Effect of Spatial Flow and Optimal Combination of New Quality Productivity Forces on High-Quality Economic Development of Coastal Regions: Evidence from China 53 Coastal Cities

Abstract

This study examines the impact of the spatial flow of new quality productive forces (NQPFS) and the optimal combination of new quality productive forces (NQPFC) on the high-quality economic development (HQMED) of China’s coastal regions. Based on panel data from 53 coastal cities (2004–2023), the research constructs comprehensive evaluation systems and employs a two-way fixed effects model for empirical analysis. The main findings are as follows: First, Spatial Evolution: The HQMED level of coastal areas shows a continuous upward trend with marked regional disparities, forming a spatial pattern of “one core, two wings” characterized by “Eastern leadership with Northern and Southern regions following.” The inter-city development gap has widened, with the overall spatial structure evolving from a “core-periphery” model toward a clustered stage of “one core, multiple poles, and networked linkage.” Correspondingly, New Quality Productive Forces have transitioned from initial single-point agglomeration to a multi-polar and ultimately networked distribution. Second, both the spatial flow and optimal combination of New Quality Productive Forces exert stable positive effects on coastal HQMED. The marginal contribution of the factor optimal combination is significantly greater than that of spatial flow. Third, two complete mediation pathways are identified: NQPFS promotes HQMED primarily by enhancing the resilience of the marine industrial chain, while NQPFC drives HQMED mainly through cultivating new-quality marine business forms. Fourth, resource misallocation exerts a significant negative moderating effect on the relationship between NQPFS and HQMED. Conversely, a sound innovation ecosystem positively moderates the impact of NQPFC on HQMED. Fifth, the effects exhibit significant regional and institutional variation. Geographically, the impact follows a pattern of “strong in the East, suppressed in the North, and insignificant in the South.” Administratively, core cities demonstrate stronger factor capture and configuration efficiency compared to ordinary cities. The study confirms that facilitating the cross-regional flow and efficient internal recombination of the New Quality Productive Force is crucial for driving coastal HQMED. Policy should focus on reducing resource misallocation to remove barriers to factor mobility, optimizing regional innovation ecosystems to enhance factor synergy, and implementing differentiated strategies that balance the radiating role of core cities with the distinctive development of ordinary cities, thereby fostering a new, coordinated pattern of high-quality development across coastal regions.

1. Introduction

The global economy stands at the dawn of a new long-term development cycle, characterized by digitalization, greening, and intelligentization, which signifies a fundamental restructuring of the global productivity system [1]. International competition has shifted its core from scale expansion to the mastery and strategic recombination of key elements underpinning new quality productive forces. Major developed economies are actively guiding the agglomeration of high-end production factors through targeted national strategies to build systemic future advantages [2]. These efforts commonly aim to accelerate the flow and optimal combination of emerging factors—such as knowledge, technology, and data—to overcome the diminishing returns of traditional inputs, foster industrial ecosystems with increasing returns, and secure strategic leadership in the new wave of economic transformation and high-quality development [3]. This reflects a broader strategic transition wherein industrial productivity is increasingly driven by the dynamic combination of factors. The capacity to dismantle institutional and structural barriers to factor mobility and to channel high-quality resources toward advanced productive forces has become crucial for sustaining national development momentum [4]. With a population exceeding 1.4 billion people, China has the world’s largest middle-income group and a mega-sized market. While its domestic demand potential is immense, challenges persist, including underdeveloped factor markets; institutional barriers hindering the flow of production factors across industries, sectors, and regions; and a limited capacity for the agglomeration and synergistic allocation of existing factors toward advanced productive forces [5].

As China’s coastal regions enter a more advanced stage of development after four decades of rapid growth, the demand for higher-quality development has become increasingly urgent. In response, a high-quality economic development model driven by new quality productive forces has been formally elevated to the level of national strategy [6]. With both maritime and terrestrial attributes, China’s current development strategy emphasizes its coastal regions. At the Sixth Meeting of the Central Financial and Economic Affairs Commission in 2025, President Xi noted that “advancing Chinese modernization necessitates promoting the high-quality development of the marine economy, forging a path of strengthening the country through the oceans with Chinese characteristics.” In this regard, driving high-quality economic development requires developing new, high-quality productive forces. This includes enhancing the efficiency of factor allocation, facilitating the flow of factors toward new-quality productive forces, and reconfiguring existing production factors across different sectors [7].

However, although the development of new quality productive forces has become a strategic national priority, the underlying mechanisms of how their spatial flow and optimal combination drive high-quality development in coastal regions remain insufficiently explored. To address this research gap, this paper delves into the mechanisms through which the spatial flow and optimal combination of new quality productive forces influence the high-quality development of China’s coastal economy. Based on panel data from 53 coastal cities spanning 2004–2023, the study constructs a comprehensive evaluation system for the dynamic evolution of new quality productive forces and regional economic quality. Following an initial exploration of the spatio-temporal dynamics across the three major marine economic circles (According to the classification in the China Marine Statistical Yearbook, the Northern Marine Economic Circle includes Liaoning, Hebei, Tianjin, and Shandong; the Eastern Marine Economic Circle includes Jiangsu, Shanghai, and Zhejiang; and the Southern Marine Economic Circle includes Fujian, Guangdong, Guangxi, and Hainan.), a two-way fixed effects model is employed to verify the significant positive driving effects of both factors on high-quality development, while heterogeneity analysis is also considered. Further mechanism analysis reveals two complete mediation pathways—“enhancing marine industrial chain resilience” and “fostering the emergence of new quality business forms in the marine sector”—confirming the logic through which new quality productive forces promote coastal economic development via the upgrading of meso-level enterprise organizational forms, as well as the moderating effects of resource misallocation and the innovation environment. Theoretically, this study enriches the research framework on new quality productive forces; practically, it provides policy-making references for coastal regions to achieve a context-appropriate, synergistically driven sustainable transformation of the marine economy.

2. Literature Review

Academic inquiry into how production factors combine and flow to stimulate growth has evolved from static, micro-level analyses toward a dynamic, multidimensional framework that accounts for global systemic interactions. Alfred Marshall laid the neoclassical foundation by conceptualizing factor combination within a static equilibrium and introducing principles such as substitution and marginal productivity [8]. Joseph Schumpeter transformed this perspective by emphasizing dynamism, positing that economic development arises from entrepreneurial “new combinations”—a process of creative destruction that disrupts equilibrium [9]. Robert Solow’s growth accounting framework identified technology as the key residual driver of output growth, quantitatively establishing technological progress as the primary force beyond capital and labor accumulation [10]. Paul Romer further endogenized technological “ideas” as a core, non-rivalrous factor shaped by intentional R&D, formalizing innovation within mainstream growth theory [11]. Paul Krugman highlighted the role of spatial agglomeration of capital and labor in driving regional development [12]. Joel Mokyr underscored the importance of conducive institutional and cultural environments in enabling factor optimal combination [13]. Finally, William Nordhaus integrated planetary biophysical systems into the analysis of factor-driven growth, incorporating environmental constraints into economic modeling [14]. This paper, through a systematic review of the theories of production factors, finds that the theoretical understanding of productive forces has evolved from the classical school’s “dichotomy” of labor and land into the neoclassical “three-factor” theory encompassing land, labor, and capital. Subsequently, the theory of productive forces has further expanded to include technology, entrepreneurial capability, and resource-environmental constraints as key elements. Among these, this paper contends that the perspective of Marx and Engels who identified the simple elements of the labor process as purposeful activity or labor itself, along with the means of labor (primarily tools) and the objects of labor—is more aligned with the current development of productive forces in the Chinese context. In other words, productive forces represent the productive power of human labor, reflecting the efficiency of human labor in producing use value. Therefore, this paper adopts the “tripartite framework” of production factors to examine the scientific essence of various categories of production factors in relation to new quality productive forces.

There is growing research on the spatial flow and optimal combination of New Quality Productive Forces in relation to coastal economic development. First, there has been a focus on defining the concepts of NQPFS and NQPFC. Developing advanced productive forces requires improving factor allocation efficiency, facilitating the flow of factors toward New Quality Productive Forces, and recombining existing production factors across sectors [15]. Zhang et al. divided New Quality Productive Forces into three dimensions: elements, structure, and function [16]. At the elemental level, New Quality Productive Forces is defined by the qualitative transformation of the elements’ intrinsic quality and their endogenous development momentum. At the structural level, New Quality Productive Forces manifest as the innovative allocation of production factors and the optimization of the spatial distribution of productive forces. In terms of methodological research, on the one hand, from the perspective of input-output analysis, existing studies frequently employ extended input-output models. Chen et al. constructed an evaluation index system for New-Quality Productive Forces encompassing input quality, production vitality, and output efficiency, subsequently analyzing NQPF levels across 30 Chinese provinces using methods such as kernel density estimation and the Dagum Gini coefficient [17]. On the other hand, from the perspective of factor mobility, scholars utilize migration indices and gravity center analysis to characterize the dynamic shifts in talent, capital, and green technology across regions, thereby capturing the spatial evolution patterns of new-quality factors as they flow from low-gradient to high-gradient areas [18]. Furthermore, from the perspective of network connectivity, network flow analysis and social network analysis have been introduced to depict the topological structure within productivity systems. By calculating node centrality and flow density, researchers can identify key hubs and spillover pathways that drive the collaborative development of new-quality productive forces [19]. Second, studies of New Quality Productive Forces in various coastal regions can provide context. Research in Portugal, for instance, has highlighted spatial differences in socioeconomic and marine environmental characteristics in its coastal areas. Its northern regions and parts of the Algarve coast experienced faster socioeconomic development but also faced higher resource demands, exerting greater pressure on marine and coastal ecosystems [20]. By contrast, New Quality Productive Forces in China’s coastal areas are evolving from agglomerated "cluster-type" distributions toward more scattered “multipoint” patterns. The spatial flow of marine New Quality Productive Forces elements has a more significant effect on China’s HQMED, with notable regional heterogeneity [21].

3. Theoretical Framework and Hypothesis Development

3.1. Conceptual Foundations: Deconstructing New Quality Productive Forces

Core Components: New-Quality Labor, Labor Materials, and Labor Objects. The concept of New Quality Productive Force can be traced back to Marx’s theory of productive forces. Marx extensively employed the term “productive forces” and extended it into a series of related concepts such as “labor productivity,” “natural productive forces,” and “social productive forces.” In September 2023, during an inspection tour in Heilongjiang Province, President Xi of China originally proposed the concept of “New Quality Productive Force.” It is fundamentally defined by the qualitative leap in laborers, means of labor, objects of labor, and their dynamic optimization and recombination. This concept is characterized by high technology, high efficiency, and high quality.

Spatial Flow vs. Optimal Combination. Based on a theoretical reconstruction of Marx’s texts, Gerald Cohen argued that “space” should be incorporated into the category of means of production, regarding it as a fundamental factor of production alongside instruments of labor and objects of labor [22]. Accordingly, space also constitutes an intrinsic element of productive forces. This extension of Marx’s theory of productive forces finds textual support. In Capital, Volume III, it is explicitly stated: “Space is an element of all production and all human activity.” Therefore, a broad understanding of the connotation of productive forces should duly consider the factor of physical space [23]. Specifically, this study draws on Arthur’s [24] discourse on the evolution of technological clusters since the 21st century to distinguish between two evolutionary modes of new-quality productive forces:

(1)

Spatial flow of New Quality Productive Forces (NQPFS): China’s vast territory and ultra-large market endow it with inherent economies of scale and international competitive advantages. However, market fragmentation and resource misallocation have hindered productivity progress. This paper defines NQPFS as the “re-territorialization” of new quality productive forces, aiming to resolve contradictions in factor combinations through the reallocation of innovative elements. By optimizing the relative proportions and coupling methods of different productive factors, it enhances factor allocation efficiency and generates “Coordinated Agglomeration Effects” for both factor-inflow regions (core pioneering areas) and factor-outflow regions (peripheral late-developing areas).

(2)

Optimal combination of New Quality Productive Forces (NQPFC): This paper defines NQPFC as a dual process encompassing the “internal replacement” of old quality productive factors through iterative upgrading, and the “structural deepening” of the productive forces system driven by the emergence of new quality productive factors. From the perspective of “internal replacement”, the development of new quality productive forces requires the optimized substitution of production factors—including laborers, means of labor, and objects of labor: First, laborers transition from low-skilled to high-skilled; Second, means of labor evolve from the legacy of “machine-industry heritage” since the Industrial Revolution to modern high-precision and advanced equipment; Third, objects of labor shift from natural materials such as petroleum and cotton to clean energy and synthetic materials. From the perspective of “structural deepening”, the development of new quality productive forces urgently requires next-generation digital-intelligent technologies, such as AI large models, cloud computing, and blockchain to be embedded throughout the entire material production process, thereby expanding the boundaries of production modes.

3.2. The Direct Impact: Mechanisms Linking New Quality Productive Forces Dynamics to Coastal High-Quality Economic Development

Knowledge Spillovers and Agglomeration Economies from Spatial Flow. First, the agglomeration and flow of high-quality marine talent form an important foundation for the marine innovation ecosystem. The spatial agglomeration of high-end talent in coastal innovation hubs not only creates a specialized human capital pool but also enhances regional innovation capability through continuous knowledge spillovers. This talent agglomeration phenomenon follows Krugman’s [12] spatial-economic laws, forming an innovation field in specific geographical spaces. The embeddedness of urban innovation networks significantly expands the breadth and depth of information acquisition by broadening the scope of enterprises ’technological cooperation. The cross-organizational flow of scientific and technological talent through formal and informal channels constructs a multidimensional knowledge-dissemination pathway. This talent-flow mechanism is particularly evident in the field of marine technological innovation. Second, the synergistic coexistence of intelligent tangible labor materials and digitalized intangible labor materials constitutes the intelligent marine technology paradigm. This paradigm enhances the ability to perceive, analyze, and respond to the marine environment. The generation mechanism of positive externalities stems from the nonrivalry and knowledge spillovers of new-quality labor materials [25]. In the marine economy sector, technological dividends rapidly penetrate industrial chain linkages through demonstration and imitation [26]. Third, according to modularity theory in industrial organization economics, innovations in new-quality labor objects drive the evolution of industrial forms. The dual structure of sustainable resource utilization and integration with the digital economy constitutes the material foundation for coastal high-quality Economic Development. This transformation overcomes the resource constraints of the traditional marine economy, opening up new frontiers [27]. At the level of spatial economics, the theory of diversified externalities explains the spatial agglomeration of marine-related enterprises along coastal economic belts. This agglomeration generates significant economies of scale and scope.

Structural Upgrading and Efficiency Gains from Optimal Combination. From an industrial organization perspective, modular decomposition lowers the innovation threshold for individual enterprises. The emergence of new organizational forms, such as marine industry internet platforms, significantly reduces transaction costs between various links of the industrial chain [28]. The development of the platform economy has created new value-creation models, forming powerful network effects [29]. In terms of the innovation ecosystem, the popularity of new-quality labor materials promotes the diversification of innovation entities. New channels such as online platforms and open-source communities accelerate technology diffusion. The optimal combination of learning by doing, learning by using, and learning by interacting improves the learning efficiency of the entire marine economy [30].

Therefore, this paper proposes

Hypothesis 1. 

The NQPFS and NQPFC have a positive direct effect on HQMED.

3.3. The Indirect Pathway: Mediation Model

Enhance the Resilience of the Marine Industrial Chain. Marine Industrial Chain Resilience refers to the systemic capacity of the marine industrial network to resist, adapt to, recover from, and even transform and upgrade in response to external shocks—such as geopolitical conflicts, natural disasters, technological blockades—as well as internal disruptions. The domestic cycle serves as the primary domain for the NQPFS. Its core function lies in optimizing the allocation of advanced factors across national territory and industrial systems, thereby stimulating endogenous momentum. The domestic flow of new-quality production factors strengthens the foundation of industrial chain resilience by optimizing spatial and industrial allocation. According to New Economic Geography, the agglomeration of advanced factors in coastal innovation hubs generates scale effects. This “agglomeration-diffusion” mechanism further promotes regional synergy and enhances the internal balance and substitutability of industrial chains. Cross-sector factor flows drive “learning-by-doing” upgrades and the integration of heterogeneous knowledge [30]. Cognitive and organizational proximity help overcome technological silos, while the guiding role of venture capital reinforces vertical integration and autonomous controllability of industrial chains [31]. The directional inflow of international advanced factors expands the external resources for industrial chain resilience by embedding into global innovation networks. Its core value is to break through domestic innovation boundaries and shape new international competitive advantages through selective and strategic absorption of global advanced factors. Introducing top talent, cutting-edge technologies, and high-quality foreign investment can fill domestic gaps and promote industrial upgrading. Furthermore, engaging in international rule-making and standard alignment enhances the institutional voice and systemic resilience of industrial chains amid global fluctuations, facilitating their transformation from rule-takers to co-creators of international norms [32]. The theoretical framework of marine industrial chain resilience is constituted by three interrelated dimensions: structural resilience [33], governance resilience, and evolutionary resilience [34]. These three dimensions collectively provide robust theoretical support for the marine economy to achieve safe, efficient, and sustainable high-quality development, addressing the physical network, organizational coordination, and dynamic evolution aspects of the industrial chain, respectively. Specifically, the structural resilience of the marine industrial chain establishes a protective barrier for the stable development of the marine economy; the governance resilience of the marine industrial chain enhances the innovative efficiency of the marine economy; and the evolutionary resilience of the marine industrial chain injects sustainable competitiveness into the marine economy. Together, these three dimensions form an integrated trinity framework that ensures the steady progress of the coastal economy in China.

Nurturing New-Quality Marine Enterprises. The core of new business formats lies in industrial transformation, representing a profound restructuring of industrial organization [35]. New-quality marine formats are rooted in the marine sector and emerge from the convergence of new technologies, new production factors, and new demands, giving rise to entirely novel forms of market organization. As advanced forms of productivity driven primarily by technological innovation, NQPFCs play a decisive role in shaping marine enterprises. Through the optimal combination and spatial transition of high-quality laborers, high-tech labor materials, and a broad range of labor objects, a New Quality Productive Force instills the core characteristics of innovation-driven development, integration and coordination, green development, and open integration in marine enterprises. This process is not merely a change in the geographical location of factors but a systematic project that triggers profound changes in the technological paradigms, organizational forms, and value-creation models of marine enterprises. According to Schumpeter, during the Second Industrial Revolution, innovation was an internal factor of economic development—namely, the optimal combination of production factors [9]. Cooke later defined a regional innovation system as a system comprising geographically interconnected and specialized enterprises, research institutions, and higher education institutions. Such a system is typically driven by interactions between innovation subjects, the innovation environment, and actors, who promote the generation, flow, renewal, and transformation of new technologies and knowledge in the region [36,37]. Breakthrough innovation drives the emergence of a New Quality Productive Force. Notably, enterprises use management innovation strategies and collaborate with the government to build governance frameworks that coordinate multilateral interests, matching consumers’ green consumption demands and deepening green identity, thus promoting corporate sustainability [38]. Furthermore, manufacturing enterprises accelerate the green upgrading of their production technologies, thereby enhancing their green competitiveness, gaining competitive advantages, and improving overall business performance [39]. The emergence of new-quality business forms in the marine sector can be explained through a dynamic chain comprising two typical organizational forms, which contribute to high-quality economic development in coastal regions. First, newly incubated enterprises, equipped with new technologies, products, and organizational structures, enter the market and continuously challenge and replace outdated economic structures. Their growth enhances the dynamic efficiency of the marine economy [40]. Second, listed companies with investment and financing qualifications inherently drive the reallocation of resources toward more efficient sectors as they develop. Their expansion elevates the total factor productivity of the marine economy [41].

Therefore, this paper proposes

Hypothesis 2. 

NQPFS boosts high-quality economic development of coastal regions via strengthening marine industrial chain resilience.

Hypothesis 3. 

NQPFC boosts high-quality economic development of coastal regions via cultivating new-quality marine enterprises.

3.4. The Transmission Pathway: A Moderation Model

Resource Misallocation. This refers to a systemic allocation distortion caused by institutional distortions, market fragmentation, and policy interventions, primarily encompassing capital misallocation and labor misallocation. First, Capital Misallocation. Its core manifestation is that substantial financial resources fail to flow to the most innovative and high-growth-potential enterprises or industries, but are instead captured by inefficient sectors or “zombie firms.” When capital cannot move freely between enterprises, the marginal output of capital in high-efficiency firms far exceeds that in low-efficiency ones. However, the latter may receive more credit resources due to ownership discrimination, implicit guarantees, or policy preferences, leading to a loss in aggregate total factor productivity (TFP). Second, Labor Misallocation. Labor misallocation refers not only to the imbalanced spatial distribution of human resources in quantitative terms but, more critically, to the failure of high-quality labor to be allocated to positions and regions where its value can be most fully realized. This form of misallocation fundamentally undermines the knowledge creation and diffusion capacities upon which New Quality Productivity relies. Labor misallocation may stem from multiple factors, including the household registration system, disparities in social security, barriers to industry entry, and information asymmetries. Research indicates that the degree of labor misallocation and its negative impact on overall resource allocation efficiency may, at times, even exceed that of capital misallocation.

Innovation Ecosystem. It emphasizes that innovation activities, much like natural ecosystems, constitute a complex system characterized by dynamic interaction and co-evolution among multiple agents, resources, and the environment. James Moore first introduced the concept of the “business ecosystem,” which was later extended to the field of innovation. First, a sound innovation ecosystem can strengthen intellectual property protection. Second, a superior innovation ecosystem can improve risk capital and financial support mechanisms. Third, a sustained innovation environment can enhance the efficiency of networked synergy among factors. All three aspects positively contribute to the high-quality development of China’s marine economy by facilitating the optimized combination of New Quality Productivity.

Therefore, this paper proposes

Hypothesis 4. 

Resource Misallocation exerts a negative moderating effect on the relationship between NQPFS and high-quality development of the coastal economy, with labor misallocation having a stronger negative impact; the Innovation Ecosystem exerts a positive moderating effect on the relationship between NQPFC and high-quality development of the coastal economy.

3.5. The Contingent Effects: Expected Heterogeneities

Geographic Heterogeneity. “Growth pole” theory in economic geography suggests that growth does not occur uniformly but first concentrates in “poles” with locational advantages or policy advantages. It then influences the surrounding areas through diffusion effects. In China, Guangzhou, Shenzhen, and Zhuhai, leveraging their export-oriented economies and flexible governance, have built agile, market-oriented knowledge systems in the fields of electronic information, modern services, and emerging marine industries. Fundamental differences in knowledge structures determine that the initial conditions, absorption capacity, and optimal combination efficiency for optimizing and combining New Quality Productive Forces vary drastically across regions [42]. Shanghai, Hangzhou, and Ningbo, relying on the Golden Waterway of the Yangtze River and their status as Asia-Pacific transportation hubs, have led to the market-based configuration of high-end factors naturally tending to converge in the eastern marine economic region, where expected returns are higher, thus forming a “Matthew effect “in the spatial structure. The matching degree and co-evolution capability of New Quality Productive Force elements are steadily improving. By contrast, Dalian, Tianjin, and Qingdao rely more on the heavy, capital-intensive knowledge pool formed by equipment manufacturing, port shipping, and heavy chemical industries. Their capacity to aggregate New Quality Productive Force elements is relatively weak, and the strength of their factor optimal combination needs further improvement.

Administrative Heterogeneity. Resource allocation in China is highly dependent on the administrative hierarchy. Shanghai, Shenzhen, Guangzhou, and Hangzhou, as municipalities directly under the central government, subprovincial cities, or provincial capitals, are the main carriers of national strategies such as pilot free trade zones and technological innovation centers. They are also the first implementers of reform policies and possess a strong capacity to absorb advanced productive forces. North’s institutional change theory suggests that institutional frameworks, by shaping the incentive structure, determine the performance of economic activities and the efficiency of production factor allocation. New economic geography further provides a spatial dimension explanation for this heterogeneity. Its core-periphery model suggests that, in the presence of increasing returns to scale and transportation costs, production factors tend to agglomerate toward the “core” with initial advantages, forming a self-reinforcing cycle.

Based on the above theoretical analysis, the cross-regional flow of New Quality Productive Force mainly moves toward institutional “lowlands” and market “highlands.” Therefore, the difference in the absorption capacity of New Quality Productive Forces between core coastal cities (municipalities directly under the central government, sub-provincial cities, provincial capitals) and general cities in China is an inevitable result of the combined effects of institutional gradients and spatial patterns.

Therefore, this paper proposes

Hypothesis 5. 

The impact of NQPFS and NQPFC is stronger in the Southern or Eastern Marine Economic Circle and in higher-tier (core) cities.

Based on the five major hypotheses proposed in this paper, combined with the fundamental concepts and the overall research framework, this study constructs the logical mechanism diagram as shown in Figure 1.

4. Research Design and Data

4.1. Study Area and Data Sources

According to the China Marine Development Report (2025), the contribution rate of the national gross ocean product to the gross regional product of coastal areas has consistently ranged between 14.6% and 17.5% throughout the past 24-year period. This stable and substantial contribution highlights the critical importance of the marine sector to coastal economic growth. Considering data availability and the handling of anomalies, 53 coastal cities are selected for analysis. Panel data and indicator system hare sourced from the following annual publications: China Marine Statistical Yearbook, China Fishery Statistical Yearbook, China Industrial Ship Statistical Yearbook, China Marine Economy Statistical Yearbook, China Statistical Yearbook, China Tourism Statistical Yearbook, China Environmental Statistical Yearbook, China Port Statistical Yearbook, DMSP/OLS nighttime light data of China’s coastal cities, the Statistical Bulletin of China’s Outward Foreign Direct Investment, and statistical yearbooks of coastal provinces and municipalities. We also refer to the National Intellectual Property Administration, the National Bureau of Statistics, and firms’ annual financial reports.

4.2. Variable Construction and Measurement

Dependent Variable: High-quality Economic Development of Coastal Regions (HQMED) Index. HQMED reflects the comprehensive measurement of the growth of the marine economic sector in the national economy. Its evaluation indicators must capture the five development concepts of innovation, coordination, greening, openness, and sharing, as well as their inherent logic. We synthesize the evaluation of development performance in coastal cities and, based on data accessibility, form a comprehensive indicator set. This set includes 5 criteria layers, 10 factor layers, and 24 specific indicators, as shown in

Table 1. Evaluation Index System for the HQMED of China’s Coastal Region.

Primary Indicator	Secondary Indicator	Specific Indicator	EM W.
Innovative City Dimension	Innovation Capability	Number of Marine AI Enterprises (unit)	0.1197
		R&D Expenditure of Marine Research Institutions (10,000 yuan)	0.0660
	Innovation Performance	Number of AI Patents Held by Listed Companies (unit)	0.1299
Coordinated City Dimension	Regional Coordination	Per Capita Gross Ocean Product (100 million yuan/10,000 people)	0.0250
		Total Retail Sales of Consumer Goods (10,000 yuan)	0.0428
		Land–Sea Economic Linkage (%)	0.0125
	Industrial Upgrading	Proportion of Marine Tertiary Industry (%)	0.0013
		Proportion of Marine Secondary Industry (%)	0.0034
		Proportion of Marine Primary Industry (%)	0.0029
Green City Dimension	Resource Endowment	Forest Coverage Rate of Coastal Cities (%)	0.0078
		Built-up Area of Coastal Cities (sq. km)	0.0423
		Marine Ranch Area (hectares)	0.1494
	Environmental Performance	Industrial Particulate Emissions per Unit of Marine-related Output (tons/100 million yuan)	0.0001
		Sulfur Dioxide Emissions per Unit of Marine-related Output (tons/100 million yuan)	0.0011
		Total Industrial Wastewater Discharge per Unit of Marine-related Output (tons/100 million yuan)	0.0001
Open City Dimension	International Trade	Total Imports and Exports (10,000 USD)	0.0872
		Level of Outward Foreign Direct Investment (10,000 USD)	0.1032
	International Cooperation	Foreign Trade Dependence of Coastal Cities (%)	0.0279
		Foreign Exchange Income from International Tourism (10,000 USD)	0.1145
Inclusive City Dimension	Development Inclusiveness	Per Capita Disposable Income (yuan/person)	0.0147
		Level of Coastal Infrastructure (km)	0.0144
		Share of Food Expenditure in Total Consumption Expenditure (%)	0.0018
	Service Inclusiveness	Per Capita Public Knowledge Dissemination in Cities (volumes/person)	0.0269
		Per Capita Urban Green Space Area (sq. meters/person)	0.0049

. To determine the weights of the indicators, an objective weighting method is employed. Given the long-term stability and objective nature of historical data on marine economic development in China’s coastal regions, the entropy weight method is adopted for measurement. The specific procedures are as follows: first, the raw data undergoes dimensionless normalization and standardization. Subsequently, the information entropy and divergence coefficient for each indicator are calculated based on the processed data. Finally, the objective weight of each indicator is determined according to its divergence coefficient. It is worth noting that the entropy weights of the Environmental Performance indicators under the Green City Dimension are relatively small, reflecting the limited degree of variation among coastal cities or across the temporal dimension during 2004–2023. This indirectly indicates that, compared with other specific indicators included in the evaluation, these indicators are more reflective of fundamental, supportive, and stable characteristics rather than growth-related disparities. Nevertheless, high-quality development cannot be achieved without the critical component of green sustainability. Thus, these indicators are indispensable in the evaluation process and are therefore retained.

Core Independent Variables: NQPFS. To assess the level of NQPFS, we consider the realities of both the inflow of internationally advanced productive factors and the endowment of domestically advanced productive factors. Accordingly, we construct three primary indicators: laborers, labor materials, and labor objects. These are further divided into six secondary indicators: human capital accumulation, high-level talent, tangible labor materials, intangible labor materials (technological innovation), resource and environmental input, and digital economy development. These are supported by 16 tertiary indicators. Human capital accumulation and high-level talent reflect improvements in the overall quality and capability of China’s workforce. Tangible labor materials and intangible labor materials represent the main innovative modes of labor materials in China today. Resource and environmental input and digital economy development are manifestations of the iterative upgrading of labor objects. The first four columns of Table 1 provide a detailed breakdown of this evaluation framework.

Measurement of NQPFS. Enterprises in developed countries possess complete location information, making it possible to gauge the direction and scale of factor flows by referencing changes in the location of industrial enterprises. However, due to the lack of detailed data on factor allocation in China, there is considerable controversy over methods for measuring factor flows. Factor flows manifest as the dynamic spatial distribution and evolution of development factors across different periods. The patterns of non-regional factor flows in China can be observed by comparing the spatial trends of non-regional factor movements (Wang & Su, 2020) [43]. Drawing on the methods for calculating industrial transfer in geographic regions proposed by Zhao & Yin (2011) [44], and extending the methodological applications of Liu et al. [45] and Zhang et al. [21], we select several years prior to the occurrence of new quality productivity forces flow as the base research period. Relative changes in economic indicators before and after the occurrence of new quality productivity forces flow are used to measure the level of new quality productivity forces flow in coastal cities. Entering the 21st century, China’s marine sector embarked on a series of institutionalized, standardized, and regulated developments. In January 2003, the implementation of the Ocean Statistical Express Report System was approved by the National Bureau of Statistics. Subsequently, in May of the same year, the State Council issued the National Marine Economic Development Plan Outline, which accelerated the systematic compilation of marine economic statistics in coastal cities. Accordingly, given this institutional inception point, data from the year 2003 is selected as the base period for this study. To eliminate interference caused by preexisting disparities between regions, we incorporate the proportion of a region’s economic output relative to the city’s total output. The improved NQPFS index is represented by Equation (1):

$$F{R}_{ri,t}={P^{\prime }}_{ri,t}-{P^{\prime }}_{ri,t_0}={\displaystyle \frac{q_{ri,t}}{{{\displaystyle \sum }}_{r=1}^n{q}_{ri,t}}}/{\displaystyle \frac{{{\displaystyle \sum }}_{i=1}^m{q}_{ri,t}}{{{\displaystyle \sum }}_{i=1}^m{{\displaystyle \sum }}_{\mathrm{r}=1}^n{q}_{ri,t}}}-{\displaystyle \frac{q_{ri,t_0}}{{{\displaystyle \sum }}_{r=1}^n{q}_{ri,t_0}}}/{\displaystyle \frac{{{\displaystyle \sum }}_{i=1}^m{q}_{ri,t_0}}{{{\displaystyle \sum }}_{i=1}^m{{\displaystyle \sum }}_{r=1}^n{q}_{ri,t_0}}}$$

(1)

where $F{R}_{ri,t}$ denotes the improved factor mobility index, m represents the total quantity of New Quality Productive Force elements under study, and $q_{ri,t}$ indicates the scale of the new qualitative productivity force element $i$ in region $r$ during year $t$. If $F{R}_{ri,t}>0$, it signifies an inflow of element i into region r in that year relative to the base period. If $F{R}_{ri,t}<0$, it indicates an outflow of element $i$ from region $r$ in that year relative to the base period. Consequently, the calculated results of the improved NQPFS index not only reveal the direction of New Quality Productive Forces elements but also quantify the magnitude of such flows. For the measurement scores of NQPFS, within the indicator system established in this study, the mobility levels of 19 specific indicators are measured, and the final composite index of NQPFS is obtained using the weighted average method, where each specific indicator is assigned an equal weight of 1/19 (0.05263), as shown in

Table 2. Evaluation Indicator System for Levels of NQPFS.

Primary Indicator	Secondary Indicator	Tertiary Indicator	Specific Indicator	EQ W.
Labor	Human Capital Accumulation	Future Labor Force Reserve	Number of Domestic Undergraduate Students (persons)	0.0526
			Number of International Students in China (persons)	0.0526
		Full-Time Equivalent of Practitioners	Number of Employees at Period-end (persons)	0.0526
	High-end Talent Level	High-quality Labor Force	Marine R&D Personnel (persons)	0.0526
			Attention to Introduction of Overseas Science and Technology Talent (times)	0.0526
Means of Labor	Tangible Means of Labor	Domestic Capital	Marine Capital Stock (10,000 yuan)	0.0526
		International Capital	Level of Foreign Capital Utilization (10,000 USD)	0.0526
	Intangible Means of Labor	Intelligent Technology	Technology Market Transaction Value (100 million yuan)	0.0526
			Contract Value of Foreign Technology Introduction (10,000 USD)	0.0526
		Patent Authorization	Number of Patents Granted (items)	0.0526
			Number of Robot Patents Granted to Listed Companies (items)	0.0526
Objects of Labor	Resource and Environment Input	Resource Input	Total Marine Energy Consumption (100 tons of standard coal)	0.0526
			Port International Standard Container Throughput (10,000 TEU)	0.0526
		Environmental Input	Number of Green Patents Granted (items)	0.0526
			Expenditure on Energy Conservation and Environmental Protection (10,000 yuan)	0.0526
	Digital Economy Development	Data Factor Application	Level of Enterprise Data Factor Utilization (times)	0.0526
			E-commerce Transaction Volume (10,000 yuan)	0.0526
		Digital Platform Construction	Number of Internet Broadband Access Subscribers (1000 households)	0.0526
			Attention to Cross-border E-commerce (times)	0.0526

. The primary, secondary, and composite indices are then synthesized level by level through the weighted mean approach, with the sum of all specific indicator weights equaling 1 (Wang & Su, 2020) [43].

Measurement of NQPFC. The objective of optimizing the combination of production factors is to enhance the allocation of existing resources and elevate productivity levels. Studying the combination efficiency of new quality production factors can help gauge the current state of resource allocation and utilization of these emerging factors. Drawing on the theory and methodology of Zhao et al. [46] and grounded in the core concepts of “innovation” and “green development,” this paper attempts to construct an “input-output” evaluation indicator system for new quality productive forces. Utilizing Max DEA 8 Ultra software and applying the super-efficiency SBM model, the optimization and combination efficiency of new quality productive forces is measured for 53 coastal cities in China from 2004 to 2023. Aligning with the model construction process and the core characteristics of new quality productive forces, an “input-output” indicator system spanning three dimensions-factor inputs, desirable outputs, and undesirable outputs—has been established. For the undesirable outputs under Tier-2 Indicators, greenhouse gas emissions and low-quality innovation outputs were selected as specific variables. In this regard, drawing on the approaches of core references such as Li & Zheng [47] and Xu et al. [48], this study categorizes patents into high-quality innovation outputs (invention patents) and low-quality innovation outputs (design patents) based on the degree of substantive innovation, as detailed in

Table 3. Input-Output Indicator System for NQPFC.

Variable Type	Tier-1 Indicator	Tier-2 Indicator	Tier-3 Indicator
Input Indicators	Labor Input	Workforce Input	Number of Employed Persons (persons)
	Capital Input	Tangible Capital Input	Capital Stock (10,000 yuan)
		Intangible Capital Input	Utilization Frequency of Enterprise Data Elements (times)
			Value of Technology Market Transactions (100 million yuan)
	R&D Input	Human Capital Input	R&D Personnel (persons)
		R&D Expenditure Input	Internal Expenditure on R&D (10,000 yuan)
	Resource Input	Water Resource Input	Urban Water Withdrawal (10,000 m3)
		Green Space Investment	Urban Park Construction Investment (hectares)
		Energy Input	Total Energy Consumption (million tons of SCE)
Output Indicators	Desirable Outputs	Innovation Output	Number of Invention Patent Grants (items)
			Number of Robot Patent Grants (items)
		Economic Output	Gross Regional Product (100 million yuan)
		Social Output	Number of Internet Broadband Subscribers (1000 households)
	Undesirable Outputs	Environmental Output	Greenhouse Gas Emissions (tons)
		Low-quality Innovation	Number of Design Patent Applications/Grants (items)

.

Control Variables. Coastal HQMED is also influenced by economic performance, urban and industrial development, and investments in science, education, and related fields. To mitigate the potential effect of omitted-variable bias on causal inference and reduce estimation errors in the regression results, we refer to Chen et al. [49], Bao et al. [50], Tan et al. [51], Wang & Zhang [52], and Zhan & Liang [53]. Our approach incorporates five control variables: economic development level, urbanization level, industrialization level, overall education investment level, and science and technology innovation level. And the descriptive statistical analysis of the variables in this study is presented in

Table 4. Descriptive Statistical Analysis.

Variable	Sample	Mean	Standard Deviation	Minimum	Median	Maximum
HQMED	1060	0.059	0.060	0.011	0.041	0.495
NQPFS	1060	0.189	0.628	−0.330	0.057	9.444
NQPFC	1060	0.826	0.323	0.079	1.001	2.027
lnDEV	1060	10.850	0.714	8.543	10.925	13.056
URB	1060	0.613	0.171	0.217	0.616	1.355
IND	1060	0.504	0.168	0.084	0.507	1.157
EDU	1060	0.192	0.050	0.057	0.191	0.368
RD	1060	0.022	0.020	0.000	0.016	0.130

.

Mediating Variable. First, enhancement of the marine industrial chain resilience (MIC). Based on evolutionary economics and adaptive resilience theory, industrial chain resilience can be understood as the capacity of an industrial chain—when facing risks and shocks—to continuously strengthen its adaptability and flexibility in responding to such risks through technological advancement and knowledge innovation. Drawing on the entropy-based indicator construction method for industrial chain resilience proposed by [54], this paper primarily employs industrial diversification and innovation capability for measurement. First, considering data availability and scientific rigor at the prefecture-level city scale, nighttime light intensity data within a 100 km coastal buffer are used as a proxy for the Herfindahl-Hirschman Index (HHI) to gauge the concentration level of marine industries. Lower nighttime light intensity values indicate lower industrial concentration, reflecting stronger diversification of marine industries, which in turn implies greater capacity of the marine industrial chain to withstand external shocks and maintain stability. Second, the number of marine R&D personnel at the city level is adopted to measure innovation capability. Finally, a comprehensive index of marine industrial chain resilience is derived by combining these measures with the proportion of gross output value of the marine tertiary industry in the Gross Ocean Product (GOP).

Second, the emergence of New Quality Business Forms in the Marine Sector (NBE). As a key mediating mechanism, the emergence of new quality business forms in the marine sector functions primarily through organizational and institutional innovation. It reconstructs originally fragmented and highly asset-specific knowledge and technology elements into new production functions and market-oriented transaction structures. To measure this mediating variable, this paper adopts an equal-weighting approach to integrate the number of new startups in the city, the number of listed enterprises with investment and financing qualifications, and their level of R&D expenditure. The composite index is then adjusted by the proportion of marine GDP to national GDP [21].

Moderating Variables. First, Innovation Ecology (IE). This paper employs the Comprehensive Urban Competitiveness Index of China to measure the level of regional innovation ecology. The Comprehensive Urban Competitiveness Index can, to a certain extent, reflect the development of a city’s business environment. A high-quality business environment provides stable institutional expectations and efficient resource-matching mechanisms for innovation activities, directly reducing institutional transaction costs and uncertainty for enterprises. This enables innovation actors to focus more resources on research, development, and collaboration, thereby constructing a city-level innovation ecosystem characterized by unimpeded factor flows, effective incentives, and controllable risks. The Comprehensive Urban Competitiveness Index is sourced from the Annual Report on Urban Competitiveness in China compiled by the Chinese Academy of Social Sciences. Second, Resource Misallocation (RM). Resource misallocation refers to a deviation from the effective allocation of resources. If resources could flow freely and achieve Pareto optimality, that would constitute effective allocation; resource misallocation represents a departure from this optimal state. Drawing on the calculation method from [55], this paper derives the extent of resource misallocation in capital and labor factor markets across various regions in China, namely the degrees of capital factor misallocation and labor factor misallocation.

4.3. Econometric Models: Baseline Regression Model

Here, we construct an econometric model to empirically examine the direct effect of NQPFS and NQPFC on the HQMED of China’s coastal regions. Specifically, using panel data from 53 coastal cities in China spanning 2004–2023, we use a two-way fixed-effects model to test the direct effects, mechanisms, and heterogeneous characteristics of NQPFS and NQPFC in relation to the HQMED of coastal areas. The baseline regression model is shown in Equations (2) and (3):

$$HQME{D}_{i,t}={\beta }_0+{\beta }_{1}NQPF{S}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(2)

$$HQME{D}_{i,t}={\beta }_0+{\beta }_{2}NQPF{C}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(3)

where $i$ and $t$ represent different cities and years, respectively. $HQME{D}_{i,t}$ denotes the level of HQMED in China’s coastal cities. $NQPF{S}_{i,t}$ indicates the level of NQPFS, while $NQPF{C}_{i,t}$ represents the level of NQPFC. $Control{s}^j$ stands for the $j$ control variable, $u_i$ captures city-fixed effects, ${\eta }_t$ denotes time effects, and ${\epsilon }_{i,t}$ is the random disturbance term. ${\beta }_1$ signifies the effect coefficient of NQPFS on the HQMED of China’s coastal cities, and ${\beta }_2$ represents the effect coefficient of NQPFC on the HQMED of China’s coastal cities. If ${\beta }_1>0$, it indicates that NQPFS promotes the HQMED of China’s coastal cities; conversely, it signifies a suppressive effect. If ${\beta }_2>0$, it means NQPFC helps enhance HQMED in China’s coastal cities; otherwise, it indicates an inhibitory effect.

4.4. Econometric Models: Mediation and Moderation Models

Then, we construct an econometric model to empirically examine the indirect effect of NQPFS and NQPFC on the HQMED of China’s coastal regions. The mediation and moderation Models are as shown in Equations (4)–(9)

$$MI{C}_{i,t}={\beta }_0+{\beta }_{3}NQPF{S}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(4)

$$HQME{D}_{i,t}={\beta }_0+{\beta }_{1}NQPF{S}_{i,t}+{\beta }_{2}MI{C}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(5)

$$NB{E}_{i,t}={\beta }_0+{\beta }_{4}NQPF{C}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(6)

$$HQME{D}_{i,t}={\beta }_0+{\beta }_{1}NQPF{C}_{i,t}+{\beta }_{2}NB{E}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(7)

$$HQME{D}_{i,t}={\beta }_0+{\beta }_{5}NQPF{S}_{i,t}+{\beta }_{6}R{E}_{i,t}+{\beta }_{7}NQPF{S}_{i,t}\ast R{E}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(8)

$$HQME{D}_{i,t}={\beta }_0+{\beta }_{8}NQPF{C}_{i,t}+{\beta }_{9}I{E}_{i,t}+{\beta }_{10}NQPF{C}_{i,t}\ast I{E}_{i,t}+{\displaystyle \sum }_j{\delta }_{j}control{s}_{i,t}^j+u_i+{\eta }_t+{\epsilon }_{i,t}$$

(9)

where $i$ and $t$ represent different cities and years, respectively. $MI{C}_{i,t}$ represents the improvement in the marine industrial chain resilience; $NB{E}_{i,t}$ represents the emergence of new-quality marine enterprises; $R{E}_{i,t}$ represents the degree of resource misallocation, including both capital and labor misallocation; $I{E}_{i,t}$ represents the innovation ecosystem.

5. Spatio-Temporal Evolution: Descriptive Evidence

5.1. Spatial Correlation Analysis of High-Quality Economic Development in China’s Coastal Areas (2004–2023)

Given the presence of certain spatial dependence in the development of the marine economy in China’s coastal regions, traditional econometric models based on the assumption of spatial independence struggle to effectively capture spatial characteristics. Accordingly, this study employs a spatial correlation test to analyze the spatial agglomeration of high-quality development levels in China’s marine economy (HQMED). The current academic literature primarily utilizes Moran’s I index—including both global and local Moran’s I—to test spatial correlation. As shown in

Table 5. Moran’s I Test Results and Statistics.

Year	Moran’s I	Z	Year	Moran’s I	Z
2004	0.177	2.735	2014	0.200	3.161
2005	0.192	2.904	2015	0.210	3.264
2006	0.195	2.953	2016	0.214	3.280
2007	0.195	2.955	2017	0.205	3.171
2008	0.207	3.125	2018	0.203	3.174
2009	0.204	3.087	2019	0.199	3.147
2010	0.203	3.086	2020	0.192	3.049
2011	0.199	3.056	2021	0.190	3.045
2012	0.196	3.025	2022	0.203	3.136
2013	0.202	3.113	2023	0.215	3.238

, from 2004 to 2023, all Moran’s I values were greater than zero and statistically significant at the 1% level, indicating a clear spatial interdependence in the high-quality development levels of the marine economy across China’s coastal cities. Furthermore, the year-on-year increase in Moran’s I suggests a corresponding annual enhancement in the spatial correlation of high-quality marine economic development across these coastal cities. Furthermore, the inverse distance spatial weight matrix aligns more closely with people’s intuitive understanding of spatial relationships and is widely regarded as the classic interpretation of the First Law of Geography—namely, the greater the distance between two locations, the smaller the weight, and vice versa. Therefore, this study adopts the inverse distance matrix as the spatial weight matrix to reflect the geographical proximity among China’s coastal cities.

Furthermore, a local spatial autocorrelation test was conducted on the high-quality development of the marine economy to examine its overall distribution pattern and characteristics. Taking the years 2004, 2010, 2016, and 2023 as examples, local Moran’s I scatter plots for HQMED were plotted. The serial numbers 1–53 represent the 53 coastal cities, respectively, as shown in Figure 2.

In 2004, high-quality marine economic development in China’s coastal regions exhibited distinct spatial heterogeneity. According to the Moran scatter plot, only a few cities—such as Dongguan, Shenzhen, Tianjin, Guangzhou, and Zhuhai—were located in Quadrant I, displaying a “high-high” clustering spatial correlation pattern. This indicates that these cities had significant positive spatial correlations in marine economic development, forming synergistic agglomeration zones among high-quality cities. In contrast, the vast majority of coastal cities were situated in Quadrant III, exhibiting a “low-low” clustering pattern, reflecting that cities with relatively lower levels of marine economic development were spatially adjacent to one another, characterized by weak spatial dependence and insufficient development momentum.

By 2023, the spatial structure of marine economic quality in coastal regions had undergone significant optimization. The number of cities in Quadrant I demonstrating the “high-high” correlation pattern increased notably, including Shenzhen, Tianjin, Guangzhou, Hangzhou, Qingdao, Dongguan, and others. This reveals a further strengthening of spatial agglomeration effects among high-quality cities, resulting in a broader pattern of regional coordinated development. Simultaneously, the number of cities in the “low-low” cluster of Quadrant III decreased, such as Cangzhou, Beihai, Jinzhou, Panjin, Dandong, and Qinzhou. Although spatial negative spillover effects persisted, this overall reflects a narrowing scope of low-quality agglomeration and a certain mitigation of interregional development disparities.

Based on the construction of the indicator system and weight settings, China’s HQMED is scientifically measured and its final scores are calculated using a combination of the entropy method. The temporal variation characteristics of the high-quality development levels of the marine economy in cities belonging to the Northern, Eastern, and Southern Marine Economic Circles during the sample period are illustrated in Figure 3.

During the period from 2004 to 2023, the HQMED level of China’s three major marine economic circles exhibited an overall upward trend, yet with notable regional growth divergence. The Eastern Marine Economic Circle recorded the highest average annual growth rate, reaching approximately 7.0%. Following 2015, its growth further accelerated, with the average annual rate exceeding 8% in the later years, continuously widening its leading advantage. The Southern Marine Economic Circle maintained steady growth, with an average annual rate of about 5.5%. In contrast, the Northern Marine Economic Circle experienced relatively slower growth, averaging around 6.4%, alongside greater volatility—for instance, its growth in 2020, while not negative, was markedly lower than in other years, reflecting relatively weaker development resilience. The absolute development gap between the Eastern Circle and the Northern as well as Southern circles continued to expand over time, further accentuating the spatial differentiation pattern. Despite progress across all regions, the “Matthew Effect” is evident, as the high-growth Eastern region achieved faster advancement, thereby reinforcing the pattern of “Eastern leadership with Northern and Southern regions following.”

5.2. Temporal Analysis of New Quality Productive Forces Spatial Flow (2004–2023)

Based on the constructed indicator system, this study employs the weighted average method to scientifically measure the NQPFS and calculate its final scores. Figure 4 illustrates the temporal variation characteristics of the NQPFS levels in cities belonging to the Northern, Eastern, and Southern Marine Economic Circles during the sample period.

Between 2004 and 2023, the spatial flow index of new quality productive forces in China’s three major marine economic circles exhibited an evolutionary pattern characterized by a shift from “North–South polarization” to “Eastern–Southern leadership” From 2004 to 2010, the Southern Marine Economic Circle was a clear net outflow region for factors, forming a pronounced spatial polarization with the Northern and Eastern circles. After 2010, the Southern circle achieved a paradigm shift from negative to positive growth, with an average annual growth rate of 18.2%. By 2018, its index surged to 0.5353, surpassing the Eastern circle to become the region with the highest level of factor mobility, marking its systemic transition from a “factor depression” to a “factor attraction hub.” The Eastern circle displayed a “pulse-type leap” pattern, with an average annual growth rate of approximately 12.5%. The peak values coincided with national-level strategic initiatives-for instance, the high level in 2014 (0.4137) likely benefited from the advancement of the Belt and Road Initiative, particularly the 21st-Century Maritime Silk Road, as well as the institutional innovation dividends released by the Shanghai Pilot Free Trade Zone, which significantly enhanced the Eastern region’s attractiveness to high-end factors. The 2018 leap further corresponded to the deepening of the “Marine Power” strategy, the launch of the Hainan Free Trade Port, and the acceleration of Yangtze River Delta integration, reinforcing the Eastern region’s pivotal role in innovation and industrial chains. The Northern circle experienced relatively moderate growth, with an average annual rate of about 13.8%, showing the least volatility but overall weaker momentum. Collectively, in the later period (2018–2023), the three regions demonstrated a trend of coordinated advancement, with their flow indices all entering a relatively high range above 0.2, reflecting an enhanced overall capacity of coastal areas to attract new quality productive forces. Nevertheless, regional disparities persisted, maintaining a gradient pattern of “Southern and Eastern leading, Northern lagging.

5.3. Temporal Analysis of Optimal Combination of New Quality Productive Forces (2004–2023)

Based on the constructed indicator system, this study employs the weighted average method to scientifically measure the NQPFC and calculate its final scores. Figure 5 illustrates the temporal variation characteristics of the NQPFC levels in cities belonging to the Northern, Eastern, and Southern Marine Economic Circles during the sample period.

During the period from 2004 to 2023, the NQPFC efficiency of China’s three major marine economic circles exhibited a distinct spatial-temporal evolution characterized by “initial alternating leadership followed by Eastern advancement and Southern progression.” In the early phase of the study, the Northern Marine Economic Circle briefly held an advantage in combination efficiency due to its scale-based input of traditional factors. However, since 2009, the Eastern Marine Economic Circle has overtaken and maintained a leading position by leveraging its structural advantages. In terms of growth trends, the Eastern Marine Economic Circle consistently remained at the forefront, with the highest average annual growth rate of its NQPFC efficiency index, approximately 3.36%. After 2015, it entered an accelerated enhancement phase. The Southern Marine Economic Circle demonstrated steady growth, with an average annual rate of about 2.40%, showcasing strong catch-up resilience. Since 2017, its index has repeatedly exceeded 1.0, gradually narrowing the gap with the Eastern region. In contrast, the Northern Marine Economic Circle experienced relatively slow growth, with an average annual rate of around 2.12%, marked by significant volatility, particularly with a noticeable decline in 2020, reflecting weaker developmental resilience and insufficient stability and sustainable optimization capability in its factor allocation system. The three major coastal economic regions displayed a pronounced “spatial polarization” in NQPFC efficiency, where the Eastern region, as the core zone, established a self-reinforcing mechanism through efficient factor synergy, leading to a continuously widening gap with the Northern peripheral region. This reveals that the development of new quality productive forces depends not only on the scale of factor inputs but, more critically, on the construction of regional innovation ecosystems capable of continuously optimizing factor structures and enhancing allocation resilience.

6. Empirical Results and Analysis

6.1. Baseline Results: Validating the Direct Effects

Based on the stepwise regression results presented in

Table 6. Benchmark Model Regression Results of NQPFS.

	(1)	(2)	(3)	(4)	(5)	(6)
	HQMED	HQMED	HQMED	HQMED	HQMED	HQMED
NQPFS	0.006 ***	0.006 ***	0.005 ***	0.005 ***	0.006 ***	0.006 ***
	(4.39)	(4.57)	(4.23)	(4.24)	(4.40)	(5.04)
In DEV		−0.022 ***	−0.021 ***	−0.025 ***	−0.019 ***	−0.015 ***
		(−4.88)	(−4.81)	(−4.92)	(−3.74)	(−3.11)
URB			−0.073 ***	−0.070 ***	−0.070 ***	−0.067 ***
			(−5.97)	(−5.71)	(−5.78)	(−5.76)
IND				0.014	0.015 *	0.007
				(1.50)	(1.69)	(0.81)
EDU					0.188 ***	0.157 ***
					(5.34)	(4.60)
RD						0.645 ***
						(9.06)
cons	0.058 ***	0.294 ***	0.332 ***	0.365 ***	0.265 ***	0.218 ***
	(72.51)	(6.07)	(6.90)	(6.90)	(4.78)	(4.07)
City Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	1060	1060	1060	1060	1060	1060
Adj. R2	0.829	0.833	0.838	0.839	0.843	0.855

Note: t-statistics in parentheses *** p < 0.01.

and

Table 7. Benchmark Model Regression Results of NQPFC.

	(1)	(2)	(3)	(4)	(5)	(6)
	HQMED	HQMED	HQMED	HQMED	HQMED	HQMED
NQPFC	0.022 ***	0.024 ***	0.023 ***	0.023 ***	0.023 ***	0.020 ***
	(7.25)	(7.86)	(7.83)	(7.69)	(7.86)	(6.78)
In DEV		−0.025 ***	−0.024 ***	−0.025 ***	−0.019 ***	−0.016 ***
		(−5.59)	(−5.53)	(−5.00)	(−3.80)	(−3.21)
URB			−0.074 ***	−0.073 ***	−0.073 ***	−0.071 ***
			(−6.18)	(−6.07)	(−6.15)	(−6.15)
IND				0.004	0.005	−0.001
				(0.39)	(0.56)	(−0.06)
EDU					0.188 ***	0.160 ***
					(5.44)	(4.76)
RD						0.550 ***
						(7.73)
cons	0.040 ***	0.306 ***	0.344 ***	0.352 ***	0.252 ***	0.214 ***
	(15.33)	(6.44)	(7.31)	(6.80)	(4.65)	(4.04)
City Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	1060	1060	1060	1060	1060	1060
Adj. R2	0.834	0.839	0.845	0.845	0.849	0.858

Note: t-statistics in parentheses *** p < 0.01, * p < 0.1.

, the coefficients for the core explanatory variables NQPFS and NQPFC are consistently positive and statistically significant at the 1% level, regardless of whether control variables are included. This indicates that both the spatial flow and optimal combination of new-quality productive forces exert a stable positive effect on high-quality economic development in coastal regions, thereby supporting Hypothesis 1. A comparison of the coefficient magnitudes shows that the coefficient for NQPFC ranges from 0.020 to 0.024, which is substantially larger than that for NQPFS, which ranges from 0.005 to 0.006. This suggests that, at the current stage, the marginal contribution of factor optimal combination to high-quality development exceeds that of spatial flow alone, highlighting the critical role of improving factor allocation efficiency. During the stepwise inclusion of control variables, EDU and RD exhibit significantly positive coefficients, indicating that regional education levels and R&D investment robustly enhance the quality of development. All models pass the Hausman test, supporting the use of a two-way fixed effects specification. The gradual increase in adjusted R² with the addition of control variables further confirms the model’s strong explanatory power and appropriate variable selection.

6.2. Endogeneity and Tests Robustness

6.2.1. Endogeneity with Instrumental Variables (2SLS)

A valid instrumental variable must satisfy the fundamental requirements of relevance and Exogeneity. First, drawing on the methodology of Huang et al. [56] this paper selects the product of the number of landline telephones per hundred people in each city in 1984 and the national internet penetration rate over the years as an instrumental variable. In terms of relevance, historically well-connected cities have gained greater “digital dividends” amid the wave of information technology revolution, thereby being more capable of attracting the flow of New Quality Productive Forces that rely on information networks. The year 1984 predates the large-scale development of China’s marine economy, and the deployment of landline telephones at that time was determined by factors such as administrative allocation during the planned economy era, initial population density, and political status, which are not directly related to unobserved factors affecting current marine economic quality. Second, following the approach of Gao &Li [57], this paper employs the number of higher education institution faculty as an instrumental variable to examine the impact of NQPFC on HQMED. University faculty serve as the core agents of knowledge production, technological research and development, and talent cultivation. Through research activities and talent training, such as the output of marine professionals, they directly drive technological progress and factor upgrading. However, high-quality marine development of coastal cities depends on factors such as marine resource endowment, industrial policies, and international market demand, which are not directly influenced by the scale of university faculty.

Table 8. Endogeneity Test.

	(1)	(2)	(3)	(4)
	NQPFS (First Stage)	HQMED (Second Stage)	NQPFC (First Stage)	HQMED (Second Stage)
NQPFS		0.024 **
		(2.51)
NQPFC				0.144 ***
				(6.51)
In DEV	0.030	−0.017 **	0.081	−0.020 *
	(0.40)	(−2.49)	(1.34)	(−1.76)
URB	−0.718 ***	−0.055 ***	0.048	−0.069 ***
	(−2.86)	(−3.42)	(0.35)	(−3.06)
IND	0.036	0.007	0.370 ***	−0.049 ***
	(0.27)	(0.65)	(4.41)	(−2.58)
EDU	−0.431	0.164 ***	−1.084 ***	0.198 ***
	(−0.63)	(3.94)	(−2.79)	(3.03)
RD	−3.277 **	0.699 ***	2.212 **	0.076
	(−2.26)	(6.41)	(2.52)	(0.44)
lniv1	0.174 ***
	(4.42)
lniv2			0.030 ***
			(7.13)
City Fixed Effects	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes
C-D Wald F statistic	75.943		79.217
K-P Wald F statistic	19.505		50.895
LM statistic	17.346 p-val = 0.0000		41.600 p-val = 0.0000
N	1060	1060	1060	1060
Adj. R2	0.006	−0.085	0.056	−1.489

Note: t-statistics in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the results of the endogeneity tests, which support the robustness of the core findings. In the two-stage least squares regressions, the Cragg-Donald Wald F statistics are 75.943 and 79.217, respectively, both exceeding the empirical threshold of 10, indicating that the selected instrumental variables are strongly correlated with the endogenous variables and that weak instrument issues are not a concern. The Kleibergen-Paap Wald F statistics are 19.505 and 50.895, further confirming the strength of the instruments. Additionally, the LM statistics are 17.346 and 41.600, with p-values of 0.0000, rejecting the null hypothesis of underidentification and supporting instrument validity.

After controlling for endogeneity, the coefficients for NQPFS and NQPFC remain significantly positive at the 5% and 1% levels, respectively, with estimated values of 0.024 and 0.144. These magnitudes are notably larger than those in the baseline regressions, suggesting that after addressing potential endogeneity bias, the spatial flow and optimal combination of new-quality productive forces continue to exert a significant—and even stronger—promoting effect on the high-quality development of coastal cities in China.

6.2.2. Robustness Test: Replace Dependent Variable and Add Control Variables

First, to enhance the robustness of the findings, we replace the dependent variable. Specifically, coastal HQMED is substituted with marine labor productivity (MLPR) for model validation. Drawing on Chen [58], who used per capita output as a proxy for economic development quality, MLPR (in yuan/person) is calculated by dividing the gross marine product of coastal cities by the year-end number of marine industry employees. The regression results in columns (1) and (2) of

Table 9. Robustness Checks.

	(1)	(2)	(3)	(4)
	MLPR	MLPR	HQMED	HQMED
NQPFS	0.001 ***		0.006 ***
	(5.15)		(5.01)
NQPFC		0.001 ***		0.019 ***
		(2.73)		(6.40)
In DEV	0.002 ***	0.002 ***	−0.024 ***	−0.019 ***
	(3.93)	(3.91)	(−3.89)	(−3.20)
URB	0.001	0.001	−0.065 ***	−0.070 ***
	(1.17)	(0.82)	(−5.62)	(−6.07)
IND	0.000	−0.000	−0.003	−0.005
	(0.14)	(−0.20)	(−0.34)	(−0.51)
EDU	−0.011 ***	−0.011 ***	0.150 ***	0.157 ***
	(−3.17)	(−3.13)	(4.41)	(4.65)
RD	−0.001	−0.006	0.572 ***	0.520 ***
	(−0.19)	(−0.88)	(7.37)	(6.75)
InECOD			0.018 **	0.008
			(2.33)	(1.03)
_cons	−0.015 ***	−0.015 ***	0.170 ***	0.192 ***
	(−2.79)	(−2.79)	(2.96)	(3.38)
City Fixed Effects	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes
N	1060	1060	1060	1060
Adj. R2	0.541	0.532	0.856	0.858

Note: t-statistics in parentheses *** p < 0.01, ** p < 0.05.

show that the coefficients for NQPFS and NQPFC remain significantly positive at the 1% level, supporting the conclusions of this study.

Second, following the approach of Zhu & Sun [59], urban economic density (ECOD) is introduced as a control variable for regional economic development, measured as the ratio of regional GDP to the city’s administrative area. Cities with higher economic density typically possess better infrastructure, lower transaction costs, and more dynamic knowledge spillovers. These factors not only facilitate the development of New Quality Productivity but may also directly contribute to high-quality development outcomes such as economic growth and industrial upgrading. Omitting economic density could lead to an overestimation of the impact of New Quality Productivity. As shown in columns (3) and (4) of Table 9, after controlling for ECOD, the coefficients for NQPFS and NQPFC are 0.006 and 0.019, respectively, both statistically significant at the 1% level. The coefficient for ECOD is also positive and significant in column (3), further corroborating the robustness of the core findings.

6.3. Mediation Mechanism Test

Based on the preceding theoretical analysis, this study posits two distinct mediation pathways: “NQPFS → MIC → HQMED” and “NQPFC → LnNBE → HQMED.” Hence, drawing on the causal inference framework proposed by Jiang [60], we employ the widely accepted two-step approach for mediation analysis to examine the underlying operational mechanisms. This method mitigates the reliability issues associated with traditional techniques that arise due to endogeneity. The test results are presented in

Table 10. Mediation Effect Test.

	(1)	(2)	(3)	(4)
	MIC	HQMED	LnNBE	HQMED
NQPFS	0.022 ***	0.005 ***
	(8.02)	(3.85)
NQPFC			0.961 ***	0.019 ***
			(3.12)	(6.45)
MIC		0.061 ***
		(4.35)
LnNBE				0.001 ***
				(3.39)
In DEV	−0.053 ***	−0.012 **	−1.537 ***	−0.014 ***
	(−4.76)	(−2.45)	(−2.99)	(−2.89)
URB	0.022	−0.068 ***	4.775 ***	−0.076 ***
	(0.85)	(−5.93)	(3.94)	(−6.56)
IND	0.094 ***	0.001	−1.853 **	0.001
	(4.77)	(0.15)	(−2.01)	(0.16)
EDU	0.284 ***	0.139 ***	7.587 **	0.153 ***
	(3.70)	(4.10)	(2.14)	(4.54)
RD	0.957 ***	0.586 ***	32.227 ***	0.517 ***
	(5.97)	(8.17)	(4.30)	(7.24)
_cons	0.574 ***	0.183 ***	26.090 ***	0.187 ***
	(4.76)	(3.41)	(4.69)	(3.52)
City Fixed Effects	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes
N	1060	1060	1060	1060
Adj. R2	0.879	0.858	0.812	0.859

Note: t-statistics in parentheses *** p < 0.01, ** p < 0.05.

.

The first-step results in Table 10 indicate that both NQPFS and NQPFC significantly drive the proposed mediating variables. Specifically, NQPFS exerts a positive impact on MIC, with a coefficient of 0.022 significant at the 1% level. Similarly, NQPFC shows a strong positive effect on LnNBE, with a coefficient of 0.961 also significant at the 1% level. These results provide preliminary evidence supporting the first step of each mediation pathway. The second-step results further confirm the mediation effects. After introducing the mediating variable MIC, the coefficient of NQPFS on HQMED decreases to 0.005 but remains significant at the 1% level, suggesting that MIC partially mediates the relationship between NQPFS and HQMED. In parallel, after including LnNBE, the coefficient of NQPFC on HQMED declines to 0.019 while remaining significant at the 1% level, indicating that LnNBE also serves as a partial mediator. The positive and significant coefficients of both MIC (0.061) and LnNBE (0.001) in their respective second-step regressions further verify the transmission roles of these two channels.

Overall, the findings validate that new-quality productive forces promote high-quality development in coastal cities through two distinct mechanisms: enhancing marine industrial chain resilience and fostering new-quality business forms in the marine sector, thereby supporting Hypotheses 2 and 3.

6.4. Moderating Effect Mechanism Test

Based on the theoretical analysis above, this paper posits that Misallocation (RM) leads to the fragmentation of the unified market, thereby hindering the spatial flow of New Quality Productivity and negatively affecting HQMED by reducing resource allocation efficiency. Resource misallocation can be further categorized into capital misallocation (KM) and labor misallocation (LM). Conversely, a favorable Innovation Ecology (IE) further enhances the cohesion and recombination capacity of New Quality Productivity, exerting a positive influence on HQMED. To test these conjectures, a moderating effect model is constructed to examine the underlying patterns of such external environmental influences.

According to the moderating effect test results presented in

Table 11. Moderation Effect Test.

	(1)	(2)	(3)
	HQMED	HQMED	HQMED
NQPFS	0.005 ***	0.005 ***
	(4.46)	(4.08)
NQPFS × KM	−0.010 ***
	(−3.73)
KM	0.029 ***
	−0.010 ***
NQPFS × LM		−0.010 ***
		(−5.27)
LM		0.014 ***
		(10.98)
NQPFC			0.009 ***
			(3.06)
NQPFC × IE			0.107 ***
			(10.50)
IE			0.062 ***
			(5.11)
In DEV	−0.005	−0.025 ***	−0.014 ***
	(−0.93)	(−5.32)	(−3.02)
URB	−0.064 ***	−0.067 ***	−0.054 ***
	(−5.70)	(−6.18)	(−5.04)
IND	0.001	0.011	−0.002
	(0.15)	(1.33)	(−0.23)
EDU	0.079 **	0.209 ***	0.085 ***
	(2.30)	(6.58)	(2.73)
RD	0.476 ***	0.706 ***	0.387 ***
	(6.61)	(10.69)	(5.78)
_cons	0.112 **	0.298 ***	0.187 ***
	(2.09)	(5.96)	(3.85)
City Fixed Effects	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes
N	1060	1060	1060
Adj. R2	0.865	0.876	0.881

Note: t-statistics in parentheses *** p < 0.01, ** p < 0.05.

, resource misallocation exerts a significantly negative moderating effect on the relationship between NQPFS and HQMED. Specifically, the interaction term coefficients for both capital misallocation (NQPFS × KM) and labor misallocation (NQPFS × LM) are −0.010, statistically significant at the 1% level as shown in columns (1) and (2). This indicates that as the degree of either capital or labor misallocation intensifies, the positive effect of NQPFS on HQMED is substantially weakened due to distorted resource allocation. Conversely, this result confirms that reducing resource misallocation and advancing market-oriented reforms in factor allocation constitute crucial institutional preconditions for effectively unleashing the role of New Quality Productivity in promoting high-quality development in coastal economies. In contrast, innovation ecology (IE) demonstrates a significantly positive moderating effect on the relationship between NQPFC and HQMED. The interaction term coefficient for NQPFC × IE is 0.107 in column (3), significant at the 1% level. This suggests that optimizing and improving the urban innovation environment can effectively enhance the recombinant and transformative capacity driven by New Quality Productivity, thereby amplifying its positive contribution to the economic quality of coastal regions. These findings robustly support Hypothesis 4.

6.5. Heterogeneity Analysis: Geography and Institutions at Play

6.5.1. Differential Impacts Across the Three Major Marine Economic Circles

The above results indicate that NQPFS and NQPFC can enhance the level of HQMED in China’s coastal regions. However, given the significant differences in geographic characteristics, marine resource distribution, and marine technological development levels across China’s coastal regions, the effect of New Quality Productive Forces spatial flow on the marine economy also exhibits heterogeneity. Geographically, China has formed three major marine economic zones: northern, eastern, and southern. In these different marine economic circles, distinct and advantageous industrial clusters are emerging, driven by the concentration of key marine factors, laying a foundation for coastal HQMED. Based on national strategies, we divide the sample into the three major marine economic circles-northern, eastern, and southern-for grouped regression analysis. The results are shown in

Table 12. Heterogeneity Test Results for the Three Major Coastal Economic Circles.

	(1)	(2)	(3)	(4)	(5)	(6)
	Northern	Eastern	Southern	Northern	Eastern	Southern
NQPFS	−0.009 ***	0.009 ***	−0.001
	(−2.95)	(4.37)	(−0.50)
NQPFC				0.008 **	0.057 ***	0.003
				(2.32)	(6.64)	(0.70)
In DEV	−0.007	−0.031 *	−0.006	−0.009	−0.021	−0.006
	(−0.82)	(−1.66)	(−1.01)	(−1.04)	(−1.21)	(−1.01)
URB	−0.018	−0.238 ***	−0.061 ***	−0.016	−0.203 ***	−0.062 ***
	(−1.51)	(−5.18)	(−3.37)	(−1.37)	(−4.64)	(−3.40)
IND	0.044 ***	−0.093 *	−0.013	0.044 ***	−0.140 ***	−0.014
	(4.60)	(−1.66)	(−1.09)	(4.59)	(−2.63)	(−1.15)
EDU	0.114 **	0.413 ***	0.170 ***	0.134 ***	0.251 **	0.173 ***
	(2.31)	(3.43)	(4.14)	(2.69)	(2.13)	(4.21)
RD	0.460 ***	0.210	0.683 ***	0.430 ***	0.102	0.677 ***
	(3.67)	(0.72)	(7.88)	(3.40)	(0.38)	(7.74)
_cons	0.086	0.528 ***	0.119 *	0.097	0.414 **	0.117*
	(0.93)	(2.72)	(1.83)	(1.03)	(2.25)	(1.78)
City Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
N	340	220	500	340	220	500
Adj. R2	0.834	0.896	0.866	0.834	0.896	0.866

Note: t-statistics in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

.

Based on the regression results in Table 12, the impact of NQPFS and NQPFC on the high-quality development of the coastal economy exhibits significant regional heterogeneity, reflecting the spatial differentiation of economic development across China’s coastal regions. Specifically, NQPFS shows a significant negative effect in the northern marine economic circle (coefficient: −0.009, significant at the 1% level), which aligns with the observed weak attractiveness and net outflow of new-quality factors in the north, suggesting a possible “backwash effect” whereby factor inflows fail to translate into local productivity gains. In contrast, NQPFS demonstrates strong positive geographical agglomeration benefits in the eastern circle (coefficient: 0.009, significant at the 1% level), indicating efficient factor attraction and strong localization capacity. Its impact in the southern circle remains statistically insignificant, pointing to underdeveloped mechanisms for trickle-down effects or spatial diffusion. In comparison, NQPFC exerts significantly positive effects in both the northern and eastern circles, with the strongest impact observed in the east (coefficient: 0.057, significant at the 1% level). This underscores the pivotal role of spatial allocation efficiency in bridging regional development gaps. The positive response in the northern circle (coefficient: 0.008, significant at the 5% level) suggests that improving factor combination efficiency can partially compensate for the outflow of spatial flows. The statistically insignificant coefficient in the south may reflect constraints from regional lock-in or path dependency.

In summary, new-quality productivity forces drive marine economic development with a distinct core–periphery structure and pronounced spatial heterogeneity. Policy implications are threefold: the eastern circle should consolidate its geographical agglomeration advantages and strengthen spatial spillover effects; the northern circle urgently needs to reverse net factor outflow, break path dependency, and enhance local anchoring through institutional innovation; the southern circle should focus on clearing spatial channels for factor flows and optimizing regional division of labor networks. These measures will help foster a synergistic, complementary, and efficiently interconnected new pattern of marine economic geography at the national scale.

6.5.2. Divergent Effects Between Core Cities and General Cities

Drawing on institutional economics and growth pole theory, China’s coastal cities can be further categorized into core cities (municipalities directly under the central government, sub-provincial cities, provincial capitals) and ordinary cities in terms of coastal HQMED [61]. Benefiting from their higher administrative levels, core coastal cities inherently occupy advantageous positions in policy access and resource allocation, creating a strong “institutional agglomeration effect” that attracts high-end factors such as talent, capital, and technology. Such differential development environments shaped by administrative forces influence NQPFC efficiency. Accordingly, we divide the 53 coastal cities into two groups: (1) core cities, including Shanghai, Tianjin, Dalian, Ningbo, Qingdao, Shenzhen, Xiamen, Fuzhou, Haikou, Guangzhou, and Hangzhou, totaling 11 cities; and (2) the remaining 42 cities, classified as ordinary cities. The regression results are presented in

Table 13. Regional Heterogeneity Analysis: Core Cities vs. General Cities.

	(1)	(2)	(1)	(2)
	Core Cities	General Cities	Core Cities	General Cities
NQPFS	0.009 ***	−0.002
	(3.92)	(−1.62)
NQPFC			0.029 ***	0.001
			(3.12)	(0.15)
In DEV	−0.058 ***	0.005 *	−0.064 ***	0.005 *
	(−3.14)	(1.77)	(−3.38)	(1.76)
URB	−0.201 ***	−0.020 ***	−0.196 ***	−0.019 **
	(−5.46)	(−2.69)	(−5.21)	(−2.55)
IND	0.022	−0.008	0.010	−0.009
	(0.64)	(−1.55)	(0.28)	(−1.62)
EDU	0.053	0.002	0.037	0.002
	(0.35)	(0.08)	(0.24)	(0.12)
RD	0.314	0.291 ***	0.170	0.293 ***
	(1.39)	(6.05)	(0.75)	(6.07)
_cons	0.897 ***	−0.003	0.955 ***	−0.004
	(4.13)	(−0.11)	(4.31)	(−0.13)
City Fixed Effects	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes
N	220	840	220	840
Adj. R2	0.878	0.828	0.878	0.828

Note: t-statistics in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

.

Based on the regression results in Table 13, the estimated coefficient of NQPFS in core cities is 0.009, significant at the 1% level, indicating that NQPFS strongly drives high-quality marine economic development in these cities. In ordinary cities, however, the coefficient of NQPFS is negative and statistically insignificant (−0.002), suggesting that constrained by limited absorptive capacity and institutional barriers, the potential of spatial factor flows has not been effectively realized. For NQPFC, its estimated coefficient in core cities is 0.029, significant at the 1% level, while in ordinary cities it is negligible and statistically insignificant (0.001). This reveals that core cities, leveraging their administrative advantages and mature institutional systems, have enhanced their capacity to capture and recombine factors. In contrast, ordinary cities face a structural contradiction between the net outflow of high-end factors and insufficient supply of innovative elements under the siphon effect of core cities.

These findings indicate that core cities should further exert the radiating and driving functions of growth poles, while ordinary cities need to break away from dependent development lock-in through institutional innovation and the creation of distinctive pathways, thereby reshaping their comparative advantages and embedded positions within regional synergy. The results support Hypothesis 5.

7. Discussion and Limitations

7.1. Discussion

This study examines the Spatial Flow and Optimal Combination of New Quality Productive Force on High-quality Economic Development of Coastal Regions, revealing the core role of dynamic factor allocation in driving the transformation and upgrading of regional marine economies. It provides new empirical evidence for understanding the patterns of productivity distribution and spatial economic development in the Chinese context. The findings demonstrate that NQPFS and NQPFC are central mechanisms driving coastal HQMED. This aligns with Romer’s classical growth theories emphasizing factor accumulation [25] and evolutionary economic geography, focusing on knowledge recombination, thereby enriching our understanding of this frontier concept in the Chinese context. We empirically validate the transmission chain of “NQPFS → MIC → HQMED” and “NQPFC → LnNBE → HQMED.”This not only confirms Schumpeter’s [9] assertion that “innovation is a new combination” but extends it from the micro to the meso and macro levels, revealing the complete pathway through which New Quality Productive Force promotes regional development by empowering enterprises and enhancing the marine industrial chain resilience.

The significant heterogeneity revealed in this study—stronger effects in the eastern region and core cities, exhibiting a Matthew effect—has implications for policy. It corroborates theories suggesting that institutional environments shape economic performance [62] and reflects the explanatory power of the “core-periphery model” of Krugman [12]. The eastern region demonstrates a stronger capacity for factor absorption, while core cities leverage administrative advantages to form an “institutional agglomeration effect.” This divergence necessitates policies that move beyond a “one-size-fits-all “approach, emphasizing regional differentiation and institutional synergy. The phenomenon of high-end factors agglomerating toward institutional and innovation hubs is universal. For instance, socioeconomic activities in Portugal’s coastal areas are highly concentrated in the northern and Algarve regions. These areas, by leveraging superior infrastructure and policy support, achieve faster growth [20]. This observation reinforces our finding regarding the “institution-led coordinated agglomeration of factors” in China’s coastal cities, underscoring a common link between institutional quality and efficient factor allocation across contexts. Furthermore, the significant moderating effect identified in this study highlights the critical role of institutional context and market integration during the transition from factor-driven to innovation-driven growth—a shift enabled by new quality productivity. This mechanism finds parallels in advanced marine economies. For instance, the sustained advantages of clusters in the U.S. Gulf Coast or the European North Sea are sustained not merely by resource endowment but through synergistic R&D and “industry–university-government” networks that continuously upgrade technologies and industrial chains—a mechanism akin to the “optimization and synergy effect of new-quality factors” we identify in China’s eastern marine economic circle [63,64]. This cross-context consistency highlights that fostering a New Quality Productive Force requires more than strategic top-down planning; it necessitates complementary micro-foundations, including a supportive and forward-looking institutional environment for innovation and the reduction in unnecessary regional market protectionism, to fully stimulate grassroots innovation vitality. Additionally, attention must be paid to the degree of environmental and resource misallocation.

7.2. Limitations

There are still some limitations in this paper that are worth further discussion.

First, although the composite indices for new quality productive forces and high-quality marine economic development constructed in this study are based on authoritative statistical sources, they may not fully capture certain intangible and rapidly evolving dimensions of “quality,” such as those related to sci-tech talent, digitalization, and the green marine economy. Future research could integrate alternative data—for example, from satellite remote sensing, web scraping, or corporate disclosures—to develop more dynamic and granular indicators. Second, while this paper attempts to examine the impact of the spatial flow and optimal combination of new quality productive factors on the high-quality development of China’s coastal economy, it does not explore the interactive relationship between spatial flow and optimal combination from the perspective of new quality productive forces. This constitutes a limitation of the study. Future work will employ principal component analysis and the Solow-Swan model to further investigate the underlying relational mechanisms, thereby deepening the theoretical foundations of the topic. Third, this paper applies a two-step mediation approach to examine the mediating roles played by the emergence of new quality marine enterprises and the resilience of the marine industrial chain. However, whether marine industrial spatial layout and marine industrial structure upgrading represent more critical mediating transmission channels remains untested. Future research should further dissect the potential chain mediation mechanisms inherent in these pathways. Finally, the dataset used in this study is current only through 2023 and thus does not capture several pivotal subsequent developments—notably breakthroughs in generative artificial intelligence and the establishment of China’s National Data Administration—which are actively reshaping the constitutive dimensions of new quality productive forces.

8. Conclusions

China’s maritime endeavors are crucial to the nation’s survival and development, bearing on the rise and fall as well as the security of the Chinese nation. New Quality Productive Forces are creating new opportunities for promoting the rational spatial layout of the human-environment system and achieving high-quality development in China’s coastal regions. The theoretical analysis in this paper demonstrates that the spatial flow of New Quality Productive Forces drives the reallocation of innovative factors through “re-territorialization”, resolving contradictions in factor combinations and enhancing factor allocation efficiency, thereby generating “coordinated agglomeration” effects in both factor-inflow regions (early-developed) and factor-outflow regions (late-developed). Meanwhile, the optimal combination of New Quality Productive Forces influences high-quality development in coastal areas through the “internal replacement” of old and new productive forces and the “structural deepening” brought about by the emergence of new production factors. Through this research, the paper derives the following main conclusions:

(1)

The high-quality development level of China’s coastal regions shows a continuous upward trend, yet with marked regional disparities, forming a spatial pattern of “one core, two wings” characterized by “Eastern leadership with Northern and Southern regions following.” The development gap between cities has gradually widened, and the overall spatial structure is evolving from a “core-periphery” single-point model toward a clustered stage of “one core, multiple poles, and networked linkage.” new quality productive forces has transitioned from initial single-point agglomeration to multi-polar distribution, eventually forming a networked pattern, reflecting the dynamic deepening of factor mobility and synergy across coastal areas.

(2)

The baseline regression results indicate that both the spatial flow and optimal combination of new quality productive forces exert stable positive effects on high-quality economic development in coastal regions. The marginal contribution of factor optimal combination is significantly higher than that of spatial flow, highlighting the critical role of improving factor allocation efficiency at the current stage. The significantly positive coefficients of education investment and R&D expenditure further affirm the fundamental supporting function of human capital and technological innovation in high-quality development. Model diagnostics support the use of a two-way fixed-effects specification, and the gradual increase in adjusted R² demonstrates the model’s strong explanatory power and reasonable variable selection.

(3)

The mechanism tests confirm that new quality productive forces promote high-quality development through two mediation pathways—“enhancing marine industrial chain resilience” and “fostering the emergence of new quality business forms in the marine sector”. Furthermore, resource misallocation exerts a significantly negative moderating effect on the role of factor flow, while the innovation ecology shows a positive moderating effect on the impact of factor configuration. These results highlight the critical boundary conditions imposed by external institutional environments in the high-quality development process.

(4)

Heterogeneity analysis reveals that the impact of new quality productive forces on high-quality coastal development exhibits significant regional and institutional differentiation. Spatially, the effects display a distinct geographic pattern: they are strongly positive in the Eastern marine economic circle, significantly negative for spatial flows yet partially positive for factor combination in the Northern circle, and statistically insignificant in the Southern circle. Institutionally, core coastal cities demonstrate pronounced positive effects from both spatial flows and factor optimization, leveraging their administrative advantages for effective factor capture and recombination. In contrast, ordinary cities show negligible or even negative responses, constrained by limited absorptive capacity and the siphon effects from core cities. These findings underscore that both the institutional environment and geographic foundations are key structural factors shaping the effectiveness and boundaries of new, quality productive forces.