Talent agglomeration and urban inclusive green growth

1 Introduction

Cities serve as the central hubs of human socioeconomic activity, contributing approximately 80% of global GDP while accounting for 70% of energy consumption and emissions (Shan et al., 2021). The imbalance between economic development and ecological conservation has become a common challenge in global urban development (Wu et al., 2025). China has achieved rapid economic growth through extensive, factor-driven development, with its contribution to global economic growth remaining at around 30% for an extended period (Wang M. et al., 2024). However, this high-speed growth has been accompanied by ecological issues such as excessive resource consumption and worsening environmental pollution (Gu et al., 2021; Xu A et al., 2025), as well as social problems including an urban-rural dichotomy, widening wealth gaps, and unequal development opportunities (Sun et al., 2020; Kebede et al., 2023). The economy exhibits significant characteristics of non-green and non-inclusive development; such issues are also present in many developed countries and have become major obstacles to the sustainable development of the global economy (Gu et al., 2021).

Against this backdrop, the United Nations' New Urban Agenda (2016) and Sustainable Development Goal 11 (SDG 11) set forth the goal of building inclusive, safe, resilient, and sustainable cities. Consequently, the concept of “Inclusive Green Growth, IGG” emerged and has become the core guiding principle for high-quality urban development worldwide. The core essence of inclusive green growth is that, while pursuing economic growth, it prioritizes equal involvement of all economic individuals in the process, sharing the fruits of economic growth, and narrowing the wealth gap. Centered on the synergy of economic, social, and environmental dimensions (World Bank, 2012), this concept aims to resolve the inherent contradictions between efficiency and equity, as well as growth and sustainability, in traditional urbanization, thereby providing a new paradigm for addressing the issues of non-green and non-inclusive urban development globally (Yang et al., 2024; Kamguia et al., 2025). Exploring the core drivers of inclusive green urban growth has become a major focus of current urban development research.

Current scholarly inquiry into IGG centers on three principal domains: conceptual foundations, measurement methods, and determinants of performance. First, regarding conceptualization, while a universally accepted definition remains elusive, the prevailing academic consensus anchors IGG in its tripartite socio-economic-ecological dimensions (Sun et al., 2020; Li et al., 2021). Second, with respect to measurement, two dominant methodological paradigms prevail: (i) the composite index approach—which constructs a multidimensional evaluation system and applies entropy-weighted method to quantify IGG (Zhou, 2022; Fan et al., 2023; Ofori et al., 2023); and (ii) the efficiency-based approach—which adopts input-output perspectives and primarily employs Data Envelopment Analysis (DEA) to assess IGG (Zhang et al., 2022; Ren et al., 2022; Fang and Wen, 2025). Third, concerning influencing factors, extant literature emphasizes policy-related drivers (Feng et al., 2024; Khan et al., 2025; Ai et al., 2025) and market-oriented mechanisms (Ma et al., 2024; Li et al., 2024; Ofori et al., 2024; De Pascale and Romagno, 2024; Kamguia et al., 2025). Notably, however, human capital—particularly talent agglomeration (Talagg)—has received scant attention in IGG research, and the interrelationship between Talagg and urban inclusive green growth (UIGG) remains empirically unexplored.

As the primary conduit for knowledge spillovers and technological innovation, the spatial concentration of talent transcends a mere aggregation of individual factors; it constitutes a dynamic process whereby human capital coalesces through synergistic interactions. This concentration furnishes dual foundational supports for UIGG: First, the innovative mindsets and advanced technologies fostered by talent concentration catalyze industrial upgrading toward low-energy-intensity, high-value-added sectors—thereby mitigating pollution, enhancing resource utilization efficiency, and alleviating resource and environmental constraints. Second, the economic transformation propelled by Talagg generates high-quality employment opportunities, advances the equitable provision of public services, narrows inter-regional and intra-urban development disparities, and broadens stakeholder participation in the benefits of sustainable development. Moreover, Talagg reconfigures the endogenous drivers of urban development. By augmenting human capital endowments and reinforcing systemic innovation capacity (Peri et al., 2015), Talagg facilitates structural transformation and sectoral diversification of urban economies—serving as a critical nexus linking economic advancement, social equity, and ecological sustainability.

The positive influence of human capital or talent factors on urban economic growth and green development is now widely acknowledged in the scholarly literature. Yet existing research largely treats talent's contributions in isolation (either as a driver of economic output or as an enabler of environmental sustainability), rarely examining how Talagg simultaneously shapes economic, ecological, and social outcomes within an integrated economy-ecology-society framework. This conceptual and empirical gap lies at the heart of our investigation. Specifically, current literature remains inconclusive regarding (i) whether Talagg fosters UIGG, (ii) through which pathways such effects operate, and (iii) under what contextual conditions these relationships hold. Moreover, conceptual fragmentation, limited theoretical depth, and insufficient empirical rigor continue to constrain our understanding of how high-end talent contributes to cities' comprehensive, high-quality development.

To bridge these gaps, this paper builds upon well-established theoretical insights concerning talent's dual role in advancing both economic productivity and ecological transition. Moving beyond unidimensional analyses, we develop an integrative analytical framework that jointly incorporates three pillars of sustainable development: economic efficiency, green low-carbon transformation, and social inclusiveness. Within this framework, we systematically investigate (1) the aggregate effect of Talagg on UIGG; (2) its heterogeneous manifestations; and (3) the underlying mechanisms. By doing so, this study overcomes the fragmentation and superficiality prevalent in prior work, enriches the literature on drivers of IGG, and offers empirically grounded policy implications for cities pursuing talent-led, high-quality, and sustainable transformation.

The marginal contributions of this paper are reflected in two points: First, it breaks through the limitations of existing research that focuses on policy and market factors, and extends the research boundaries of UIGG from a new perspective of Talagg, enriching the theoretical analysis framework of economic-social-environmental coordinated development. Second, it provides a systematic empirical investigation into the overall effect, heterogeneous patterns, and underlying mechanisms through which Talagg influences UIGG. In doing so, it elucidates the “black box” of how human capital drives urban green growth, strengthens the empirical foundation for talent-led urban sustainability, and offers actionable evidence to inform strategic talent planning and high-quality urban development both domestically and internationally.

2 Theoretical analysis and research hypothesis

2.1 Direct impact

Lucas argues that the accumulation of human capital is the source of economic growth, and that talent—as the primary agents of R&D and innovation—can drive urban green development by influencing production, consumption, investment, and savings. Porter's theory of competitive advantage also suggests that advanced factors centered on talent are more effective in promoting regional development than primary factors centered on natural resources. It is therefore evident that talent is a key driver of UIGG.

First, Talagg can boost the momentum of urban economic growth (Farndale et al., 2022). From an agglomeration perspective, Talagg and economic growth are parallel processes. Agglomeration can significantly enhance social productivity and promote regional economic growth (Davis et al., 2014; Fontagné and Santoni, 2019). Talent provides cities with abundant human capital and intellectual support, serving as the primary driver of technological advance (Squicciarini and Voigtländer, 2015). Talagg not only enhances learning effects in corporate production processes (Zhou et al., 2018; Zhang et al., 2024), optimizes management models, reduces operating costs, and improves production efficiency. But also drives technological renewal and information sharing within the region, promotes the specialized division of labor and improves the efficiency of resource allocation. Thereby fostering efficient and sustainable urban economic growth (Hsieh et al., 2019).

Second, Talagg contributes to deepening the inclusiveness of urban development. Talagg can rapidly attract and stimulate the influx of various factors of production and related enterprises into cities. While broadening the path toward diversified urban development, it accelerates the commercialization of R&D projects and the market application of innovative outcomes, effectively narrowing development disparities within cities. At the same time, the cutting-edge management concepts brought by Talagg can play a significant role across various sectors, including government, business, and social organizations. They assist the government in achieving collaborative governance, elevating decision-making standards, refining governance systems, enhancing governance efficiency, and reducing governance costs. Consequently, this significantly improves the level of urban governance (Ren Y et al., 2024), strengthens the resilience of cities to external risks and shocks, and amplifies the inclusiveness of urban development.

In addition, Talagg helps accelerate the green transition of cities. Not only does Talagg speed up improvements in preliminary production methods and spur the transformation and upgrading of traditional industries, but it also plays a key role in promoting green technologies and improving environmental quality. Furthermore, by influencing consumer preferences, investment philosophies, employment patterns, and environmental awareness, Talagg guides urban residents toward greener and low-carbon lifestyles. This, in turn, compels highly polluting and energy-intensive industries to undergo green transformation, gradually reducing the city reliance on resource-driven industries. It advances industrial restructuring toward rationalization and high-end development, cultivates an urban atmosphere conducive to green development, elevates the efficiency of the green economy, and ultimately drives the urban green transformation and development (Zhang et al., 2024). So, this paper proposes the following hypothesis 1.

• H1: Talagg can promote UIGG.

2.2 Mechanisms analysis

2.2.1 Green innovation (Greeinn)

Innovation is the driving force behind economic growth and social advancement. Talent is the foremost core factor in enhancing technological innovation capacity of region (Yang and Pan, 2020). Marshall's theory of externalities posits that agglomeration economies promote innovation mainly due to positive externalities arising from agglomeration. Since knowledge and skills transfer alongside the movement of talent (Chemmanur et al., 2018; Teslenko et al., 2021), agglomeration can strengthen interaction and learning effects among innovators within a city (Atkin et al., 2022), while simultaneously reducing the costs of acquiring technological innovations and the risks of failure (Moretti, 2021), thereby creating favorable conditions for the conduct of innovative activities. Green technological innovation as a primary engine for advancing urban sustainable development (Liu et al., 2023), plays a crucial role in reconciling the tensions between economic growth and the ecological environment, as well as in addressing global warming and natural disasters.

Talagg mainly elevates urban green innovation capabilities by increasing inputs and expanding output. On the one hand, Talagg not only meets the demand for high-quality labor inputs required for urban green development, but also drives the concentration of innovation resources—such as knowledge, technology, and capital—in cities, consequently compelling and incentivizing governments, enterprises, and other stakeholders to increase funding for green innovation R&D projects (Zheng et al., 2021; Cai et al., 2025), thus providing policy support and financial guarantees for urban green transition. On the other hand, by building tight-knit knowledge networks and specialized social circles, Talagg facilitates the efficient coordination of internal and external resources, which helps to rapidly resolve challenges in green technology R&D (Squicciarini and Voigtländer, 2015) and precisely capture green development opportunities. At the same time, increased intellectual input strengthens the matching efficiency of production factors such as capital, equipment, and land (Zhang and Guo, 2025), accelerates the R&D of new green knowledge, products, and technologies (Kerr et al., 2017; Tang et al., 2024), and effectively improves the level of low-carbon production and green living (Shan et al., 2021), ultimately driving the green transformation and development of cities. In summary, this paper considers the following hypotheses 2a.

• H2a: Talagg can impel UIGG through Greeinn.

2.2.2 Entrepreneurial activity (Startup)

Urban public service systems (such as makerspaces and startup cafes) provide a comprehensive platform for brainstorming and technical exchange among talent. These systems not only nurture numerous potential entrepreneurial opportunities, but also boost communication efficiency among entrepreneurs (Garay et al., 2017), contributing to fostering a favorable urban entrepreneurial ecosystem and enriching the capabilities of local entrepreneurs (Glaeser et al., 2015). Consequently, this enhances the appeal of cities to entrepreneurial actors, driving an increase in the number of entrepreneurs and the expansion of entrepreneurial activities in the region (Jian et al., 2024).

Firstly, Talagg can reduce the probability of adverse selection. Talagg not only provides startups with the skilled workers they need, thereby lowering recruitment and training costs, but also helps overcome technical barriers and budgetary challenges during the startup process, laying a solid foundation for the growth of these enterprises (Rong et al., 2024). Meanwhile, the high-quality social networks of this talent pool provide startups with scarce business information and channels, helping entrepreneurs quickly connect with partners, investors, and customers, thus shortening the startup cycle and reducing risks, which in turn attracts more people to engage in entrepreneurial activities (Gennaioli et al., 2013).

Secondly, entrepreneurial activities can optimize the allocation of market resources and guide the growth of green industries. While accelerating economic development, Talagg also intensifies both competition and cooperation within cities (Yang and Pan, 2020). Market entities continuously enhance their competitiveness by collaborating to share resources and bear risks jointly (Combes et al., 2012). Entrepreneurial firms, in particular, under the influence of market competition mechanisms, guide resources toward promising green industries and emerging sectors. This effectively improves resource utilization efficiency, optimizes industrial structures, reduces urban dependence on high-pollution and energy-intensive industries, propels the green transformation of traditional industries, and ultimately achieves urban green growth. In summary, we present the following hypothesis 2b.

• H2b: Talagg can promote UIGG via Startup.

2.3 Moderating effect

2.3.1 “Soft” environment: Talagg, industry agglomeration (Indagg) and UIGG

For Talagg to exert its influence on UIGG, it requires a mediating platform and a tangible vehicle. Indagg happens to fulfill this role. Indagg is a key characteristic of modern economic growth and an external manifestation of productivity agglomeration. The clustering of enterprises within the same or related industries in a specific region inevitably leads to the flow of talent toward that region. Talagg, in turn, further drives the formation of enterprise clusters (Jian et al., 2024), creating a mutually reinforcing developmental dynamic between the two.

To begin with, Indagg attracts Talagg by creating job opportunities, while simultaneously encouraging formal and informal exchanges within the industry, thereby sparking intellectual exchanges among highly qualified professionals and enabling the refinement of craftsmanship by highly skilled workers (Squicciarini and Voigtländer, 2015; Moretti, 2021). By leveraging the “local buzz” effect, it facilitates the rapid dissemination and sharing of knowledge and technology, stimulates the innovative potential of talent, refines urban social networks and innovation ecosystems, improves industry collaboration and innovation efficiency (Huggins and Thompson, 2023), and enhances development potential of cities.

Furthermore, Indagg can expand the impact of green innovation of Talagg, promoting the rapid adoption and widespread use of green technologies. Talagg plays a significant role in advancing the development of knowledge-intensive business services (KIBS). The “sticky knowledge” generated by KIBS clusters forms a dynamic match with a highly skilled workforce, accelerating the transformation of human resources into human capital. This not only enhances production efficiency but also speeds up green technological innovation (Zhang and Guo, 2025). Early adopters of new green technologies gain a first-mover advantage, attracting the clustering of industry-related talent, knowledge, and technology. This further enhances industry specialization, reduces factor search costs and risks (Wang M. et al., 2024), drives the clustering of related industries, and achieves energy conservation and efficiency gains in urban economic growth (Giuliano et al., 2019).

In addition, compared to other regions, the environmental standards applied to industrial clusters will be more stringent. According to Porter's hypothesis, strict environmental regulations can compel talent to develop clean technologies (Ouyang et al., 2020; Wang et al., 2024), thereby reinforcing the orientation toward green growth. Concurrently, the synergistic effects of industrial and talent policies can further invigorate green innovation incentives among talent, boost the enthusiasm for green innovation among highly qualified and skilled professionals (Moretti, 2021), and ultimately enhance UIGG. Thus, we put forward the following hypothesis 3a.

• H3a: Indagg has a substantial moderating effect in the process of Talagg affecting UIGG.

2.3.2 “Hard” systems: talagg, labor marketization (Labmark) and UIGG

Promoting the free flow of labor and optimizing the allocation of human capital constitute critical institutional enablers of healthy, sustainable, and high-quality regional economic development. The degree of Labmark serves as a pivotal institutional determinant of talent mobility efficiency and factor resource allocation efficiency, directly shaping the effectiveness of Talagg and regional industrial innovation (Cai et al., 2025), and underpinning the high-quality development of urban economies (Sequeira et al., 2020).

However, in certain regions of China, persistent labor market segmentation and inefficient factor allocation remain salient challenges. Inadequate progress in Labmark constitutes a fundamental driver of cross-regional labor mobility barriers and allocative frictions in labor markets. Under low marketization conditions, labor flows between urban and rural areas and across sectors face implicit institutional constraints. Moreover, labor pricing fails to emerge autonomously from supply-demand dynamics, leading to price distortions in human capital and suboptimal spatial distribution of skilled labor (Guo et al., 2024). Consequently, talent decisions regarding interregional employment are constrained, bargaining costs for scarce labor rise significantly (Beaudry et al., 2018), and enterprises—facing heightened cost pressures—tend to scale production through low-cost, routine labor rather than investing in high-caliber talent. This dynamic dampens incentives for enterprise transformation and upgrading, suppresses R&D investment in green and low-carbon technologies (Lin et al., 2021), and ultimately impedes regional green innovation and inclusive growth.

As Labmark advances, institutional barriers to cross-sectoral and interregional labor mobility have been progressively dismantled. This has effectively mitigated distortions in factor markets and alleviated human capital misallocation (Yang and Pan, 2020), thereby enhancing the allocative efficiency of labor. Talent flows are increasingly guided by market signals, yielding dual benefits: First, improved matching between job requirements and human capital—amplified by skill complementarities and human capital spillovers—which strengthens the catalytic role of Talagg in urban innovation and green production (Zhang and Guo, 2025; Cai et al., 2025). Second, the expansion of non-agricultural employment facilitates structural upgrading of industries, establishing a robust industrial foundation for inclusive and green growth.

Concurrently, labor market deepening is closely intertwined with the equalization of public services. As labor mobility reshapes spatial population distribution, market-driven resource reallocation promotes a more equitable geographic distribution of public service infrastructure. This helps mitigate welfare disparities rooted in occupational or hukou status, reduces income and benefit losses among migrant workers, and strengthens their sense of urban belonging. When equitable access to public services is ensured, talent retention improves, unlocking the full demographic dividend and supporting cities' coordinated advancement toward inclusiveness and sustainability.

In summary, advancing Labmark optimizes the spatial allocation of human capital, activates innovation spillovers from Talagg, fosters social integration of migrant workers, and thereby provides comprehensive support for UIGG.Therefore, we suggests the following hypotheses.

• H3b: Advancing Labmark can positively moderate the relationship between Talagg and UIGG.

Based on the above theoretical analysis, we draw the logic mechanism diagram (Figure 1).

3 Estimation strategy and data

3.1 Baseline model

In order to study the impact of Talagg on UIGG in Chinese, the following baseline regression model is constructed:

UIGGit=α0+α1Talaggit+α2Controlit+CityFE +YearFE+εit(1)

Equation 1 where i is the city, t is the year, and α₀ is the constant term. α₁ is the estimated coefficient we interest. UIGG_*it* is the level of IGG in city i in year t. Talagg_*it* is the level of Talagg in city i in year t. Control_*it* is a collection of control variables. CityFE is the city fixed effect, YearFE is the year fixed effect, and ε_*it* is the random disturbance term. The coefficient α₁ measures the average difference in the impact of Talagg on UIGG.

3.2 Variables

3.2.1 Explained variable: UIGG

Grounded in the theoretical foundations of sustainable development, inclusive growth, and the green economy, UIGG integrates three interdependent pillars: economically efficient growth, universally accessible social opportunities, and ecologically sustainable development. Accordingly, we conceptualize UIGG as a holistic developmental paradigm that synergistically advances ecological sustainability, high-quality economic advancement, and social equity. Its defining objective is to transcend the longstanding urban development trade-offs—namely, prioritizing economic expansion at the expense of ecological integrity, or pursuing efficiency gains while undermining distributive fairness. To this end, UIGG positions green and low-carbon transformation as its strategic orientation, quality-driven economic upgrading and productivity enhancement as its operational foundation, and universal participation and equitable sharing of developmental benefits as its ultimate aim, thereby enabling balanced, coordinated progress across all three dimensions.

Drawing upon established literature, we have developed a comprehensive, multi-dimensional evaluation index system comprising three domains: economic development, social inclusiveness, and green livability. The conceptual rationale underpinning each domain is articulated as follows:

First, the economic development dimension furnishes the material basis for UIGG implementation. Regional economic output reflects the capacity to generate wealth. Sustained expansion of economic scale strengthens local fiscal capacity, thereby ensuring stable public financing for both the equalization of essential services and integrated ecological governance. That prerequisites indispensable to achieving inclusive and green transitions.

Second, the social inclusiveness dimension embodies the people-centered ethos of UIGG. It centers on three key aspects: equitable income distribution, universal access to basic public services, and robust social protection systems. This dimension distinguishes UIGG from conventional green development models by explicitly embedding inclusivity into environmental progress, thus operationalizing UIGG core commitment to the broad-based sharing of developmental gains.

Third, the green livability dimension establishes the ecological boundary conditions necessary for UIGG long-term viability. It emphasizes the provision and equitable distribution of ecological public goods, and underscores the principle of shared ecological welfare. By capturing cities' ability to simultaneously enhance residents' quality of life and safeguard regional ecological stability, this dimension ensures coherence between environmental sustainability and spatial–social equity.

From a systems perspective, fiscal resources generated through economic growth serve as critical inputs for enhancing public service delivery and ecological restoration, thereby advancing social inclusion and fostering the development of green, livable communities. In turn, equitable access to public services strengthens societal inclusiveness and facilitates a strategic shift in economic development—from extensive, resource-intensive expansion toward intensive, efficiency-driven growth. Meanwhile, green regulatory frameworks constrain the unregulated expansion of high-energy-consuming industries, catalyzing regional industrial transformation and upgrading toward sustainability. These three interdependent dimensions mutually reinforce one another and collectively constitute the theoretical foundation of IGG, grounded in tripartite economic–social–ecological co-evolution. So, referring to Wang D et al. (2022), we construct a UIGG index system based on three dimensions: economic development, social inclusion, and green livability in Table 1, and measure it adopting the entropy method.

Table 1. UIGG indicator system.

The above data are all from EPS.

3.2.2 Explanatory variable: Talagg

In extant literature, several constructs (including human capital agglomeration, Talagg, high-skilled labor agglomeration, and innovation talent agglomeration) are employed to characterize the spatial concentration of regional human resources. However, these constructs differ fundamentally in conceptual scope and applicability. Specifically: (1) Human capital agglomeration is a macro-level construct grounded in the production-factor paradigm, representing the aggregate stock of human capital across the entire labor force—not individual workers—and primarily reflecting the region's overall labor input intensity; (2) Talagg, by contrast, adopts a categorical approach within the human capital framework. It deliberately excludes low-skill, routine labor and defines inclusion based on formal educational attainment and occupational credentials. It captures the spatial distribution and structural optimization of professional human resources across all industry sectors—extending beyond, but not limited to, science, technology, and innovation domains; (3) High-skilled labor agglomeration constitutes a subset of Talagg, focusing narrowly on practitioners possessing advanced vocational competencies aligned with real-economy production contexts, emphasizing applied proficiency rather than foundational technological breakthroughs; (4) Innovation talent agglomeration represents the most specialized and narrowly defined tier. It comprises individuals whose core functions encompass R&D, technological invention, and knowledge commercialization, serving as the principal human agents driving regional technological advancement and industrial upgrading.

Given these distinctions, this study adopts Talagg as its analytical lens. Four interrelated rationales underpin this choice: First, Talagg avoids the excessive breadth inherent in human capital concentration, thereby enhancing conceptual precision and empirical tractability. Second, it mitigates the narrow sectoral bias associated with high-skilled labor and innovation talent agglomeration, ensuring broader representativeness across industries. Third, Talagg occupies a pivotal mediating position in the hierarchical transformation from aggregate human capital to domain-specific high-end talent. Fourth, while high-skilled and innovation talents predominantly serve select advanced sectors, Talagg encompasses the professional workforce essential to all industrial ecosystems. Consequently, this perspective can not only circumvent the overgeneralization of macro indicators but also avoid the undue specialization of high-end proxies. Accordingly, consistent with established research, we selects the number of students in regular higher education institutions, which is a widely recognized conventional indicator in the academic community, as the proxy variable for Talagg.

• 3. Control Variables

We referred to existing studies and selected 9 control variables. Detail of variables is provided in Table S1.

• 4. Data Description

Combining the comparability and availability of data, and excluding cities with serious missing data, this paper finally compiles and obtains balanced panel data for 280 cities from 2003 to 2023. The sample data are mainly from the Express Professional Superior (EPS), China Entrepreneur Investment Club (CEIC) and China Stock Market & Accounting Research (CSMAR) databases. In addition, we applied the following to the data treatments: (i) All variables involving values are deflated using 2003 as the base year. (ii) All variables, excluding ratios, are logarithmically treated to reduce the problem of heteroskedasticity. (iii) baseline variables are supplemented with linear interpolation and moving average methods. The descriptive statistics of the baseline datas is provided in Table S2.

4 Result and discussion

4.1 Baseline regression

We first conducted VIF tests on the baseline variables to avoid multicollinearity issues. The results show that all VIF values do not exceed four (Table S3). Subsequently, adopting two-way fixed effects model, we proceed stepwise regression to examine the direct effect of Talagg on UIGG. Table 2 shows the results of the baseline regression.

Table 2. Baseline results.

Standard errors in parentheses, “^***, ^**, and ^* indicate passing significance tests at the 1%, 5%, and 10% levels, respectively.”

As can be seen from columns (1)–(5) of Table 2, the estimated coefficients of Talagg are all significantly positive at the 1% level, which strongly suggests that an increase in Talagg can significantly and positively contribute to UIGG. H1 has been verified. At the same time, except Property, the estimated coefficients of all control variables are significantly positive at a statistical level of at least 5%. This also indicates the validity of the variables selected in this study. These factors can significantly and positively promote UIGG.

4.2 Endogeneity

This paper uses the instrumental variable (IV) way to test the possible endogeneity among variables. To enhance the robustness of our empirical findings, we employ 2SLS estimation model and select two instrumental variables (IV) grounded in established literature: (i) the number of universities located in each city, and (ii) the historical count of jinshi scholars produced by each city during the Qing Dynasty (1644–1912).

• (1) IV1: Number of universities. On the one hand, relevance. Universities serve as primary institutions for cultivating high-caliber human capital. Their presence and scale directly shape the local capacity for higher education and long-term talent supply. Empirical evidence confirms a statistically significant association between university density and Talagg. On the other hand, exogeneity. University establishment and restructuring are subject to centralized approval by China's Ministry of Education, and the spatial distribution of universities exhibits strong historical path dependence. Consequently, municipalities cannot adjust university counts endogenously in response to contemporaneous UIGG. That satisfies the exclusion criterion.
• (2) IV2: Number of Qing Dynasty jinshi scholars. First, it meets the relevance requirement. The jinshi scholars represent a canonical measure of premodern elite human capital. Regions with higher historical jinshi output tend to exhibit deeper cultural traditions emphasizing scholarly achievement and educational investment—factors that persistently shape intergenerational educational attainment and contribute to contemporary Talagg (Chen et al., 2020). Second, it further satisfies exogeneity. The jinshi examination system was abolished over a century ago. Its outcomes reflect institutional and socioeconomic conditions entirely external to modern policy levers, such as industrial strategy, environmental regulation, and green development initiatives. As such, historical jinshi counts are plausibly uncorrelated with current UIGG determinants. Their influence on contemporary outcomes operates solely through persistent cultural and human capital transmission mechanisms—not via direct effects on resource allocation, environmental management, or inclusive growth pathways. Given its fixed cross-sectional data, referring to the common practice, we multiply it by the annual time term to generate a time-varying instrumental variable suitable for the panel model.

Table 3 shows the results of 2SLS estimation. All estimated coefficients are all significantly positive at the 1% level. From the first stage results, it can be seen that there is a significant positive correlation between IV1 and IV2 and Talagg, which is in line with the previous analysis and confirms the relevance. And the LM value of the under-identification test and the F value of the weak identification test are much larger than the empirical critical value, rejecting the original hypothesis. The second stage results show a substantial and positive correlation between Talagg and UIGG. This is consistent with the baseline results. The above analysis shows that our main conclusions still hold after the introduction of IV to mitigate the impact of potential endogeneity.

Table 3. IV estimates results.

Standard errors in parentheses, ^*p < 0.1, ^**p < 0.05, ^***p < 0.01.

4.3 Robustness

H1 is proved in the baseline test. In order to strengthen the robustness of this conclusion, we conduct a series of robustness tests as follows, namely, replacing the core variable, tailing reduction treatment, lagging the dependent variable by one period, modifying the time interval, and adding control variables. All the robustness tests were passed at the 1% statistical significance level. The details is provided in Table S4.

4.4 Heterogeneity

Cities exhibit systematic heterogeneity in location attributes, factor endowments, and industrial composition, leading to differential enabling effects of Talagg on UIGG. This paper conducts heterogeneity analysis from three directions: geographical location, city size, and city type, and the results are shown in Table 4.

• (1) Geographical Location: Coastal (riverine) cities/Inland cities

Table 4. . Heterogeneity test.

Standard errors in parentheses, ^*p < 0.1, ^**p < 0.05, ^***p < 0.01. The P value for the inter-group difference test was calculated using Fisher's Combined Probability Test (Bootstrap 1,000).

Coastal (riverine) cities possess distinct locational advantages, including superior multimodal transport connectivity and accelerated demographic agglomeration. As talent mobility broadly conforms to population mobility patterns, skilled labor tends to concentrate in such advantaged regions, potentially amplifying the impact of Talagg on UIGG in these areas. Accordingly, we sets China's coastal cities and 70 cities distributed along the Yangtze River Economic Belt as coastal (riverine) cities (the list of cities is provided in the Supplementary Material), and the remaining cities as the inland cities.

Columns (1) and (2) of Table 4 report statistically significant positive Talagg coefficients for both groups at the 1% level. The coefficient for coastal (riverine) cities (0.0012) is approximately twice that for inland cities (0.0006), and the intergroup difference is statistically significant at the 10% level (p = 0.08). This suggests that Talagg exerts a substantially stronger marginal effect on UIGG in coastal (riverine) cities.

From the perspective of open economic theory and factor agglomeration theory, coastal (riverine) cities benefit from institutional openness, port infrastructure, and integrated logistics networks—facilitating higher foreign trade intensity, deeper marketization, and more mature industrial ecosystems. Such conditions enhance knowledge spillovers from talent, thereby accelerating green technology commercialization and catalyzing structural green transitions in energy-intensive sectors. Concurrently, their export-oriented economies generate diversified employment opportunities across skill levels, promoting income convergence between urban and rural residents and strengthening social inclusiveness. In contrast, inland cities face structural constraints—including fragmented supply chains, weak upstream-downstream industrial complementarity, and elevated transaction costs—hindering the commercialization of talent-driven innovations. Consequently, institutional frictions and factor-market inefficiencies attenuate the talent dividend, yielding a comparatively muted marginal effect. Nevertheless, consistent with the ongoing integration of China's domestic market, Talagg retains a robust and statistically significant positive effect on UIGG in inland cities.

• (2) City size: small cities/large cities and megacities

During the period of 15th Five-Year Plan for National Economic and Social Development, China will continue to implement the new urbanization strategic goals and build a new pattern of coordinated urban development. The coordinated development of cities requires the realization through the free flow of population and resources. The policy support from the government and the talent volume required by regional development strategies need to be compatible. To test whether there are heterogeneities in the inclusiveness and green growth brought about by Talagg for different city size, this paper is based on the relevant urban classification standards of the Chinese government. The sample data is divided into small cities (<1 million population) and large and large cities and megacities (≥1 million population) according to the permanent resident population.

Columns (3) and (4) of Table 4 present the estimated Talagg coefficients across city-size categories. The coefficient for large cities and megacities is 0.0008, statistically significant at the 1% level. In contrast, the coefficient for small cities is negative but statistically insignificant. The test of group difference yields p = 0.019, indicating statistical significance at the 5% level. These results suggest that the effectiveness of Talagg exhibits substantial heterogeneity across city sizes.

Drawing on urban agglomeration economics and the core-periphery framework, large and megacities benefit from pronounced scale economies (high-quality public services, advanced scientific and technological infrastructure, and complete industrial clusters). Within such contexts, Talagg fosters green decarbonization and employment expansion concurrently, thereby reconciling economic growth, ecological sustainability, and social welfare enhancement. By contrast, small cities constrained by small industrial scale, a single industrial chain, and insufficient supply of high-quality public resources. These limitations hinder their capacity to attract and retain skilled talent, increase the risk of underutilized human capital, elevate the marginal cost of human capital investment relative to productivity returns, and often precipitate sustained out-migration of locally trained high-skilled workers. Consequently, small cities struggle to generate a critical mass of innovation-driven spillovers, undermining the ability of Talagg to effectively catalyze UIGG. This explains the insignificance of the estimated coefficient for this group.

• (3) City Type: Industrialized cities/Non-industrialized cities

In fact, compared with non-industrialized cities, industrialized cities face heightened challenges, including talent outflow, elevated employment pressure, and intensified environmental regulatory burdens. These structural constraints tend to undermine UIGG. Consequently, the impact of Talagg on UIGG exhibits marked heterogeneity across city types. To operationalize this distinction, we classify sample in accordance with the “Notice on the Adjustment and Transformation Plan for China's Old Industrial Bases (2013–2022)” issued by the National Development and Reform Commission. This yields a subsample of 95 industrialized cities¹ and 185 non-industrialized cities.

As reported in Columns (5) and (6) of Table 4, the estimated Talagg coefficient for non-industrialized cities is 0.0008, statistically significant at the 1% level. In contrast, the coefficient for industrialized cities is 0.0002 and statistically insignificant. The test of group difference yields p = 0.015, confirming significance at the 5% level. This provides robust empirical support for differential effects of Talagg across city type.

From the perspective of factor input structure, non-industrialized cities are predominantly anchored in modern service sectors, high-tech industries, and energy-saving and environmental protection industries. The industries belong to the knowledge-intensive production mode, and human capital is the core production input factor. In such contexts, talent-driven technological iteration simultaneously enhances industrial upgrading, reduces environmental externalities, and fosters inclusive employment, thereby exerting a pronounced and multifaceted positive effect on UIGG. By contrast, traditional industrialized cities rely heavily on capital-intensive sectors such as heavy industry and raw material processing, whose production functions exhibit strong dependence on physical capital inputs (fixed assets, mineral resources, and energy). Human capital contributes relatively little to output in these settings, and entrenched industrial structures impede rapid technological upgrading. Consequently, the investments in talent-based innovation fail to translate efficiently into improvements in green production efficiency. Therefore, the positive effect of Talagg cannot be manifested statistically.

In summary, the effect of Talagg on UIGG is heterogeneous, varying systematically with urban characteristics due to the geographical location, city size, and city type. Specifically, Talagg exerts a significantly stronger positive influence on UIGG in coastal (riverine) cities, large and megacities, and non-industrialized cities.

5 Mechanisms

The baseline results show that Talagg can provide high-quality human capital for urban sustainable development and promote UIGG. According to the theoretical analysis in section 2.2, we will launch the verification of H2a and H2b in this section.

5.1 Mediating variables

• (1) Green innovation (Greinn)

Innovation is the first driving force of development. The improvement of urban innovation capability can inevitably impel the high-quality development of the regional economy. Most scholars use patent applications to measure the level of regional technological innovation. Thus, we adopt the ratio of green patent applications to the total number of patent applications in a given year as an indicator of Greinn. The larger its value, the higher the ability of urban green innovation.

• (2) Entrepreneurial Activity (Startup)

Startup reflects the overall intensity of entrepreneurial activities in a specific region. In addition to start-up capital, entrepreneurship requires excellent professional skills, which is the characteristic of highly-skilled talents. The entrepreneurial atmosphere of a city can directly reflect its economic vitality and social inclusion. Therefore, we refer to the research method of (Xu F et al., 2025). We choose to use the quantity of newly created enterprises per 100 people in a city to measure Startup. This approach avoids to a certain extent the problem of measurement bias due to the heterogeneity of enterprise sizes in the region, and is able to portray the Startup of cities more accurately.

5.2 Model setting and results

Referring to Guo et al. (2024), we construct the following model to test the mechanisms of the role of Talagg on UIGG.

Mit=γ0+γ1Talaggit+∑Controlit+CityFE+YearFE +μit(2)

In Equation 2, M_*it* is the mechanism variable, including Greinn and Startup. The rest of the variables are defined the same as in (Equation 1).

The estimation results of mechanisms test are shown in Table 5, Column (1) and (2) are green innovations. Columns (3) and (4) are entrepreneurial activity. The regression results for both are significantly positive at the 1% statistical level. It confirms that Talagg can indeed promote green innovation technology, increase entrepreneurial activity, and thus enhance UIGG. H2a and H2b of this paper are confirmed.

Table 5. Mechanisms results.

Standard errors in parentheses, ^*p < 0.1, ^**p < 0.05, ^***p < 0.01.

6 Expanded analysis

In order to reveal more comprehensively and deeply the influence of Talagg on UIGG, referring to the existing studies, we select the two variables of Industrial Agglomeration (Indagg) and Labor Marketization (Labmark) as moderating variables. Incorporating them into the baseline model with the interaction term of Talagg to further conduct the empirical test.

6.1 Moderate variables

• (1) Indagg

Indagg measures the degree of concentration of an industry in a specific geographical area. We use the number of industrial enterprises above designated size to represent Indagg.

• (2) Labmark

This study employs the share of non-agricultural employment in total employment as a proxy for regional Labmark. This indicator captures the extent to which labor reallocates from traditional agricultural sectors to modern non-agricultural industries, thereby reflecting the degree to which labor resources are allocated through market-driven mechanisms.

6.2 Empirical regression

Referring to existing researches, the interaction terms of Talagg and Indagg and Talagg and Labmark were respectively included in Equation (1) for regression, and the moderating effects of Indagg and Labmark were tested.

The test results of the moderating effects are reported in Table 6. As shown in column (1) and (2) of Table 6, the coefficients of Talagg^*Indagg are all significantly positive at the 1% statistical level. It fully explains that the conjugated effect of Indagg and Talagg can promote UIGG. In other words, Indagg can play a positive moderating function in the process of Talagg on UIGG.

Table 6. Moderate results.

Standard errors in parentheses, ^*p < 0.1, ^**p < 0.05, ^***p < 0.01.

In columns (3) and (4) of Table 6, the coefficients of all Talagg^*Labmark are obviously positive at the 1% statistical level. This fully indicates that the policy synergy effect of household registration system reform and talent introduction can enhance UIGG. The increase of Labmark level can play a significant moderating effect in the process of Talagg on UIGG.

In summary, the synergistic effect of Indagg and Talagg, as well as the policy cooperation effect of household registration system reform and Talagg, making the urban economic development more inclusive and greenning. This finding also supports the hypotheses H3a and H3b.

7 Discussion

As outlined in the introduction, extant research on IGG has predominantly centered on conceptual clarification and measurement evaluation. In contrast, scholarly attention to the determinants of IGG remains comparatively limited. With regard to conceptual foundations and measurement frameworks, this study adheres to well-established approaches validated in prior literature. Our principal contribution lies in advancing causal inquiry into IGG. Concerning determinants, existing studies primarily examine the roles of digital economy agglomeration (Huang et al., 2023; Ren et al., 2022), innovation capacity (Jia et al., 2023), and the impacts of environmental and fiscal policy interventions (Ai et al., 2025; Mu and Zou, 2025; Xu A et al., 2025). Notably, only a small number of works have investigated endogenous drivers of IGG through the perspective of agglomeration effects (Ren et al., 2022; Lu et al., 2025). Aligning with this emerging strand, our study likewise adopts an agglomeration perspective. However, the difference is that we combine the theory of high-end human capital with agglomeration to empirically examine the causal effect of Talagg on UIGG from a new direction. Regarding transmission mechanisms, we identify technological innovation and industrial transformation as the two primary pathways through which Talagg exerts influence on UIGG. This conclusion also echoes the research of Ren Y et al. (2024) and Xu A et al. (2025).

Theoretically, this study enriches the IGG literature by deepening the conceptual linkage between human capital and sustainability-oriented urban development, thereby addressing a critical gap in the causal analysis of Talagg and UIGG relationships. Practically, it offers actionable insights for policymakers seeking to advance the United Nations Sustainable Development Goals (SDGs), particularly those related to sustainable cities and communities.

Nonetheless, several avenues for future research remain. Firstly, in terms of analytical depth, subsequent studies could further study different types of industry talents, such as innovative talent agglomeration and technical talent agglomeration. Secondly, regarding research scope, while our analysis focuses on urban units, extending the investigation to rural and peri-urban contexts would enhance the generalizability and policy relevance of findings. Thirdly, from a spatial-econometric perspective, future work may explore cross-regional spillover effects between Talagg and UIGG. These directions constitute key priorities for our ongoing and forthcoming research agenda.

8 Conclusions and implications

8.1 Main conclusions

Talent, as the first resource for economic and social growth, can provide the necessary intellectual support required for the long-term development of global cities. The agglomeration of highly skilled labor can also fuel urban economic growth, increase social diversity, and encourage the green transformation of industry. This is also the new direction outlined in the UN SDG declaration for future urban development.

Based on the panel data of 280 cities in China from 2003 to 2023, this study investigate the impact of Talagg on UIGG, integrating both theoretical analysis and empirical validation. The findings yield four principal conclusions: First, Talagg exerts a statistically significant and robust positive effect on UIGG. Second, heterogeneity analyses reveal that the positive effect of Talagg on UIGG is significantly stronger in coastal (riverine) cities, large and megacities, and non-industrialized cities. Third, mechanism tests confirm that Talagg fosters UIGG primarily through two interrelated pathways: stimulating green technological innovation and enhancing entrepreneurial activity. Fourth, extended analyses demonstrate that both horizontal business environment and vertical institutional governance critically shape urban development outcomes.

Specifically, the conjugate effect of Talagg and Indagg, as well as the policy synergy between Talagg and Labmark, can better boost UIGG.

8.2 Policy implications

Drawing on the preceding empirical findings and theoretical frameworks, this paper proposes the following key priorities to advance UIGG.

First, a high-level, integrated development framework tailored to UIGG should be established, with Talagg serving as its foundational enabler. The strategic role of Talagg in catalyzing UIGG must be systematically embedded into the city's long-term master development plan. This entails coordinating education, scientific research, and industrial development in a synergistic manner; establishing robust institutional linkages between talent ecosystems and objectives of urban green transition and inclusive advancement; and ensuring that human capital and intellectual resources are deeply integrated across all phases of urban green transformation and diversified, equitable development. Such top-down strategic anchoring lays a rigorous foundation for talent-driven, high-quality urban development aligned with overarching sustainable development goals.

Second, a differentiated, region-specific coordination strategy must be implemented. For coastal (riverine), large, and megacities, the comparative advantage conferred by Talagg should be leveraged to deepen talent-industry integration and strengthen innovation system coherence, thereby amplifying positive effects of Talagg on UIGG and sustaining its catalytic pull. For non-industrialized cities, development pathways should be calibrated to local endowments: talent resources must be strategically deployed to restructure industrial composition, upgrade development models, and pioneer context-appropriate green and inclusive growth trajectories. Cross-regional collaboration mechanisms must be strengthened to facilitate the efficient, equitable allocation and shared utilization of talent assets, thereby mitigating interurban disparities and fostering balanced national development.

Third, ensure the effective dissemination of core information and reinforce institutional mechanisms that support green technological innovation and entrepreneurship. Establish a cross-sectoral innovation collaboration platform to channel talent resources toward green technology R&D, thereby continuously enhancing green innovation capacity. In this context, innovation serves as a critical intermediary in empowering Talagg impact UIGG. Concurrently, foster an inclusive institutional environment for development to safeguard the orderly advancement of entrepreneurial activities and harness entrepreneurial dynamism to fully realize talent value.

Fourth, advance institutional design for multi-factor integration and maximize the synergistic impact of coordinated policy instruments. Systematically align talent attraction frameworks, industrial development strategies, and labor market institutions to enable dual reinforcement. Optimize talent resource allocation efficiency through structural enhancements within the industrial system. Moreover, strengthen supporting labor market institutions, deepen inter-system coordination, and amplify policy coherence to sustainably advance UIGG via holistic, system-based institutional development.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frsc.2026.1869544/full#supplementary-material

Supplementary material Supplementary material at the source

Supplementary material Supplementary material at the source

Footnotes

Edited by: Ziad Mouhanna, Arab International University, Syria

Reviewed by: Yang Chen, Southwest Jiaotong University, China

Ran Zhang, Qingdao University, China

Footnotes

https://www.gov.cn/gongbao/content/2013/content_2441018.htm