traditional productivity metrics centered on factor inputs and total factor productivity (TFP), new quality productivity emphasizes advanced technologies, digitalization, and green production, which represents a qualitative leap in productivity [38]. This conceptualization resonates strongly with Schumpeterian innovation-driven growth theory and aligns closely with the principles of SDG 8 (Decent Work and Economic Growth) and SDG 9 (Industry, Innovation, and Infrastructure) [39]. As a new paradigm of economic growth, new quality productivity underscores the transition from scale-oriented expansion to quality-oriented upgrading, relying on digital, intelligent, and green drivers to balance efficiency with sustainability.

In this context, new quality productivity plays a pivotal role in enhancing economic efficiency and fostering long-term sustainable development. Empirical studies indicate that digital infrastructure and the digital economy substantially improve resource allocation efficiency and stimulate green growth [40]. This pattern has been confirmed by empirical evidence from both OECD countries and emerging economies [41]. Furthermore, digital infrastructure reduces information transmission costs, improves supply chain coordination, and strengthens firms’ operational efficiency and innovation capacity [42]. Green technologies and the energy transition are also critical channels for driving qualitative changes in productivity, as they not only ease environmental constraints but also create new sources of competitive advantage [43,44]. In the long run, the development of new quality productivity contributes to building a high-quality growth engine, thereby fostering sustainable economic growth and enhancing international competitiveness [38]. This implies that, in both advanced and developing economies, policy packages centered on digital transformation, green transition, and innovation system building constitute key pathways for shaping the future landscape of productivity.

2.3. Hypothetical Development

At the macroeconomic level, population aging shows a significant correlation with declining labor supply, thereby exerting downward pressure on economic growth and productivity. Maestas et al. [45] demonstrate that a 10-percentage-point increase in the proportion of the population over 60 years of age in the United States is associated with a reduction in GDP per capita of approximately 5.5 percent. The decline in GDP per capita is estimated to result primarily from a slowdown in the growth rate of labor productivity, accounting for about two-thirds of the total reduction. However, from a micro perspective, some studies reveal opposite effects. Tan et al. [46], using firm-level data, find that population aging—by intensifying labor costs and scarcity—may stimulate labor-substituting innovation, thereby enhancing firms’ innovation output.

In summary, existing literature presents mixed findings regarding the relationship between aging and productivity. These findings provide a valuable foundation for exploring the implications of aging for productivity improvement. However, existing studies have predominantly focused on macroeconomic growth rates or general productivity indicators, while the internationalized discussion of new quality productivity remains at a nascent stage [47].

H1. 

Population aging is significantly negatively associated with the development of new quality productivity.

In the dimension of land use, population aging may affect land productivity by altering the allocation patterns between labor and arable land. Land productivity, in turn, has been recognized as a core efficiency indicator in sustainable agriculture and food supply systems, consistent with the objectives of SDG 2 (Zero Hunger) and SDG 12 (Responsible Consumption and Production). Jian et al. [48] use panel data on Chinese farm households and find that rural population aging significantly facilitates land turnover and enhances factor utilization efficiency, thereby indirectly improving productivity. This mechanism is consistent with the “factor reallocation” logic in Schumpeterian growth theory. Moreover, given the close interlinkages between agriculture and urban economies through industrial chains and factor flows, changes in land use efficiency not only influence rural production performance but also generate spillover effects on urban economies and innovation ecosystems via agricultural supply, capital transfers, and labor mobility. Accordingly, this mechanism is not unique to China but demonstrates cross-regional applicability. For instance, in Latin America and Southeast Asia, population aging and rural labor outmigration have likewise facilitated land consolidation and mechanization [49,50], which are comparable to China’s experience. At the same time, some studies caution that land integration and efficiency gains, if not accompanied by proper institutional constraints, may entail ecological risks and environmental pressures [51]. The following research hypothesis is proposed:

H2. 

Land productivity is a significant mechanism through which population aging affects the development of new quality productivity.

In the context of urbanization, labor migration and factor agglomeration can help mitigate the pressures of a declining labor supply and evolving consumption patterns associated with population aging [52]. The positive effects of aging on labor productivity are mainly observed in highly urbanized cities, while they are insignificant in less urbanized areas. This suggests that in regions with larger economic scale, stronger industrial upgrading, and greater capital investment, the impacts of aging are more readily offset through technological substitution and scale externalities [53]. This mechanism is consistent with the “urban attractiveness” and “migration adaptation” theories [54], and it highlights the pivotal role of urbanization in advancing Sustainable Development Goal 11 (Sustainable Cities and Communities).

H3. 

Urbanization moderates the impact of aging on new quality productivity.

As shown in research by Liu et al. [55], there are substantial regional disparities in both economic structure and the efficiency of factor allocation. Labor misallocation is most pronounced in less-developed regions, whereas economically stronger regions exhibit clear advantages in enhancing new quality productivity. Such regional disparities are evident not only in China but also in Latin American and Eastern European countries [56]. Moreover, industries respond differently to population aging. Scholars [57] find that rising aging levels exacerbate labor shortages, which reduce the capacity for industrial upgrading. In the current macroeconomic environment, the net effect of aging on technological advancement remains uncertain. However, technological progress continues to play a pivotal role in driving structural transformation. In conclusion, regional and industrial structural differences shape the pathways through which population aging and urbanization influence new quality productivity. Based on these insights, the following hypothesis is proposed:

H4. 

Population aging, urbanization, and new quality productivity are interrelated through complex and heterogeneous mechanisms.

3. Methods and Data

3.1. Data Source

The research dataset constructed in this paper comprises a balanced panel of 30 provinces in China from 2011 to 2022, resulting in a total of 360 sample observations (the details are as shown in Appendix A). Notably, Tibet, Hong Kong, Macau, and Taiwan were excluded from the empirical analysis owing to significant data gaps. In cases where individual indicators exhibited missing values, this paper follows the approach proposed by Zhang et al. [58], applying linear interpolation to fill the gaps, thereby ensuring data consistency and integrity.

The primary data sources include authoritative publications from the National Bureau of Statistics of China, such as the China Statistical Yearbook, China Energy Statistical Yearbook, China Industrial Statistical Yearbook, China Environmental Statistical Yearbook, China Financial Yearbook, and various provincial statistical yearbooks, which collectively ensure the accuracy and credibility of the dataset.

Descriptive statistics of variables are shown in

Table 1. Descriptive statistics of variables.

Category	Indicators	Variable	Count	Mean	Std. Dev.	Min	Max
Dependent Variable	New Quality Productivity	nqp	360	0.2852	0.1307	0.1019	0.7473
Independent Variable	Population Aging	aging	360	15.5624	4.4060	7.4000	28.8000
Mediating Variable	Land productivity	lp	360	0.5766	0.5043	−0.6579	1.9308
Control Variable	Government Intervention Degree	ur_gap	360	2.5468	0.3789	1.8266	3.6716
	Industrialization Level	digint	360	0.0669	0.1401	0.0151	2.5204
	Financial Development Level	in_fdi	360	11.4778	1.4639	7.9477	15.5508
	Urbanization Level	in_elec	360	7.4349	0.6964	5.2219	8.9708
moderating variable	Human Capital Level	urban	360	0.6012	0.1205	0.3500	0.9000

. To address potential heteroskedasticity and reduce variability due to differences in indicator scales, this study applies logarithmic transformation and standardization to selected continuous variables [59]. This procedure enhances the comparability of variables across provinces and over time.

3.2. Variable Description

3.2.1. Dependent Variables

In this study, the development of new quality productivity serves as the dependent variable. Its essence lies in the integrated manifestation of production efficiency and development potential, driven by innovation, green orientation, and digital empowerment. Drawing on existing literature and integrating the two-factor productivity theory with the Schumpeterian perspective of innovation-driven growth [4], this study conceptualizes new quality productivity across five primary dimensions: (1) Human Capital and Labor Productivity, (2) Workforce Engagement and Culture, (3) Physical Infrastructure and Resource Efficiency, (4) Innovation and Digital Capabilities, and (5) Environmental Sustainability. Within this framework, further secondary and tertiary indicators are specified—such as economic output, human capital, industrial structure, innovation capacity, and digital infrastructure (see

Table 2. Evaluation indicators of digital economy development level.

Indicator Dimension	Primary Indicator	Secondary Indicator	Measurement Method	Direction of Effect
Basis for quality development	Economic development	Economic output	GDP	+
		Economic revenue	Average Wage of Employed Staff	+
		Employment structure	Employment in Tertiary Industry/Total Employment	+
	Human capital	Educational attainment	Average Years of Schooling per Capita	+
Basis for quality development	Human capital	Cultural resources	Education Expenditure/Fiscal Expenditure	+
		Knowledge potential	Number of Enrolled Students/Total Population	+
	Innovative entrepreneurship	Innovative spirit	Full-time Equivalent R&D Personnel in Above-scale Industrial Enterprises	+
		Entrepreneurship	Number of New Enterprises per Million People	+
Green development capacity	Digitalization and future industries	Level of informatization	Number of E-commerce Active Enterprises/Total Enterprises	+
		Future industries	Regional Industrial Robot Installations × (Regional Industrial Employment/ National Industrial Employment)	+
	Green development	Green development	Forest Coverage Rate	+
			Environmental Protection Expenditure/General Fiscal Expenditure	+
		Green production	COD Emissions/GDP	−
			SO2 Emissions/GDP	−
			Number of Green Patent Applications/Total Patent Applications	+
The Driver of Innovation	Factor Support	Infrastructure	Highway Mileage	+
			Railway Mileage	+
			Fiber Optic Cable Length	+
			Internet Access Ports per Capita	+
		Energy efficiency	Energy Consumption/GDP	−
		Energy use potential	Exhaust Gas Treatment Capacity	+
	Innovation drive	Technological Innovation Level	Number of Patent Grants/Total Population	+
			Expenditure on New Product Development/GDP	+
		Digitalization level	Digital Economy Index	+
			Enterprise Digitalization Level	+

)—to comprehensively capture both the structural characteristics and external manifestations of new quality productivity.

For the weighting methodology, this paper employs the entropy method. First, raw data are standardized to eliminate the influence of differing units of measurement. Then, following the principle of information entropy, weights are assigned to each indicator, giving greater emphasis to those with higher variability and differentiation to reduce subjective bias. Finally, standardized indicators are weighted and aggregated to generate a composite score of new quality productivity for each province and year [60]. Furthermore, principal component analysis (PCA) was employed to reveal the underlying dimensional structure of the indicator system [61]. The PCA results reveal that the eigenvalues of the first four components are all greater than one, collectively accounting for 62.5% of the total variance. Notably, the first three components can be interpreted as corresponding to the dimensions of digitalization, green development, and innovation, thereby confirming the statistical validity and robustness of the new quality productivity indicator system.

This evaluation framework integrates key dimensions, including GDP, human capital, educational quality, innovation output, digital economy development, pollution control efforts, green production capacity, and infrastructure completeness. Each indicator is selected for its systematic structure, representativeness, and policy relevance, enabling a comprehensive and robust measurement of new quality productivity across China. This framework aligns closely with key Sustainable Development Goals, including SDG 8 (Decent Work and Economic Growth), SDG 9 (Industry, Innovation, and Infrastructure), and SDG 11 (Sustainable Cities and Communities), thereby underscoring its value for cross-national comparison and policy learning.

3.2.2. Independent Variables

The core independent variable selected for this study is population aging. This variable reflects structural demographic shifts, characterized by a gradual decline in the proportion of the working-age population and a continuous increase in the elderly population, driven by persistently low birth rates and rising life expectancy [62]. Population aging is a critical societal factor that constrains labor supply, technological adaptability, and innovation potential [14].

The old-age dependency ratio is a widely adopted indicator in the existing literature for measuring the extent of population aging. This study follows that convention, using the ratio of individuals aged 65 and over to those aged 15–64 to quantify aging. This metric captures the proportion of the elderly population economically supported by the working-age group and serves as a key variable for assessing structural demographic pressures and productivity constraints [63].

3.2.3. Intermediary Variables

To examine the mechanisms through which population aging influences new quality productivity, this study incorporates land productivity as a mediating variable. Land productivity reflects the agricultural output generated per unit of arable land area and measures the efficiency with which land resources are utilized [64]. Population aging may lead to a reduced labor supply and decreased agricultural inputs, which in turn intensifies land use pressure and reduces land productivity, thereby indirectly inhibiting the development of new quality productive forces [65]. This paper uses the ratio of total agricultural output to arable land area as a proxy to examine the mediating role of land productivity in the pathway through which population aging affects new quality productivity.

3.2.4. Control Variables

The urban–rural income gap (ur-gap) reflects the pattern of income distribution within a region and may indirectly shape the formation and development of new quality productivity by influencing human capital accumulation, consumption capacity, and the stability of social structure [66]. This study uses the ratio of urban residents’ disposable income to that of rural residents as an indicator to measure disparities in income levels between urban and rural areas.

The level of informatization (digint) captures the strength of a region’s digital infrastructure and information services [67]. This paper employs the natural logarithm of the ratio of postal and telecommunications business volume to regional GDP as a measurement variable, reflecting the density of information services and the penetration of the digital economy. This approach is widely adopted in domestic regional economic research and effectively reflects relative differences in regional informatization development.

The level of foreign direct investment (in-fdi) serves as a key indicator of a region’s openness and its capacity to attract foreign capital. Beyond providing financial resources, foreign investment can significantly impact local production efficiency through technology spillovers, improved management practices, and competitive pressures [68]. This study adopts the natural logarithm of the total assets of foreign-invested enterprises to capture both the degree of dependence on foreign capital and the intensity of regional openness [69].

Finally, energy consumption (in-elec) reflects, to a certain extent, the level of industrialization and the structural characteristics of the regional economy. It is also closely linked to technological advancement and the efficiency of resource allocation [70]. To measure regional energy intensity, this paper uses the natural logarithm of total electricity consumption across all sectors in each province. This variable serves as a proxy for the foundational role of the energy system in supporting the development of new quality productivity.

3.2.5. Mechanism Variables

To further elucidate the mechanisms through which population aging influences new quality productivity, this paper introduces urbanization as a moderating variable. Urbanization facilitates population concentration, factor integration, and infrastructure development, thereby enhancing regional productivity and resource allocation efficiency. In the context of population aging, urbanization may help offset the adverse effects of labor shortages by optimizing spatial structures and industrial layouts. This study uses the proportion of the urban population to the total population as an indicator of urbanization level, capturing the degree of spatial concentration and the extent of transformation in socio-economic organizational forms across regions [71]. By including an interaction term between population aging and urbanization, the analysis identifies the moderating role of urbanization in the pathway from demographic structural changes to shifts in productive efficiency.

3.2.6. New Quality Productivity Calculation Method

To assess provincial-level new quality productivity in a rigorous and objective manner, this paper draws on relevant domestic and international literature, aligns the research objectives with data availability, and employs the Entropy Weight Method (EWM) to construct a multidimensional new quality productivity index. This method effectively captures the actual variation characteristics of each indicator, mitigates biases associated with subjective weighting, and ensures the objectivity and methodological rigor of the calculated results. The specific steps are as follows:

First, an indicator system is established, and the directionality of indicators is defined. All indicators are categorized as either benefit-type (gain-type) or cost-type (loss-type) based on economic theory and principles of sustainable development. To eliminate the effects of differing units and scales, and to ensure comparability across multidimensional data, this study applies range normalization to standardize all indicators. Specifically, for gain-type indicators, the following Formula (1) is used:

$$x_{ij}^{*}={\displaystyle \frac{x_{ij}-\mathrm{m}\mathrm{i}\mathrm{n}\left(x_j\right)}{\mathrm{m}\mathrm{a}\mathrm{x}\left(x_j\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left(x_j\right)}}$$

(1)

For loss-type indicators, the following Formula (2) is used to perform standardization:

$$x_{ij}^{*}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}\left(x_j\right)-x_{ij}}{\mathrm{m}\mathrm{a}\mathrm{x}\left(x_j\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left(x_j\right)}}$$

(2)

where $x_{ij}$ is the raw value of indicator $j$ for province $i$, and $x_{ij}^{*}$ is the value after normalization.

Then, calculate the information entropy weighting based on the normalized data to determine the weighting of each indicator. The steps are as follows:

Calculate the ratio value, as shown in Formula (3):

$$p_{ij}={\displaystyle \frac{x_{ij}^{*}}{{\sum }_{i=1}^n{x}_{ij}^{*}}}$$

(3)

Calculate the entropy value, as shown in Formula (4):

$$E_j=-{\displaystyle \frac{1}{\mathrm{l}\mathrm{n}n}}{\sum }_{i=1}^n{p}_{ij}ln{p}_{ij}$$

(4)

Calculate the redundancy and weight, as shown in Formulas (5) and (6):

$$d_j=1-E_j$$

(5)

$$w_j={\displaystyle \frac{d_j}{{\sum }_j{d}_j}}$$

(6)

Finally, the composite index is calculated by synthesizing and weighting the new quality productivity index, as shown in Formula (7):

$${nqp}_i={\sum }_{j=1}^m{w}_j{x}_{ij}^{*}$$

(7)

where ${nqp}_i$ denotes the composite score of new quality productivity for province $i$. The information entropy weighting method is based entirely on the distribution of information within the sample, thereby avoiding subjective interference and enhancing the objectivity and scientific validity of the results. The new quality productivity index constructed in this paper integrates multiple dimensions, including economic development, innovation-driven growth, and digital transformation. This index is fully consistent with the theoretical connotation of high-quality development and holds substantial theoretical and practical significance.

3.3. Model Settings and Methodology

This paper employs a high-dimensional panel fixed effects model to investigate the direct impact of demographic aging across Chinese provinces on new quality productivity. Compared with traditional fixed effects models, high-dimensional fixed effects approaches not only retain the ability to control for individual- and time-specific heterogeneity, but also offer greater computational efficiency [72]. The specific baseline regression model is specified as Formula (8):

$$nq{p}_{it}={\alpha }_1+{\beta }_1{aging}_{it}+{\gamma }_1{x}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(8)

where $nq{p}_{it }$ denotes the new quality productivity of province $i$ in year $t$; ${aging}_{it}$ represents the level of population aging; $x_{it}$ is the vector of control variables; ${\mu }_i$ and ${\lambda }_t$ denote province and year fixed effects, respectively; and ${\epsilon }_{it}$ is the error term capturing unobservable random disturbances.

To better elucidate the internal mechanisms through which population aging influences new quality productivity, this paper introduces land productivity as a mediating variable and constructs the following mediation effect model:

$$l{p}_{it}={\alpha }_2+{\beta }_2{aging}_{it}+{\gamma }_2·x_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(9)

The second step is to examine its transmission effect on new quality productivity based on the control of mediating variables:

$$nq{p}_{it}={\alpha }_3+{\theta }_1{aging}_{it}+{\theta }_{2}l{p}_{it}+{\gamma }_3{x}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(10)

In Equation (9), ${\beta }_2$ represents the effect of population aging on land productivity, while in Equation (10), ${\theta }_2$ reflects the impact of land productivity on new quality productivity. If both ${\beta }_2$ and ${\theta }_2$ are significant, and the coefficient ${\theta }_1$ in Equation (10) is smaller than the direct effect in the first-stage regression, this indicates that land productivity mediates the relationship between population aging and new quality productivity.

In this study, the urbanization variable ${ urban}_{it}$ is treated as a moderating variable to examine whether it influences the strength of the effect of population aging on new quality productivity. Accordingly, an interaction model is constructed as Formula (11):

$$nq{p}_{it}={\alpha }_4+{\delta }_1{aging}_{it}+{\delta }_2{urban}_{it}+{\delta }_3({aging}_{it}\times {urban}_{it})+{\gamma }_4{x}_{it}+{\mu }_i+{\lambda }_t+{\epsilon }_{it}$$

(11)

The interaction term ${ aging}_{it}\times {urban}_{it}$ captures the synergistic mechanism between aging and urbanization (Figure 1 and Figure 2). The significance and sign of ${\delta }_3$ indicate the direction and strength of the moderating effect of urbanization on the impact of aging.

4. Empirical Result Analysis

4.1. Analysis of Benchmark Regression Results

This paper employs a two-way fixed effects model at both the provincial and annual levels.

Table 3. Benchmark regression result.

Variable	(1)	(2)	(3)	(4)	(5)
aging	−0.007 *** (0.002)	−0.007 *** (0.002)	−0.007 *** (0.002)	−0.007 *** (0.002)	−0.006 *** (0.001)
ur-gap		0.026 (0.036)	0.025 (0.036)	0.030 (0.035)	0.033 (0.032)
digint			−0.007 ** (0.003)	−0.008 ** (0.003)	−0.006 * (0.004)
in-fdi				0.004 (0.003)	0.003 (0.004)
in-elec					0.051 * (0.028)
Constant	0.393 ***	0.332 ***	0.334 ***	−0.271 **	−0.119
N	360	360	360	360	360
Within R-sq.	0.120	0.123	0.124	0.128	0.151

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

illustrates the impact of population aging on new quality productivity. Model (1) presents the baseline specification, incorporating only the aging variable and year fixed effects. The results show that the coefficient of aging is −0.007 and negative at the 1% significance level, indicating that a higher degree of population aging is significantly associated with a lower level of new quality productivity. This finding is consistent with existing studies that link demographic aging to reduced labor supply and weakened innovation capacity. Model (2) introduces a control variable for the urban-rural income gap to the baseline model, and the aging coefficient remains significant at the 1% level. Models (3)–(5) progressively add controls for information technology development, foreign direct investment, and energy consumption. The coefficient for information technology is negative and statistically significant at the 5% level in some specifications, suggesting that digital infrastructure may play a role in shaping new quality productivity.

Importantly, the coefficient on aging remains consistently negative and statistically robust at the 1% level across all model specifications, with only minor variations. In addition, the within R² of the model increases from 0.120 to 0.151 as more control variables are introduced, indicating that the model’s explanatory power improves with the inclusion of relevant covariates. Overall, these regression results provide strong empirical support for Hypothesis H1: population aging has a significant and robust negative effect on the development of new quality productivity.

4.2. Robustness

To further test the robustness of the above conclusions, this paper employs five robustness checks, with the specific estimation results summarized in

Table 4. Robustness test results.

Variables	Winsor	Lagged	TimeTrend	FGLS	IV
aging	−0.006 *** (0.002)	−0.005 *** (0.001)	−0.003 ** (0.001)	−0.008 *** (0.001)	−0.007 *** (0.002)
Controls	Yes	Yes	Yes	Yes	Yes
Fixed Effects	Prov + Year	Prov + Year	Prov + YearTrend	Prov + Year	Prov + Year
Within R-sq.	0.149	0.114	0.084		0.675
Observations	360	330	360	360	360
Kleibergen-Paap rk LM					15.655
Kleibergen-Paap rk Wald F					165.485
Cragg-Donald Wald F					494.622

Note: *** and ** indicate significance at the 1% and 5% levels, respectively.

. These methods extend the main regression model by adjusting model specification, variable treatment, and sample construction. The results show that the sign and statistical significance of the population aging coefficient remain largely stable across different specifications, indicating that the conclusions are highly robust and possess strong explanatory power.

• Winsorization: Given that regional economic development and new quality productivity indicators may be influenced by unobservable shocks, such as statistical errors, policy interventions, or short-term fluctuations, extreme values in the sample could introduce bias into the regression results. To mitigate this, all continuous explanatory variables are winsorized at the 1st and 99th percentiles, and the model is re-estimated using a two-way fixed effects specification. The results show that the coefficient of population aging remains negative and significant at the 1% level.
• Lagged core explanatory variable: To address potential simultaneity bias between the dependent variable and the core explanatory variable, the model is re-estimated using a lagged value of population aging, while keeping all control variables at their contemporaneous levels. This approach helps strengthen causal interpretation. The results confirm that the lagged aging variable maintains a significant negative effect on new quality productivity, again at the 1% level.
• Alternative time fixed effects specification: While the benchmark model employs two-way fixed effects, this test replaces the year fixed effects with a national uniform linear time trend (province + year-trend). This specification captures time evolution through a linear trend, preserving degrees of freedom while minimizing the risk of misidentifying macro trends. The results show that the direction and significance of the aging and urbanization coefficients remain consistent, indicating that the findings are not sensitive to the time fixed effects specification.
• Model replacement using FGLS: The two-way fixed effects model may produce inefficient estimates when random disturbances exhibit intra-group, inter-group, or temporal correlation. Although clustered standard errors have been applied in the baseline regressions to address heteroskedasticity, autocorrelation, and cross-sectional dependence, this correction pertains only to standard errors and not to the model structure. To further enhance robustness, the model is re-estimated using feasible generalized least squares (FGLS). The results remain significant and directionally consistent.
• Endogeneity identification via instrumental variables: Considering potential endogeneity due to reverse causality or omitted variable bias, the study adopts the total dependency ratio (TDR) as an instrumental variable for population aging, following Angrist et al. The TDR, being structurally rigid and less responsive to current economic conditions, satisfies both the relevance and exogeneity conditions. The instrumental variable regression confirms that population aging has a significantly negative effect on new quality productivity at the 1% level. Additionally, both the weak instrument test and the over-identification test yield statistically significant results, confirming the validity and strength of the chosen instrument.

4.3. Mediation Analysis

To further reveal the internal mechanism through which population aging affects new quality productivity, this paper introduces land productivity as a mediating variable. A two-way fixed effects panel mediation model is constructed, and both the three-step regression procedure and the Bootstrap method are employed to systematically examine the mediating effect. When using land productivity as the dependent variable to examine the effect of aging, the regression coefficient is −0.028, statistically significant at the 1% level. This indicates that population aging significantly reduces the efficiency of intensive land use, thereby validating the first stage of the mediation pathway (the “aging → land productivity” link).

Further analysis reveals that in a joint regression model incorporating both aging and land productivity, the coefficient for land productivity is 0.059 and remains significant at the 1% level, whereas the coefficient for aging continues to be significantly negative. This confirms the second stage of the mediation mechanism and supports the existence of an indirect transmission effect.

To ensure the robustness of the mediating effect, this paper further applies the Bootstrap resampling method, conducting 1000 replications to estimate the indirect effect. As shown in

Table 5. Test results of intermediary effect.

Variable	nqp	lp	nqp
aging	−0.006 *** (0.002)	−0.028 *** (0.007)	−0.004 *** (0.001)
lp			0.059 *** (0.014)
Control variable	Yes	Yes	Yes
Fixed Effects	Prov + Year	Prov + Year	Prov + Year
Constant	−0.048	1.691 *	−0.218
Observations	360	360	360
Within R-sq.	0.151	0.259	0.197

Note: *** and * indicate significance at the 1% and 10% levels, respectively.

, the results show that the indirect effect is −0.002, significant at the 1% level, with a 95% confidence interval of [−0.0020657, −0.0006918], which does not include zero. This indicates that land productivity exhibits a significant partial mediating effect in the pathway from population aging to new quality productivity. In addition, both the direct effect and total effect are statistically significant at the 1% level.

In conclusion, population aging indirectly suppresses the development of new quality productivity by significantly weakening the efficiency and intensity of land resource allocation and utilization. This finding reveals a negative transmission mechanism driven by land factors, suggesting that in response to demographic structural shifts, greater efforts should be made to promote land system reform and the modernization of agriculture. Enhancing the efficiency of land resource use is essential for mitigating the adverse impacts of aging on the accumulation of new production factors and the advancement of technological productivity.

4.4. Moderation Analysis

To delve deeper into the complex interplay between population aging and urbanization on new quality productivity, this paper introduces an interaction term between population aging and urbanization levels to explore whether a structural buffering effect exists between the two. This framework rests on the premise that the impact of aging on productivity varies across regions and that its transmission pathways are shaped by differences in development stages, industrial structures, spatial population distributions, and infrastructure levels. In particular, within the context of China’s uneven urban-rural development, rural areas exhibit higher aging rates and weaker industrial absorption capacity, which is usually accompanied by a noticeable decline in the level of new quality productivity as the labor supply contracts. However, by promoting industrial agglomeration, the urbanization process enhances capital substitution and improves infrastructure provision. The urbanization process has the potential to alleviate the factor constraints imposed by aging, thereby generating a synergistic mitigation effect.

As shown in

Table 6. Interaction effects.

Variables	Baseline Interaction Model	Lagged Controls Model
aging	−0.013 *** (0.003)	−0.015 *** (0.003)
urban	0.487 *** (0.129)	0.475 *** (0.131)
aging × urban	0.011 ** (0.005)	0.015 *** (0.004)
Controls	Yes	Yes (lag1)
Fixed Effects	Prov + Year	Prov + Year
Within R-sq.	0.222	0.183
Observations	360	330

Note: *** and ** indicate significance at the 1% and 5% levels, respectively.

, the benchmark regression results support this hypothesis. The coefficient for the effect of population aging on new quality productivity is −0.013, statistically significant at the 1% level, suggesting a substantial adverse association. The coefficient for urbanization is 0.487, also significant at the 1% level, reflecting its positive role in enhancing new quality productivity. The interaction term coefficient is 0.011, significant at the 5% level. Although the statistical strength is moderate, the positive sign suggests that urbanization can partially offset the negative effects of aging on new quality productivity. After conducting a robustness test using lagged control variables, the interaction term coefficient increases to 0.015, with statistical significance improving to the 1% level. The direction and significance of the core explanatory variables remain consistent, further confirming the robustness of the findings.

In summary, a significant synergistic relationship exists between population aging and urbanization. Urbanization can effectively mitigate the adverse effects of population aging by optimizing spatial structures and facilitating factor agglomeration. This finding expands the analytical perspective on the economic consequences of aging and offers new insights for population governance strategies. Rather than focusing solely on demographic controls or elderly care services, future governance efforts should prioritize improving the quality of urbanization, promoting spatial restructuring of industries, and advancing integrated urban-rural development. These efforts are vital for building a resilient productivity support system that can effectively address the profound challenges posed by population aging.

4.5. Heterogeneity Analyses

4.5.1. Regional Heterogeneity Analysis

To further examine the regional heterogeneity in the synergistic effects of population aging and urbanization, this study explicitly introduces an interaction term of aging × urbanization into the regression model and conducts separate estimations for eastern and central–western regions, equivalent to an interaction-effect test. Given the substantial differences across China’s eastern, central, and western regions in terms of economic foundation, factor agglomeration, and resource allocation efficiency, the sample is divided into eastern and central–western subgroups. A group regression analysis is subsequently conducted. Leveraging their first-mover advantage, the eastern regions exhibit stronger capacities for industrial upgrading and factor reallocation, allowing them to partially offset the adverse effects of population aging. By contrast, the central and western regions, constrained by limited capital accumulation, a less diversified labor structure, and relatively weaker innovation capacity, are more vulnerable to the direct pressures of aging. In the central and western regions, the regression coefficient of population aging is –0.017 and significantly negative at the 5% level, indicating that the productivity-suppressing effect of aging is more pronounced where resource allocation and innovation activities are constrained. By contrast, in the eastern region, the coefficient is negative but not statistically significant, suggesting a weaker impact. Regarding urbanization, the coefficient in the central and western regions is 0.440 and significantly positive at the 5% level, implying that urbanization enhances new quality productivity through factor agglomeration, infrastructure improvement, and optimized population mobility. Moreover, the interaction term between aging and urbanization is 0.021 and significant at the 5% level, confirming the moderating role of urbanization in mitigating the adverse effects of aging, particularly in less developed central and western regions.

4.5.2. Industrial Structure Heterogeneity Analysis

Industrial structure not only reflects the fundamental characteristics of regional economic development but also determines its resilience and pathways in responding to the challenges of population aging. To examine this, the study employs an aging × urbanization interaction term and conducts comparative regressions between high-tech service-oriented provinces and traditional industry-oriented provinces. High-tech service regions, supported by the digital economy and modern services, possess stronger factor integration advantages but rely heavily on knowledge capital and high-skilled labor, making them more vulnerable to aging shocks. By contrast, traditional industry regions, dominated by resource-based and low-cost labor, exhibit relatively weaker short-term sensitivity to aging. Based on the ratio of tertiary to secondary industry output, the provinces are classified into two categories: high-tech service-oriented and traditional industry-oriented.

As shown in

Table 7. Heterogeneity test results.

Variables	Eastern Region	Central and Western Region	High-Tech Service	Traditional Industry
aging	−0.001 (0.008)	−0.017 ** (0.006)	−0.020 *** (0.004)	−0.001 (0.006)
urban	0.415 ** (0.173)	0.440 ** (0.192)	0.593 ** (0.203)	0.614 ** (0.285)
aging × urban	0.008 (0.010)	0.021 ** (0.008)	0.018 *** (0.005)	−0.007 (0.008)
Controls	Yes	Yes	Yes	Yes
Fixed Effects	Prov + Year	Prov + Year	Prov + Year	Prov + Year
Observations	132	228	180	180
Within R-sq.	0.402	0.172	0.236	0.3037

Note: *** and ** indicate significance at the 1% and 5% levels, respectively.

, the regression results show that in high-tech service-oriented provinces, the coefficient of aging is –0.02 and significantly negative at the 1% level, indicating a stronger suppressive effect of aging on new quality productivity. In contrast, the coefficient in traditional industry-oriented provinces is not statistically significant, suggesting a more limited impact. Regarding the interaction term, the coefficient of aging × urbanization is 0.018 and significant at the 1% level in high-tech service-oriented provinces, demonstrating that urbanization mitigates the negative effects of aging through factor agglomeration and innovation support. However, the interaction effect is not significant in traditional industry-oriented provinces, implying weaker buffering capacity.

In sum, while industrial upgrading enhances new quality productivity, it simultaneously intensifies dependence on high-skilled labor, making aging shocks more pronounced in high-tech service-oriented provinces. Although the short-term impact is weaker in traditional industry-oriented provinces, their long-term resilience remains at risk if industrial upgrading and improvements in urbanization quality are not achieved.

The revealed mechanism exhibits a degree of cross-regional transferability. In areas such as demographic transitions, the agriculture-to-services shift, and urban spatial governance, emerging economies—including those in Latin America and Southeast Asia show constraint patterns similar to those observed in China. However, the extent of extrapolation is contingent upon factors such as institutional quality, the maturity of factor markets, and housing and commuting costs in urban areas. Therefore, the findings should be interpreted with consideration of these contextual limitations.

5. Discussion

5.1. Conclusions

This paper examines the relationship between China’s aging population and the development of new quality productivity, utilizing panel data from 30 provinces between 2011 and 2022. Through mediation and moderation effect analyses, it explores the pathways and mechanisms through which population aging influences the formation of new quality productivity. The study reaches the following main conclusions:

First, Population aging is significantly negatively associated with the development of new quality productivity. Specifically, aging exerts a direct negative effect on new quality productivity, a conclusion further confirmed through a series of robustness tests. The reduction in the working-age population, decline in human capital quality, and increased fiscal burden associated with aging directly hinder technological innovation and industrial upgrading, thereby constraining the economy’s transition toward a more advanced and innovation-driven productivity structure.

Second, this study validates the mediating role of Land productivity as a transmission channel: population aging is negatively associated with land productivity, land productivity is positively associated with new quality productivity, and the indirect effect is statistically significant. Specifically, aging significantly reduces the efficiency of land resource utilization, further impeding the improvement of new quality productivity. This finding suggests that enhancing land productivity is a key channel for mitigating the adverse effects of aging and promoting the formation of new quality productivity.

Third, the improvement of urbanization levels can effectively alleviate the negative impact of population aging on new quality productivity. Urbanization supports the mitigation of challenges such as labor shortages and weak innovation momentum caused by aging through multiple mechanisms, including agglomeration effects, optimized resource allocation, and technology diffusion, particularly in the relatively underdeveloped central and western regions.

Finally, the impact of population aging on new quality productivity exhibits significant regional and industrial heterogeneity. Eastern regions have effectively mitigated the adverse effects of aging through technology-intensive industries and stronger resource adjustment capabilities. In contrast, central and western regions, due to weaker innovation capacity and lower resource allocation efficiency, are more severely affected. Additionally, different industries vary in their sensitivity to aging, with high-tech and service sectors demonstrating greater resilience compared to traditional industries.

5.2. Policy Implications

The evidence of this study demonstrates a robust negative association between population aging and the development of new quality productivity. The main transmission channels include labor supply contraction, declining innovation activities, and rising fiscal pressures. Land productivity functions as a key mediating mechanism, while urbanization exerts a conditional moderating effect, particularly in the central–western and high-tech service–oriented regions. Based on these findings, policy efforts need to adopt a coordinated approach across the dimensions of “people–land–city,” and align goals with mainstream theories and the Sustainable Development Goals (SDGs).

Positioning and Theoretical Linkages. The results are consistent with Schumpeterian innovation-driven growth theory, which highlights factor renewal and technology diffusion as determinants of long-term productivity, as well as with the extended TFP framework that emphasizes efficiency in resource allocation across sectors and regions. By incorporating digitalization, greening, and innovation into productivity measures, the study directly corresponds to SDG 8 (Decent Work and Economic Growth), SDG 9 (Industry, Innovation, and Infrastructure), and SDG 11 (Sustainable Cities and Communities). Policy frameworks should therefore establish clear outcome indicators and evaluation standards.

Integrated “People–Land–City” Policy. On the human dimension, policies such as delayed retirement and reemployment incentives can mitigate labor supply contraction, while skill training and investments in aging-friendly technologies enhance productive participation among the elderly. On the land dimension, ecological redlines should be strictly enforced alongside the promotion of smart and digital agriculture, crop structure adjustment, and targeted imports. Given varying levels of mechanization, grain crops are less affected by aging, whereas labor-intensive cash crops face stronger risks; optimizing crop structures and leveraging international trade can thus alleviate demographic pressures while maintaining ecological sustainability. On the urban dimension, affordable housing and public transport are crucial to reducing commuting burdens, while digital urban governance and differentiated regional strategies should be adopted—emphasizing high-end factor agglomeration in the east, and infrastructure and industrial digitalization in the central–west.

Firm-Level Actions. Enterprises should treat aging as a driver of transformation by promoting age diversity in recruitment and performance evaluation, adapting workplaces with aging-friendly support, and expanding investments in energy efficiency, process automation, and digital management. Financing can be accessed through green credit and transition bonds. Supply chain management should incorporate green procurement and carbon tracking, while partnerships with agricultural producers can advance smart farming. Labor structure, retraining, and resource efficiency indicators should be integrated into corporate disclosure practices, in line with SDG 8, SDG 9, and SDG 12 (Responsible Consumption and Production).

International Transferability. The mechanisms identified in this study have cross-national applicability. Emerging economies in Latin America and Southeast Asia face similar challenges of demographic aging, inefficient land use, and urban housing pressures. China’s integrated “people–land–city” framework provides valuable insights that can be adapted to local contexts through targeted interventions in housing policy, public transportation, land circulation, and smart agriculture. Among advanced economies, Japan has mitigated aging shocks through delayed retirement, the silver economy, and community-based services, while Germany and the Nordic countries have sustained productivity growth through innovation incentives and technology transfer. These experiences provide transferable references for both China and other emerging economies.

Academic Contributions. For the scholarly community, this study enriches the literature on the relationship between aging and productivity by highlighting the “population–land–city” channel of resource allocation. Unlike prior studies that focused narrowly on labor quantity, the findings reveal the mediating role of land productivity and the moderating role of urbanization. This perspective provides a new framework for international comparative research and contributes a replicable indicator system for measuring productivity.

Overall, population aging should not be regarded solely as a burden but also as a catalyst that drives economies to engage in institutional innovation, factor reallocation, and sustainable governance. Through fiscal incentives, land system reform, and improvements in urban livability, it is possible to find a balance between addressing demographic challenges and advancing new quality productivity within the triple transition of greening, digitalization, and innovation. These insights hold practical relevance not only for China but also for other countries confronting similar challenges.