Digital Creative Industries in the Yangtze River Delta: Spatial Diffusion and Response to Regional Development Strategy

Abstract

The digital creative industries have emerged as a critical driver of regional economic transformation, upgrading, and sustainable development. While previous research has primarily focused on creative industry layout and agglomeration in urban areas, with the integration of digital technology and the creative industry, existing research has an insufficient explanation for the digital creative industry. Specifically, few people have studied the spatial distribution and diffusion of digital creative industries in emerging economies from the macro-regional level. To address this gap, this study analyzes the spatial diffusion mode and regional spatial response law of digital creative industries in the Yangtze River Delta during three critical time windows (2016, 2019, and 2022) in the context of national strategy implementation. A range of spatial analysis technologies is utilized to process the full sample of big data from digital creative industries. This study utilizes OLS and a quantile regression model to determine the dominant factors that affect spatial diffusion and response in the digital creative industries. The results demonstrate that, against the backdrop of regional development strategies, digital creative industries exhibit a variety of diffusion modes, including contagious, hierarchical, corridor, and jump diffusion. The response of industries to regional strategies has different rules in terms of regional space, urban development, and sub-industries. Furthermore, the comprehensive influence of institutional environment, urban economy, development and innovation significantly impacts industrial spatial diffusion and regional response. Among them, government investment in science and technology and the number of universities have consistently been important influencing factors, and policy exhibits nonlinear effects and asymmetric characteristics on industry agglomeration and diffusion. This study enhances the understanding of digital creative industry development in the YRD and offers a theoretical basis for optimizing regional industrial spatial structure and promoting the sustainable development of digital industries.

1. Introduction

With the rapid advancement of new technologies, including cloud computing, virtual reality, and 5G, the convergence of creative industries and the digital economy has significantly deepened, and the digital creative industries have emerged and experienced explosive growth [1,2,3,4]. Digital creative industries have multiple attributes, such as high added value, a long industrial chain, and deep integration and have become an important path toward driving consumption upgrading and promoting economic transformation and regional sustainable development [5,6].

Digital creative industries encompass a new economic field produced through the deep integration of digital technologies and creative industries. In academia, there remains a lack of consensus regarding the definition of digital creative industries, as well as the concepts surrounding digital copyright industries [7]. The concepts of digital copyright industries [7], digital content industries [8], and digital cultural industries [9] vary across different countries. However, their industrial domains are fundamentally similar, encompassing sectors, such as publishing, advertising, design, software, film, television, broadcasting, animation, gaming, music, and others [5,10]. The study of digital creative industries can be traced to creative industries, which originated in Australia’s “Creative Country” report in 1994 [11]. The United Kingdom (UK) proposed a “creative economy” in 1997, and the British Government paved the way toward a global creative economy by formulating relevant industrial policies [12]. Today, with the continuous progress of digital technology, digitalization has become inseparable from creative industries [13,14], so the digital creative industries are a new type of economic activity based on digital technology and centered on creative formats, including cultural creation, content production, copyright utilization, animation, and games. They are also regarded as a key driver of the green economy and sustainable transformation.

In recent years, digital creative industry has developed rapidly and has become one of the most dynamic industries in the world economy, providing a new avenue through which developing countries can grow significantly quickly into emerging fields across the world [5]. Developed countries, such as the United States (US), the UK, and Japan, attach great importance to the development of digital creative industries, regarding them as crucial to promoting economic growth, consolidating technological innovation advantages, and expanding cultural influence [8]. China has listed the digital creative industries as a strategic emerging field, having become the primary battlefield and strategic direction in the promotion of the high-quality development of China’s economy [15]. They have also been incorporated into the core policy agenda for advancing regional coordination and sustainable development. According to statistics, in 2017, the output value of digital copyright-related industries in the US was USD 2247.4 billion, accounting for 11.59% of the gross domestic product (GDP). Meanwhile, in 2017, the UK’s creative industries were valued at USD 124.4 billion, accounting for 5.5% of the GDP, and in 2018, the output value of China’s digital creative-related industries reached USD 503.3 billion, accounting for 3.5% of the GDP.

As a new type of converged industry, digital creative industry has impacted the traditional boundaries of creative industry, has promoted the innovation of the creative industry business model, and has been reshaping the development pattern of the city [14,16,17,18]. For this research, on the one hand, the continuous agglomeration of digital creative industries in large cities has promoted the organic renewal of central urban areas [11]; enhanced the development of urban architecture, culture, environment, systems, and other aspects; advanced the urban culture; and driven “urban rejuvenation” [19,20]. On the other, with the continuous development of large cities, the competition and cooperation between creative cities are getting stronger and stronger, and the industries of large cities are gradually spreading and shifting to neighboring areas, forming a close geographical region through connectivity among industrial chains [21]. Some scholars believe that with the continuous development of digital technology, the time and space of creative industry elements will be compressed, and the boundaries of geographical regions will gradually disappear [22]. However, some scholars also believe that geographical areas are still crucial to the development of creative industries [21,23], providing a broad spatial hinterland for industrial development. These views offer a larger spatial perspective for further research, of which the proliferation of digital creative industries within the regional space is an important topic.

Regional industrial diffusion is a dynamic process with both temporal and spatial dimensions. It is an important factor in the interregional industrial division of labor and industrial transformation and upgrading and also a key pathway through which developed core regions promote the development of peripheral areas via diffusion effects. In 1953, Swedish geographer Hägerstrand first proposed the concept of spatial diffusion, after which the diffusion process became a research focus in many disciplines and attracted extensive attention in the field of geography [24]. Economic geographers categorized the basic spatial forms of regional industrial diffusion theory as contagious diffusion and hierarchical diffusion [25,26], where the former refers to the process of enterprise space expansion from near and far and the latter to the diffusion of enterprises, with reference to the scale of the city’s hierarchy. In this line of research, many scholars have built applied models, such as enterprise spatial evolution models, based on firm-level studies [27]. Focusing on the digital creative industry, numerous empirical studies have examined the spatial diffusion and structural evolution of sectors, including cultural creativity, design services, film and television, and animation and gaming. Research on the cultural and creative industries has shown that these industries tend to cluster in the core areas of large cities [28], but there is a noticeable trend of relocation to neighboring regions [29]. This phenomenon reflects a significant proximity-based diffusion effect, whereby an increase in the number of cultural and creative industry (CCI) firms in a core area promotes growth in the number of such firms in adjacent regions [30]. Other studies have found that subsectors such as digital content and creative design do not follow proximity-based diffusion. Instead, under policy interventions, they diffuse directly from urban centers to peripheral areas, creating new economic clusters and media cities in these peripheral regions [31]. In addition, many studies, using the film industry as an example, have found that film production centers are shifting from traditional hubs such as Hollywood in the United States to peripheral and semi-peripheral regions in Europe, Asia, and Australia [32,33,34,35,36]. However, these studies focus on single subsectors of the digital creative industry and fail to draw definitive conclusions on the spatial diffusion of the industry as a whole. For a complex and integrated industry such as the digital creative industry, the spatial diffusion of individual subsectors cannot accurately reflect the diffusion characteristics of the entire industry, particularly at a large regional scale.

Since the late 20th century, China’s rapid urbanization has fueled remarkable economic growth [36]. However, unbalanced regional development persists [37], which undermines the sustainability and coordinated progress of regional systems. In China, government policies play a pivotal role in industrial agglomeration and urban growth [38]. Through the implementation of regional integration development strategies and industrial guidance policies, the Chinese Government can further promote industrial agglomeration and diffusion, narrow regional development gaps, and promote coordinated regional development. Among these, the spatial diffusion of the digital creative industry provides a new path for promoting regional integration. By optimizing the spatial allocation of industrial factors, this spatial diffusion helps build a more balanced, resilient, and sustainable regional development pattern. However, existing studies have mainly examined the spatial diffusion of individual subsectors of the digital creative industry itself, without placing them in a comprehensive regional context and have seldom analyzed the spatial diffusion of the industry as a whole or its responses to regional development strategies.

In the context of national regional development strategies, what is the spatial diffusion mode of regional digital creative industries? How does digital creative industry respond to the national regional development strategy? Why is there such a pattern of spatial diffusion? The existing research on digital creative industries is relatively fruitful, but few people have conducted such comprehensive studies. Therefore, from the perspective of geography, this study deconstructs the spatial diffusion mode of digital creative industry against the backdrop of national development strategies on a regional scale using geospatial analysis methods, as well as studying the response law of digital creative industries to national development strategies, and it explores the factors influencing their spatial diffusion, seeking to fill this research gap. The research can provide valuable references and insights for countries and regions aiming to strengthen transformation and upgrades to the digital industry through the introduction of relevant policies, thereby facilitating regional coordination and sustainable development.

The structure of this paper is as follows. The next section introduces the theoretical framework of the research, followed by the research methodology and data processing. The empirical research is then divided into three parts: (1) the spatial diffusion mode of digital creative industries in the YRD; (2) the response law of digital creative industries to the regional development strategy; and (3) factors influencing the spatial diffusion of the digital creative industry. Finally, this paper summarizes the research results, presents the theoretical and practical contributions of the research, and discusses the research shortcomings and next steps.

2. Conceptual Framework

Digital creative industry strengthens the connection between regional industries through digital technology, and it promotes the change to regional industrial chains and urban networking [39]. Endowment conditions, such as location, transportation, population, and facilities, affect the spatial distribution of digital creative industries in metropolises [40]. However, as market competition intensifies, digital creative enterprises have begun to shift from regional central cities to non-central cities under the influences of cost, competition, and other factors [8]. With continuous improvements in transportation, medical care, education, and other facilities and services, this trend is continually expanding. Through the continuous flow of industry, information, capital, and personnel, central and non-central cities can jointly form a closely linked geographical area and gradually promote the development of underdeveloped areas in non-central cities. Therefore, the agglomeration and diffusion of digital creative industries help promote the efficient allocation of regional resources and elements, narrow regional development gaps, and further advance integrated and sustainable regional development.

The factors influencing the spatial diffusion of digital creative industries are complex, among which policy is important, affecting the spatial agglomeration and development of digital creative industries [8,41]. Policies include regional development strategies and industrial guidance policies. For example, Europe has promoted the development of digital clusters in creative industries through the EU integration development strategy and the introduction of relevant creative industry policies [5]. China has also implemented a regional development strategy that promotes industrial agglomeration and diffusion. On 11 May 2016, China issued the YRD Urban Agglomeration Development Plan, which proposed fostering the growth of emerging industries, with a focus on developing cultural creativity, digital content, and characteristic cultural and creative services. It also encouraged the integration of culture with the internet, science and technology, and finance. On 1 December 2019, China issued the Outline of the YRD Regional Integration Development Plan, which called for joint efforts to build a number of high-level service industry clusters and innovation platforms in fields, such as scientific and technological services, software and information services, e-commerce, and cultural creativity. It also proposed the establishment of several national-level bases for strategic emerging industries. The introduction of national regional development strategies has significantly influenced the spatial diffusion of digital creative industries, expanded their spatial hinterland for development, and provided institutional support for achieving regional coordination and sustainable development goals. Simultaneously, other such factors as the urban economy, the institutional environment, and technological innovation also affected the spatial agglomeration and transfer of digital creative industries, and the close connection among industrial organizations has further promoted the development of regional integration [42].

This study uses several conceptual terms, such as digital creative industry spatial diffusion (DCISD), national regional development strategies, and regional integration development. The primary aim of this research is to explore the spatial diffusion mode of digital creative industries against the backdrop of the regional development strategy and to study the response law of digital creative industries to this strategy using the years 2016 and 2019, when the national major strategy was issued, and 2022 as time points. In addition, through the selection of relevant impact indicators, the effects of urban economy, institutional environment, and development innovation of the spatial diffusion of regional digital creative industries are further explored. To examine the impact of national strategies on the development of the digital creative industry in the Yangtze River Delta (YRD), we selected three key time points: 2016, when the YRD Urban Agglomeration Development Plan was introduced; 2019, when the Outline of the YRD Regional Integration Development Plan was released; and 2022, representing the current stage.

Therefore, this study established a conceptual framework, aiming to (Figure 1) (1) through a spatiotemporal analysis, study the spatial diffusion mode of digital creative industries in the YRD against the backdrop of the regional development strategy; (2) through a response analysis, summarize the response law of digital creative industries in the YRD to regional development strategies; (3) through a regression analysis, determine the factors influencing the spatial diffusion of digital creative industries in the YRD.

3. Materials and Methods

3.1. Study Area

The YRD is located in East China (Figure 2), and it includes Shanghai, Jiangsu, Zhejiang, and Anhui provinces. It is one of the regions with the fastest economic development and the strongest innovation capacity in China, accounting for a quarter of China’s economic aggregate [43].

The YRD city cluster is one of the six major urban agglomerations in the world, the only representative from a developing country, and it enjoys a high international reputation [44]. From the end of the last century to this century, the YRD region has promoted rapid regional development through two transformation modes: industrial economic transformation and regional development transformation.

The first method is the transformation of the YRD from an export-oriented manufacturing economy to an innovation-driven economic model [27]. This is primarily reflected as follows: (1) Since the early 1980s, export-oriented economic industrial development has become an important engine driving the economic growth of the YRD. (2) In the 1990s, the upgrading of modern service industries, represented by the tourism sector, propelled the YRD to become the “Golden Triangle” of China’s tourism industry. (3) Since 2010, the advancement of the digital economy, represented by digital creative industries, has accelerated the transformation of old and new drivers of growth, adding informatization wings and strong new drivers to the high-quality economic development of the YRD. In 2021, the Yangtze River Delta Digital Creative Industry Alliance was established to foster sustainable development of the digital creative industries by mobilizing high-quality resources and cultivating industrial clusters.

The second method is the transformation of regional integration development. The YRD is the first area in China to implement regional integration and coordinated development [45]. On 11 May 2016, the government issued the YRD City Cluster Development Plan, which confirmed the path and mechanism guarantee for the integrated and coordinated development of the YRD city cluster. On 1 December 2019, the government issued and implemented the Outline of the YRD Regional Integration Development Plan, for which the YRD G60 Science and Technology Innovation Corridor is important in terms of cooperation. The release of the national development strategy has promoted the regional transformation of the YRD and promoted the balanced and coordinated development of regional integration.

At present, the world is facing major, unprecedented changes, and the global economic situation is still severe and complex, with geopolitical protectionism on the rise [46]. The YRD must urgently accelerate the transformation of the economic development mode and promote high-quality economic development. Second, as one of China’s economically developed regions, the YRD also faces the problem of unbalanced regional development [35]. Strengthening the research on spatial diffusion and the regional response of digital creative industries in the YRD against the backdrop of regional integration strategy implementation can provide a theoretical reference for the industrial development of similar regions in developing countries.

This study takes the YRD as a case study and, according to the specific scope determined in the Outline of the Regional Integrated Development Plan of the YRD, it includes 4 provinces and municipalities (Shanghai, Zhejiang, Jiangsu, and Anhui provinces) and 41 municipal administrative units (Figure 2) in total, for a combined area of 35.9 km². According to the Chinese administrative division system, it includes one municipality directly under the central government (Shanghai) and 40 prefecture-level cities.

3.2. Research Methodology

3.2.1. Kernel Density Estimation (KDE)

Kernel density estimation (KDE) is used to characterize the spatial density characteristics and distribution trends of the study object, and the results can be used to smoothly identify and represent the agglomeration and dispersion of the samples in the study area [47]. KDE is used to estimate the spatial distribution density of digital creative industries in the YRD. The formula is given as follows:

$$f(x)={\displaystyle \frac{1}{nh}}{\displaystyle \sum _{i=1}^{n}k\left(\frac{x-x_i}{h}\right)}$$

(1)

where: h is the bandwidth; x − xi is the distance between the digital creative enterprise and the measurement punctuation enterprise.

To ensure the scientific validity and rationality of the estimation, the kernel function k(x) must satisfy the following conditions: the domain x ∈ (0, 1), symmetry about the origin, and the normalization condition with an integral equal to 1. Given that the choice of kernel function exerts little influence on the estimation results, this study employed the Gaussian kernel for estimation. The formula is given as follows:

$$K(x)={\displaystyle \frac{1}{\sqrt{2\pi }}}\mathit{exp}\left(-{\displaystyle \frac{x^2}{2}}\right)$$

(2)

Previous studies have shown that the bandwidth h is highly sensitive to the estimation results of the kernel density function [47]. A large bandwidth reduces the variance in the estimation results and produces smoother curves, but it may obscure local fluctuations in the data and affect estimation accuracy. A small bandwidth increases the variance and enhances curve fluctuations, yet it can reflect the actual distribution structure of the data in greater detail. Theoretically, when h is considered as a function of the sample size n, it should satisfy h(n)→0 and nh(n)→∞ as n→∞ to ensure estimation consistency. Considering the large sample size in this study and the requirement for computational efficiency in kernel density estimation, the empirical rule (Rule of Thumb) proposed by Silverman was adopted for bandwidth selection. Based on the practical situation of the digital creative industries in the Yangtze River Delta and after multiple adjustments, a fixed bandwidth of 3000 m was chosen for ease of comparative analysis.

3.2.2. Policy Effectiveness Model (PEM)

Policy effectiveness refers to the content validity and influence of the policy text. We establish a quantitative evaluation model of policy text from the four dimensions of Policy Power, Policy Goal, Policy Method and Policy Feedback [48] and evaluate the effectiveness of policies in the field of digital creative industry. After determining the criteria for policy effectiveness scoring, from 5 to 9 August 2025, we invited six experts from universities, businesses, and government sectors specializing in digital creative industry policies and public policies to be divided into three groups for synchronous scoring. After the first round of scoring, the results will be compared. When there are differences, we will discuss and determine them together and finally complete the scoring of each policy document. The formula for the policy effectiveness model is:

$$\begin{array}{l}PE{M}_i={\displaystyle \sum _{j=1}^N(m_j}+b_j+f_j)p_j\\ APE{M}_i={\displaystyle \frac{{\displaystyle \sum _{j=1}^N(m_j}+b_j+f_j)p_j}{N}}\end{array}$$

(3)

where: i is the year of policy implementation; N is the policy effectiveness implemented in year i; j is the jth policy implemented from the i-th year; (m_i + b_i + f_i) is the score of Policy Method, Policy Goal and Policy Feedback of policy; p_i is the score of policy power of policy j; PEM_i is the overall effectiveness of the policy of creative industry in the i-th year; APEM_i is the average effectiveness of the policy of digital creative industry in the i-th year.

3.2.3. Ordinary Least Squares (OLS)

The OLS model is widely used to study the factors influencing the spatial distribution of enterprises [49]. In this study, OLS was used to analyze the impact of the urban economy, institutional environment, development, and innovation on the spatial diffusion of digital creative industries. The formula is given as follows:

$$Y={\beta }_0+{\beta }_1{X}_1+{\beta }_2{X}_2+\dots \dots +{\beta }_n{X}_n+\epsilon$$

(4)

where: x_n is the observed value of the nth factor affecting the spatial diffusion of digital creative industries; β_n (n = 1, 2, 3, 4, 5, 6) is the regression coefficient of the influencing factors; ε is a random interference term of the influencing factors.

3.2.4. Quantile Regression Model (QRM)

Quantile regression estimates the relationship between a set of explanatory variables X and the dependent variable Y at different quantile points (0, 1) by minimizing the asymmetrically weighted absolute residuals. Compared with the ordinary least squares (OLS) model, it is less affected by outliers, offers stronger estimation robustness, and requires less restrictive assumptions on the error term [50]. This method enables the examination of detailed distributional characteristics in the data, thereby uncovering the underlying mechanisms that link variable distributions to spatial patterns. By contrast, OLS estimates the conditional mean of the dependent variable by minimizing the sum of squared residuals and is suitable when the error term follows a normal distribution and the data meet linearity assumptions. In empirical studies of industrial spatial patterns, however, the effects of explanatory variables on the dependent variable often exhibit heterogeneity and may differ across development levels. In this study, the large-sample data on digital creative enterprises in the Yangtze River Delta exhibit skewness and local extreme values, and the analysis focuses on capturing heterogeneous effects across development levels. Therefore, quantile regression was adopted as a complement and extension to OLS to enhance the completeness and explanatory power of the estimation results. The formula is given as follows:

$${\mathrm{y}}_i={x^{\prime }}_i{\beta }_{\theta }+{\epsilon }_{\theta i},Quan{t}_{\theta }\left({\epsilon }_{\theta i}|x_i\right)=0,i=1,2,\dots ,n$$

(5)

where: y_i is the vector of the dependent variable; x_i is the vector of explanatory variables; θ is the estimated quantile; β_θ is the estimated regression coefficients as the quantile varies within the interval (0, 1); n is the sample size.

3.2.5. Robustness and Sensitivity Tests

(1).

Kernel Density Estimation Test with Variable Bandwidths

To examine the sensitivity of the kernel density estimation results to bandwidth settings, this study used h = 3000 m as the baseline and compared it with two alternative bandwidths, h = 2000 m and h = 5000 m. The Gaussian kernel was adopted, and the estimation formula is given as follows:

$$\widehat{f}(s)={\displaystyle \frac{1}{n{\hslash }^2}}\sum _{i=1}^{n}K\left({\displaystyle \frac{\Vert s-s_i\Vert }{\hslash }}\right)$$

(6)

where: s denotes the spatial location and s_i represents the i-th enterprise location. The spatial diffusion trends of the digital creative industry were compared under different bandwidths h to evaluate the robustness of the conclusions drawn from the kernel density estimation.

(2).

Ridge Regression as an Alternative Estimation

Since OLS and quantile regression may produce unstable parameter estimates when affected by multicollinearity or extreme values, ridge regression was employed as an alternative estimation to verify whether the signs and significance levels of the coefficients remain consistent. Ridge regression alleviates multicollinearity and improves generalization stability through an L₂ penalty term. The formula is given as follows:

$${\widehat{\beta }}^{\mathrm{r}\mathrm{i}\mathrm{d}\mathrm{g}\mathrm{e}}=\mathit{arg}\underset{\beta }{\min }\left\{\sum _i(y_i-x_i^{\top }\beta {)}^2+\lambda \sum _j{\beta }_j^2\right\}$$

(7)

where: λ ≥ 0 is selected via K-fold cross-validation; in this study, 10-fold cross-validation was used to determine λ.

(3).

Lag Effect Test

This study introduced the policy effectiveness variable X4. As there is often a transmission period from policy formulation and release to actual implementation, a lag effect model was constructed to verify the robustness of the results by replacing the contemporaneous X4 in the baseline OLS and quantile regression models with its one-period lag (X4_lag1). Due to the lack of prior-period observations for 2016, the lag test was only conducted for 2019 and 2022. To further enhance robustness, the lag test was repeated using pooled samples with year effects controlled.

The lag effect model for OLS is:

$$\begin{array}{c}{Y}_{it}={\beta }_0+{\beta }_1{X}_{1,it}+{\beta }_2{X}_{2,it}+{\beta }_3{X}_{3,it}+\gamma X_{4\_lag1,it}+{\beta }_5{X}_{5,it}+{\beta }_6{X}_{6,it}+{\mu }_i\hfill \\ +{\tau }_t+{\epsilon }_{it}\hfill \end{array}$$

(8)

where: Y_it is the dependent variable; X_1,it − X_6,it are control variables; X4_lag1_it is the one-period lag of the policy effectiveness variable X4; μ_i represents individual fixed effects; τ_t denotes year fixed effects; and ε_it is the random disturbance term.

The lag effect model for quantile regression is:

$$\begin{array}{cc}{Q}_{\tau }(Y_{it}\mid X)=\hfill & {\beta }_0\left(\tau \right)+{\beta }_1\left(\tau \right)X_{1,it}+{\beta }_2\left(\tau \right)X_{2,it}+{\beta }_3\left(\tau \right)X_{3,it}\hfill \\ & +\gamma \left(\tau \right)X_{4_{-}lag1,it}+{{\beta }_5\left(\tau \right)X_{5,it}+\beta }_6\left(\tau \right)X_{6,ii}+{\tau }_{t,\tau }\hfill \end{array}$$

(9)

where: Qτ(Yit∣X) denotes the conditional quantile function of Y_it given quantile τ, γ(τ) measures the impact of the lagged policy effectiveness at different quantiles, and τ_t(τ) is the coefficient of the year dummy variables at quantile τ.

3.3. Data Source and Processing

Relevant studies have shown that the spatial diffusion of enterprises can reflect spatial changes in industry, urban, and regional development [38]. Many cities in China regard the number of enterprises as an important indicator of industrial development [51]. The development of regional integration, urban renewal, and industrial iteration in cities in the YRD region has been rapid, highlighted by the increase in digital creative enterprises. Therefore, the number of enterprises is utilized to analyze the spatial diffusion characteristics of industries against the backdrop of regional development strategies. The existing research primarily analyzes the spatial agglomeration of industries using data, such as digital creative listed companies and state-owned enterprises [23], but it cannot reflect the real development of regional industries. With the ongoing advancements in technology, big data methodologies offer new avenues for our research.

Qixinbao (https://www.qixin.com/, accessed on 1 March 2023.) is a well-known enterprise information inquiry website in China, providing users with a real-time search tool for massive enterprise information services. Using network crawler technology, from March to September 2023, we climbed all the data of enterprises in 41 cities in the YRD, totaling 21 million+data points, including name, address, classification (national economic sectors), registered capital, number of operators, and other information. According to the classification standards of digital creative industries, such as “Classification Standard of China’s Strategic Emerging Industries: Digital Creative Industry (2018)” (https://www.gov.cn/zhengce/zhengceku/2018-12/31/content_5433037.htm, accessed on 1 March 2023) combined with the classification function of Qixinbao, the digital creative industries studied were classified into eight categories: digital creative technology equipment manufacturing, creative software engineering, digital content production services, new media services, radio and television services, other creative activities, digital design, and digital creativity and integration (

Table 1. Classification of the digital creative industry.

Digital Creative Sector	Subsector	Corresponding National Economic Sectors
Digital Creative Technology Equipment	Digital creative technology equipment manufacturing	Film machinery, broadcasting and video equipment, professional audio, smart consumer devices manufacturing
Digital Cultural and Creative Activities	Creative software engineering	Application software development
	Digital content production services	Animation and gaming content services
	New media services	Other digital content services, digital publishing
	Radio and television services	Cable and wireless TV transmission services
	Other creative activities	Geographic remote sensing, program production, performing arts
Design Services	Digital design	Engineering, urban planning, and professional design services
Digital Integration Services	Digital creative and integration	Online advertising, tourism exhibition services, travel agency services

Data sources: the National Bureau of Statistics of China.

). It is noteworthy that the Chinese Government has incorporated the manufacturing of digital creative technology equipment into the digital creative industry system, as this sector provides the foundation for realizing digital creative content and plays a key role in promoting the integration of creative design and digital technology. Based on this classification standard, we matched and cleaned each firm’s national economic industry classification with the corresponding categories defined by the National Bureau of Statistics, removing enterprises unrelated to the digital creative industry as well as those with incomplete or abnormal records. In total, 923,759 digital creative enterprises were included in this study.

The data cleaning workflow and quality control procedures were as follows:

(1).

Industry classification. Qixinbao classified enterprises according to the national economic industry classification standard (GB/T 4754-2017 [52]) issued by the National Bureau of Statistics of China. Based on the industry category registered with market supervision authorities and the scope of business operations, the main industry of each enterprise was determined through unified matching rules. For enterprises engaged in multiple business activities, the primary business was used to determine the main industry, ensuring consistency and rationality in classification. Using Qixinbao’s classification data, we matched the collected data one to one with the large, medium, and broad industry categories defined by the National Bureau of Statistics for the digital creative industry, ultimately identifying the number of enterprises in eight major categories.

(2).

Outlier detection. Variables such as registered capital that were far above the upper limit of the normal range for the industry or far below the industry average were considered potential outliers and removed. The proportion of missing data for each variable was also checked. If a field had more than 30% missing data and key information such as address, establishment date, and industry classification was missing, it was treated as an abnormal record and removed.

(3).

Duplicate removal. To avoid duplication, each enterprise’s registration information was identified by a unified social credit code, which was verified via official government websites. If two or more records shared the same identifier and had identical name, address, and other information, they were considered duplicates, and only the most complete and accurate record was retained.

(4).

Quality control. Data quality checks were first conducted to ensure consistency of information for the same enterprise across multiple fields (e.g., name, business scope, registered capital). Reasonable value ranges were defined, such as non-negative registered capital and a business scope consistent with the industry classification, to detect unreasonable data, which were removed. Finally, manual verification and sampling were performed by randomly selecting samples from the cleaned dataset to ensure that the standards and rules were applied accurately during the cleaning process.

Geographical coordinates of each digital creative enterprise were obtained using the Baidu Maps API and further verified, screened, and filtered according to the spatial extent of the Yangtze River Delta to remove invalid samples. Finally, ArcGIS 10.8 was used to convert the coordinates of 923,759 digital creative enterprises across 41 cities in the Yangtze River Delta into point vector data. Some partial descriptive statistics are shown in

Table 2. Sample profile.

Indicator	Abbreviation	Unit	Minimum	Maximum	Mean
Longitude	LON	Degree	114.94	122.76	119.88
Latitude	LAT	Degree	26.67	35.09	31.36
Opening years	YOE	Year	1	32	6.50
Registered capital	REC	Million	0.01	99,990	1850.73
Area code	ARC	Number	310,000	341,881	325,940.50
Valid Sample No.	923,759

Data sources: Qixinbao (https://www.qixin.com/, accessed on 1 March 2023.).

.

Table 2 shows several characteristics of the digital creative industry. First, digital creative enterprises are widely distributed in space. Digital creative enterprises in the YRD are not limited to a certain city or region. Second, the average number of operating years of digital creative companies is 6.5, indicating that most companies were registered after 2015. This shows that after the release of the regional strategy in 2016, the digital creative industry developed rapidly.

Regarding the influencing factors of the spatial diffusion of the digital creative industry, previous studies have shown that digital creative enterprises, as knowledge-intensive tertiary-sector firms, are closely linked to urban industrial structures. The proportion of the tertiary industry in GDP reflects structural shifts and provides diverse services that facilitate the growth of digital creative industries [40]. Meanwhile, the level of urbanization lays the foundation for digital creative development, making urban economic strength a key factor influencing firm distribution [40]. The institutional environment also plays a crucial role in attracting firms, as the development of creative industries often depends on fiscal support from local governments. In China’s administrative context, industrial policies and regional strategies significantly shape the spatial direction and distribution of digital creative enterprises [41,53]. Fundamentally, the digital creative industry emerges from the deep integration of creative sectors and digital technologies. Indicators such as the number of universities and patent applications reflect regional innovation capacity and provide essential support for digital creative development [54]. Therefore, we selected three dimensions—urban economy, institutional environment, and development and innovation—to investigate the factors influencing the spatial diffusion of the digital creative industry (

Table 3. Factors affecting the spatial diffusion of digital creative industries.

Type of Variable	Influence Factor	Conventional Letter	Index Interpretation
Urban economy	Industrial structure	X1	Share of tertiary industry in the GDP
	Urbanization level	X2	Urbanization rate
Institutional environment	Government support	X3	Government investment in science and technology
	Policy guidance	X4	Policy effectiveness
Development and innovation	Innovation support	X5	Number of universities in the city
	Innovation ability	X6	Number of patents authorized in the city

Data sources: the National Bureau of Statistics of China, Statistical Yearbooks of Each City.

). The data are sourced from the National Bureau of Statistics of China, “China Science and Technology Statistical Yearbook,” “China Statistical Yearbook,” and “China Urban Statistical Yearbook.” To address the large differences in magnitude among variables, this study applied a logarithmic transformation to the relevant variables.

Further, within the institutional environment, to promote industrial development, local governments have introduced numerous redundant and even convoluted industrial policies, leading to a vast policy framework. However, we remain uncertain about the actual effects of these policies on industrial development. Existing research and analyses of the impact of policies on digital creative industries are mostly qualitative [41]. We start with the content and characteristics of policy texts, establish an evaluation model, and quantitatively analyze and evaluate the effectiveness of government policies guiding the development of digital creative industries. By 2022, the research systematically combed 573 policies related to digital creative industries from the official websites of the national government, relevant ministries and commissions, as well as municipal governments in the Yangtze River Delta, and built a policy library of digital creative industries. Drawing on methods from relevant experts [48], we rate the Policy Power, Policy Goal, and Policy Feedback on a scale of 1 to 5 (

Table 4. Quantitative criteria of Policy Power, Policy Goal and Policy Feedback.

Project	Assignment and Scoring Criteria
Policy Power	5	Laws promulgated by the National People’s Congress and its Standing Committee;
	4	Regulations, regulations, and decisions issued by the State Council;
	3	Provisional regulations, plans, decisions, opinions, methods, and standards issued by various ministries;
	2	Opinions, methods, plans, guidelines, temporary regulations, detailed rules, conditions, and standards issued by provincial governments and departments;
	1	Notices, announcements, plans and measures issued by municipal governments and departments;
Policy Goal	5	The Policy Goals are clear and quantifiable, with clear indicators and quantities, and clear numerical standards;
	3	Policy Goals are clear, but there are no quantitative criteria;
	1	Just a macro way to describe the policy vision and expectations;
Policy Feedback	5	There are clear supervision methods and responsible departments, and regular feedback documents;
	3	There is a clear supervision method and a responsible department, but there is insufficient feedback;
	1	No oversight or feedback was available.

Note: In order to facilitate our understanding and grasp of the quantitative standards of the scorer, the Policy Goal and Policy Feedback give three obvious differences of 5,3 and 1 points, and 4-2 points are between the adjacent standard scores.

). Considering the varying emphases of different Policy Methods, we categorize them into three types of policy instruments based on their commonalities [48], assigning scores from 1 to 5 accordingly (

Table 5. Comparison of three different policy instruments.

Type of Policy Tool	Behavior Dynamics	Behavior Restriction	The Degree of Coercion	Main Form of Expression
Command-and-control Policy	Administration of government power	Laws and regulations, norms, systems and other constraints	High	Laws, regulations, mandatory standards, norms, plans, etc.
Economic-incentives Policy	Economic interests closely related to industrial development	Economic cost, financial strength and industrial development budget constraints	Middle	Price policy, subsidy policy, tax policy, talent policy, land policy, etc.
Environmental-impact Policy	The factors affecting the development environment, indirectly affect the industry	Other development environment constraints	Low	There are policies related and cross forms to the industry

and

Table 6. Quantitative criteria of Policy Method for different policy instruments.

Type of Policy Tool	Assignment	Scoring Criteria
Command-and-control Policy	5	Established mandatory entry conditions, thresholds, and standards for enforcement; formulated measures for assessment, inspection, supervision and inspection related to industrial development; formulated mandatory management measures, opinions, plans, etc., to promote industrial development;
	3	Clearly require the entry conditions, thresholds, and standards of the industry; clearly require the implementation of industrial development related assessments, supervision and inspection; clearly requiring the formulation of relevant policies or systems to promote industrial development, but no relevant plans have been formulated;
	1	The government has relatively loose control over industrial development, only mentioning the clauses in the 5-point and 3-point evaluation criteria for command and control policy tools;
Economic-incentives Policy	5	Provide strong support in finance, finance, land, talent, taxation, recruitment, and rewards, and propose the amount or support method of rewards;
	3	it is clearly proposed to provide strong support in finance, finance, land, talent, taxation, recruitment, and rewards, but no relevant implementation measures or measures have been formulated;
	1	Only mention or involve clauses in the 5-point and 3-point evaluation criteria for economic incentive policy tools;
Environmental-impact Policy	5	Policies have been introduced to promote the development of industries that are highly overlapping with this industry type, providing strong support in finance, finance, land, talent, taxation, recruitment, and rewards, and proposing the amount of rewards or support methods;
	3	Clearly propose to provide strong support in finance, finance, land, talent, taxation, recruitment, and rewards for industries that have a high degree of intersection with this industry type, but no relevant implementation measures or measures have been formulated;
	1	Only mention the clauses in the 5-point and 3-point evaluation criteria for environmental impact policy tools.

Data sources: The official website of the Chinese Government.

).

To more intuitively reflect the spatial diffusion mode of digital creative industries in the YRD against the backdrop of the regional development strategy, the research takes 2016, 2019 and 2022 as time points and uses the ArcGIS10.8 software to draw a distribution map of the current situation of digital creative industries in 3D (Figure 3), the KD evolution (h = 3000, Figure 4), the SDE and WMC evolution (Figure 5), trend of the number of enterprises (Figure 6), the KD evolution (h = 2000/5000, Figure 7). Figure 3 shows the spatial distribution of digital creative industries. In Figure 4, the KD evolution map provides valuable observations of the geospatial diffusion of digital creative industries. In Figure 5, the SDE evolution shows that digital creative industries are gradually changing from east to southwest and northwest. The evolution chart of the WMC shows that the overall trend of digital creative industries and regional development strategies is obvious, and there is a certain coupling relationship. Figure 6 shows the changing trend in the number of digital creative industries in each city. Figure 7 shows the KD evolution of the digital creative industry under different bandwidths.

4. Results

4.1. The Spatial Diffusion Characteristics of Digital Creative Industry

4.1.1. Before 2016: The Diffusion Characteristic Dominated by Contagious Diffusion

Figure 3, Figure 4 and Figure 8 show that the digital creative industry in the YRD spatially formed an initial “polar core” diffusion pattern centered on Shanghai in 2016. During this period, the spatial diffusion of digital creative industries presented the characteristics of domination by contagious diffusion, with Shanghai as the center. According to the law of geographical proximity, it first spread in Shanghai and gradually spread to Suzhou, Jiaxing, and other directions, showing a large-scale industrial agglomeration core in space. At the same time, a small range of industrial agglomeration points also formed in Hangzhou, Nanjing, and Hefei, which were also spread throughout the cities through contagious diffusion. In the early days, the digital creative industries were primarily concentrated in Shanghai, because as the center of China’s economic strength and science and technology innovation, Shanghai has a long history of development, a strong economic foundation, scientific and technological strength, and strong policy support, which have laid a solid foundation for the development of digital creative industries.

4.1.2. 2017 to 2019: The Diffusion Characteristic Dominated by Hierarchical Diffusion

Figure 3, Figure 4 and Figure 8 show that the spatial diffusion of digital creative industries in the YRD from 2017 to 2019 presents the characteristics of domination by hierarchical diffusion. Based on the initial diffusion pattern, with Shanghai as the center, it spread from west to south, along roads and rivers, and it formed agglomeration centers in Hangzhou, Nanjing, Hefei, and other provincial capitals through hierarchical diffusion. Simultaneously, the industrial agglomeration centers of Shanghai, Hangzhou, Nanjing, and Hefei spread from the center to the periphery through contagious diffusion. The agglomeration scope of Shanghai expanded to Suzhou, Nantong, Jiaxing, and other cities, forming a large-scale industrial agglomeration pole. Nanjing expanded to Zhenjiang, Ma Anshan, Chuzhou, and other cities, while Hangzhou expanded to Huzhou and Shaoxing to form a secondary agglomeration center. Hefei’s industrial effect has also been continuously strengthened, spreading to Huainan, Lu’an, and other cities. In terms of the industrial spatial distribution, digital creative industry has formed a “Z-shaped corridor” distribution characteristic along “Hefei-Nanjing-Suzhou -Shanghai-Hangzhou-Ningbo”, and the overall diffusion mode is dominated by hierarchical diffusion.

4.1.3. 2020 to the Present: Mixed Diffusion Is the Dominant Diffusion Mode

Figure 3, Figure 4 and Figure 8 show that the spatial diffusion mode of digital creative industries in the YRD from 2020 to 2022 reflected a combination of hierarchical, contagious, corridor, and jump diffusion. The industrial agglomeration centers diffuse along the “Z-shaped corridor”. A small-scale industrial cluster center has emerged in Xuzhou, Suqian, and other regional peripheral cities through hierarchical diffusion to the north, and the development of the digital creative industry has broken through the central region of the YRD. Meanwhile, various industrial clusters have formed in Yangzhou, Yancheng, Huai’an, Lianyungang, Suzhou, and other cities in the north; Wuhu, Wuxi, and other cities in the middle; and Jinhua, Ningbo, and other general cities in the south through contagious diffusion. However, in Fuyang in the northwest and Wenzhou in the southeast, there are “enclave-type” cluster centers, and the “jump” diffusion mode of industries is obvious. Fuyang has a large population and a broad market, attracting the agglomeration of digital creative industries. Wenzhou, on the other hand, is on the southeast coast, with a superior location and great development potential. In terms of the industrial spatial distribution, the digital creative industries have gradually formed the characteristics of “networked” distribution, generally presenting the diffusion mode dominated by a combination of hierarchical, contagious, corridor and jump diffusion.

4.2. The Spatial Response of the Digital Creative Industries to National Regional Development Strategies

4.2.1. The Overall Spatial Response of the Digital Creative Industries

On 11 May 2016, China issued the YRD Urban Agglomeration Development Plan, which, for the first time, included the Hefei metropolitan area in Anhui Province in the YRD city cluster, totaling 26 cities, the aim of which was to take the lead in developing the new economy and promoting collaborative innovation in key areas, such as information technology. On 1 December 2019, China issued the Outline of the YRD Regional Integration Development Plan, which included the whole of Anhui in the YRD region, as well as 41 municipalities, and it proposed to implement an innovation-driven development strategy and take the road of “science and technology innovation + industry.” The introduction of the national development strategy has gradually expanded the spatial scope of the YRD, providing policy opportunities for the expansion of the hinterland of the digital creative industry market.

Figure 5 shows that SDE and WMC of the digital creative industries gradually expanded to the southwest and northwest, coupled with the spatial scope expansion of regional integration strategies. In 2016, the YRD boasted a total of 284,738 digital creative enterprises, and this number reached 728,456 in 2019, an increase of 443,718 compared with 2016. In 2022, it reached 923,759, an increase of 19,5303 compared to 2019. Looking at the rate of enterprise growth in the digital creative industries, it was found that it accelerated after the introduction of the regional development strategy in 2016 and slowed in 2022, primarily attributed to the disruptive effects of COVID-19. This further underscores the significant impact of regional development strategies on the advancement of digital creative industries.

4.2.2. The Urban Spatial Response of the Digital Creative Industry Layout

As depicted in

Table 7. Descriptive statistics of various digital creative industry regions in the YRD.

Time	2016		2019		2022
NO.	City	NOC	City	NOC	City	NOC
1	Shanghai	100,370	Shanghai	179,844	Shanghai	222,401
2	Hangzhou	25,204	Hangzhou	79,887	Hangzhou	101,409
3	Nanjing	24,217	Nanjing	53,636	Suzhou	74,496
4	Hefei	14,542	Suzhou	51,832	Nanjing	72,160
5	Suzhou	14,160	Hefei	36,528	Hefei	52,137
6	Wuxi	9395	Wuxi	35,182	Wuxi	46,348
7	Changzhou	8406	Jinhua	32,567	Jinhua	42,567
8	Jinhua	7804	Changzhou	26,348	Changzhou	39,545
9	Ningbo	6731	Ningbo	25,214	Ningbo	33,182
10	Wenzhou	5428	Xuzhou	18,510	Xuzhou	27,223

Data sources: Qixinbao (https://www.qixin.com/, accessed on 1 March 2023.).

, Shanghai, Hangzhou, Nanjing, Hefei, and Suzhou emerged as the top five cities in terms of the number of digital creative enterprises in 2016. In 2019, Shanghai, Hangzhou, Nanjing, Suzhou, and Hefei ranked in the top five, and by 2022, Shanghai, Hangzhou, Suzhou, Nanjing, and Hefei ranked in the top five. From the ranking situation concerning Shanghai, Hangzhou and other provincial capitals and against the backdrop of regional development strategies, the agglomeration effect of the digital creative industries is becoming stronger and stronger. Suzhou responded most clearly to the regional development strategy, ranking fifth in 2016, fourth in 2019, and third in 2022, and the development of digital creative industries continues to accelerate.

Figure 6 illustrates the changing trends in the number of digital creative industries in all cities of the Yangtze River Delta from 2016 to 2023, together with their spatial response characteristics. From a regional perspective, cities in the southern, central, and northern parts of the Yangtze River Delta exhibited differing responses to the regional development strategy. The central region, led by Shanghai City and distributed along the Yangtze River in riverside cities, demonstrates the most evident response to regional development strategies. This is related to development policies such as the G60 Science and Innovation Corridor. The southern region is mainly centered on Hangzhou, extending southward along the coastline to coastal cities such as Ningbo and Wenzhou. These cities have relatively high scientific and technological strength and economic levels. The northern region is primarily centered around Xuzhou City, with fewer industries in surrounding cities overall. This indicates a relatively weaker response to regional development strategies in the northern part of the YRD.

4.2.3. Response Comparison of Different Digital Creative Industries

Figure 9 shows that the number of subsectors of the digital creative industries from 2016 to 2022 has shown an increasing trend year over year, but the quantitative response to regional development strategies between industries is obviously different. Specifically, the number of companies in each segment from the 2016 to 2019 rankings is as follows: digital creativity and integration > digital design > creative software engineering > other creative activities > digital content production service > new media services > digital creative technology equipment manufacturing > radio and television services. Meanwhile, the number of companies by industry segment in 2022 is as follows: CAI > DD > OCA > SE > CPS > NMS > TEM > RATS. Among them, OCA developed rapidly from 2019 to 2022, surpassing SE and becoming the most sensitive to regional development strategies. In general, CAI, DD, OCA, SE, and other industry enterprises are distributed in the YRD, and their response to regional development strategies is more obvious. However, the number of enterprises in RATS, TEM, and other industries is relatively small, and the response to regional development strategies is relatively weak. This has a certain relationship with industry attributes, and such industries as RATS and TEM have high requirements for cultural ideology, so the quantitative response is relatively weak. From the perspective of time, enterprises in various industry segments were in a stage of rapid growth from 2016 to 2019, and from 2020 to 2022, affected by COVID-19, the growth in the number of enterprises slowed and gradually changed from high-speed to high-quality growth.

4.3. Factors Influencing the Regional Spatial Diffusion of the Digital Creative Industries

This study compares the relationship between the spatial distribution of the digital creative industry and various influencing factors using OLS and quantile regression models (

Table 8. Results of OLS and quantile regression models for digital creative industry spatial diffusion.

Time	EV	OLS	Quantile
			Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
2016	X1	1.561 (0.003 ***)	2.556 (0.009 *** )	2.063 (0.027 **)	2.354 (0.011 **)	2.382 (0.008 ***)	1.559 (0.060 *)	1.457 (0.057 *)	1.215 (0.143)	1.163 (0.223)	0.068 (0.953)
	X2	−0.416 (0.440)	0.006 (0.996)	−0.310 (0.781)	−1.020 (0.425)	−1.045 (0.420)	−1.323 (0.281)	−1.283 (0.268)	−0.784 (0.512)	−0.569 (0.566)	−0.532 (0.584)
	X3	0.292 (0.007 ***)	0.311 (0.000 ***)	0.255 (0.000 ***)	0.251 (0.244)	0.286 (0.150)	0.361 (0.166)	0.343 (0.153)	0.318 (0.110)	0.327 (0.108)	0.395 (0.101)
	X4	−0.114 (0.543)	−0.389 (0.272)	−0.162 (0.639)	−0.153 (0.667)	−0.158 (0.655)	0.135 (0.709)	0.186 (0.619)	0.126 (0.768)	0.193 (0.051 **)	0.480 (0.012 **)
	X5	0.241 (0.000 ***)	0.746 (0.000 ***)	0.786 (0.000 ***)	0.813 (0.000 ***)	0.844 (0.000 ***)	0.936 (0.000 ***)	0.905 (0.000 ***)	0.781 (0.000 ***)	0.735 (0.000 ***)	0.622 (0.000 ***)
	X6	0.267 (0.320)	0.220 (0.046 *)	0.230 (0.058 *)	0.240 (0.038 **)	0.258 (0.043 **)	0.130 (0.303)	0.157 (0.229)	0.248 (0.124)	0.187 (0.289)	0.220 (0.301)
	R2	0.735	0.745	0.744	0.761	0.772	0.771	0.783	0.790	0.793	0.809
Time	EV	OLS	Quantile
			Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
2019	X1	2.054 (0.022 **)	2.368 (0.067 *)	1.978 (0.079 *)	1.371 (0.072 *)	1.645 (0.084 *)	2.310 (0.021 **)	1.418 (0.223)	1.374 (0.247)	1.497 (0.250)	1.729 (0.128)
	X2	−1.766 (0.103)	−2.843 (0.364)	−1.220 (0.598)	−1.409 (0.409)	−1.573 (0.291)	−1.884 (0.208)	−1.338 (0.339)	−0.387 (0.801)	−0.929 (0.533)	−1.011 (0.427)
	X3	0.270 (0.048 **)	0.266 (0.067 *)	0.240 (0.039 *)	0.413 (0.027 **)	0.231 (0.313)	0.267 (0.243)	0.431 (0.193)	0.347 (0.165)	0.212 (0.390)	0.226 (0.346)
	X4	0.443 (0.321)	0.733 (0.475)	0.176 (0.828)	0.142 (0.827)	0.194 (0.735)	0.106 (0.847)	0.072 (0.901)	0.191 (0.078 *)	0.248 (0.073 *)	0.534 (0.035 **)
	X5	0.442 (0.014 **)	0.664 (0.046 **)	0.652 (0.017 **)	0.635 (0.020 **)	0.586 (0.042 **)	0.473 (0.176 )	0.452 (0.111)	0.498 (0.160)	0.435 (0.199)	0.353 (0.103)
	X6	0.419 (0.005 ***)	0.336 (0.163)	0.327 (0.105)	0.273 (0.179)	0.449 (0.040 ** )	0.524 (0.020 **)	0.437 (0.068 *)	0.363 (0.009 ***)	0.483 (0.033 **)	0.372 (0.117)
	R2	0.789	0.618	0.681	0.699	0.711	0.722	0.736	0.746	0.765	0.808
Time	EV	OLS	Quantile
			Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
2022	X1	1.160 (0.182)	2.166 (0.198)	1.299 (0.296)	1.293 (0.243)	1.212 (0.285)	0.914 (0.431)	1.800 (0.158)	1.130 (0.407)	1.624 (0.276)	0.784 (0.585)
	X2	−1.905 (0.008 ***)	−3.576 (0.037 **)	−1.745 (0.277)	−1.266 (0.375)	−1.147 (0.397)	−2.004 (0.157)	−2.101 (0.161)	−2.429 (0.141)	−2.665 (0.220)	−3.575 (0.102)
	X3	0.241 (0.021 **)	0.013 (0.048 **)	0.137 (0.091 *)	0.128 (0.395)	0.162 (0.341)	0.217 (0.249)	0.304 (0.141)	0.120 (0.622)	0.048 (0.868)	−0.032 (0.921)
	X4	0.232 (0.410)	0.541 (0.362)	0.104 (0.859)	0.074 (0.894)	0.095 (0.851)	0.060 (0.906)	0.232 (0.063 *)	0.516 (0.046 **)	0.765 (0.037 **)	1.160 (0.021 **)
	X5	0.336 (0.016 **)	0.798 (0.002 ***)	0.507 (0.044 **)	0.349 (0.141)	0.314 (0.198)	0.392 (0.109)	0.254 (0.307)	0.317 (0.229)	0.158 (0.535)	0.290 (0.230)
	X6	0.614 (0.000 *** )	0.633 (0.007 ***)	0.618 (0.012 **)	0.744 (0.005 ***)	0.725 (0.005 *** )	0.700 (0.008 ***)	0.587 (0.031 **)	0.790 (0.007 ***)	0.865 (0.007 ***)	0.917 (0.005 ***)
	R2	0.718	0.726	0.727	0.730	0.735	0.733	0.747	0.754	0.774	0.902

Note: ***, **, and * denote statistical significance at the 1%, 5% and 10% level, respectively; parentheses show p-value. The abbreviations of the segments are as follows: EV (explanatory variable). Explanatory variables are shown in Table 2. X1 (share of the tertiary industry in the GDP), X2 (urbanization rate), X3 (government investment in science and technology), X4 (policy effectiveness), X5 (number of universities in the city), and X6 (number of patents authorized in the city). Data sources: Qixinbao (https://www.qixin.com/, accessed on 1 March 2023.), the National Bureau of Statistics of China, Statistical Yearbooks of Each City.

). The R² values of the OLS model for the years 2016 to 2022 are 0.735, 0.789, and 0.718, respectively. For the quantile regression model, the R² ranges from 0.744 to 0.809, 0.618 to 0.808, and 0.726 to 0.902. The models exhibit good goodness of fit and statistical significance.

The model results (Table 8) indicate that the effects of various factors on the diffusion of the digital creative industry varied across periods. In 2016, the spatial distribution of the digital creative industry in the Yangtze River Delta was significantly and positively correlated with X1, X3, and X5. In 2019, significant positive correlations were observed with X1, X5, X6 and X3. In 2022, the spatial distribution was significantly and positively correlated with X6, X5, and X3, while showing a significant negative correlation with X2. As shown in Table 8, X3 and X5 consistently served as key factors influencing the spatial distribution and diffusion of the industry. Relevant government policies helped foster a favorable innovation environment and creative atmosphere, while government funding for science and technology further promoted changes in traditional industrial organization and development models, providing fertile technological and creative ground as well as financial support for enterprises entering local markets. Universities represented the local supply of talent and the capacity for knowledge production, serving as an important source of creative professionals and attracting the clustering of digital creative enterprises. Overall, in the early stage, industrial structure and institutional environment were the primary factors affecting the agglomeration and diffusion of the digital creative industry. In the later stage, with the advancement of the economy and technology and the intensification of competition among enterprises, technological innovation increasingly became the dominant factor influencing the industry’s agglomeration and diffusion.

Further quantile regression using the same set of variables provided more detailed results (Table 8). For example, from 2016 to 2019, the share of the tertiary industry in GDP (X1) was significantly and positively correlated with the lower and middle quantiles of the digital creative industry distribution (Q10–Q60, Q10–Q50). This indicates that X1 had a stronger effect on low-agglomeration industries than on high-agglomeration industries. The digital creative industry is closely linked to the tertiary sector, and adjustments in industrial structure promote synergetic development between upstream and downstream enterprises, thereby attracting the early-stage clustering of digital creative enterprises. Another noteworthy example is the number of universities (X5). In 2016, the estimated coefficients of X5 first increased (Q10–Q50) and then decreased (Q60–Q90) while remaining significantly positive across all quantiles. However, from 2019 to 2022, the coefficients of X5 showed a downward trend and were significantly positive only in the lower quantiles of the industry distribution (Q10–Q40, Q10–Q20). This suggests that, in both the early and later stages of industrial development, the denser the enterprise distribution, the weaker the effect of X5. The number of universities reflects a region’s innovation capacity, and because the digital creative industry has high requirements for modern digital technologies, it tends to locate in areas with a greater number of universities in the early development stage to enhance competitive advantages. In the later stage, however, as competition intensifies and the number of universities remains relatively fixed, while the industry grows rapidly, the effect of universities may gradually diminish due to issues such as underdeveloped channels for industry–university–research integration. It is also noteworthy that in 2022, the spatial distribution of the digital creative industry was negatively correlated with the urbanization rate (X2), indicating that areas with lower urbanization rates exhibited higher industry agglomeration. This may be because such areas generally have smaller populations and lower land rents, making them suitable for site selection and clustering in the later stages of industrial development.

On the other hand, policy effectiveness (X4) was not significant in the OLS model but exhibited a nonlinear effect on industrial spatial diffusion in the quantile regression results. From 2016 to 2022, X4 was significantly and positively correlated with the higher quantiles of the digital creative industry distribution (Q80–Q90, Q70–Q90, Q60–Q90). This indicates that policy effectiveness played a stronger role in regions with high industry agglomeration. A possible explanation is that the implementation of a large number of policies requires time to release their dividends and generate tangible impacts. However, it also reflects that existing digital creative industry policies are subject to pronounced lag effects, an insufficient policy system, and imprecise policy instruments. Overall, the direction of the effect of digital creative policies on industrial development is clear. However, the impact of policy effectiveness depends not only on the orientation and intensity of the policy itself but also on the stage of development at which the industry is situated when the policy is implemented. The same policy (with identical direction and intensity) may exert different practical impacts on enterprises across different development stages and quantiles, indicating that the effectiveness of digital creative industry policies is asymmetric across different periods.

In addition, the number of patents (X6) also showed a significant influence on the spatial distribution of the industry in the quantile regression results. In 2016, X6 was significantly and positively correlated with the lower quantiles (Q10–Q40) of the digital creative industry distribution. In 2019, it was significantly and positively correlated with the middle to upper quantiles (Q40–Q80), and in 2022, it was significantly and positively correlated across all quantiles (Q10–Q90). These results indicate that the number of patents is also an important factor affecting the agglomeration and diffusion of the digital creative industry. A higher number of local patent applications reflects stronger innovation capacity, which exerts greater promotion effects on industries dominated by digital technologies. Particularly in the middle and later stages of industrial development, as the scale of agglomeration expands and competition among enterprises intensifies, the industry’s dependence on innovation capacity, such as the number of patents, becomes increasingly pronounced.

4.4. Robustness Tests

4.4.1. Kernel Density Estimation Test with Variable Bandwidths

To assess the impact of bandwidth settings on kernel density estimation results, this study used 3000 m as the baseline and compared it with two alternative bandwidths of 2000 m and 5000 m (Figure 7). The results showed that, although changes in bandwidth affected the smoothness of the density maps, the overall spatial pattern and diffusion trends of the digital creative industry in the Yangtze River Delta remained largely consistent. This indicates that the spatial analysis results in this study are robust to different bandwidth settings.

4.4.2. Ridge Regression Test

Ridge regression was employed as an alternative estimation to the OLS and quantile regression models (

Table 9. Results of ridge regression test.

EV	Ridge
	2016	2019	2022
X1	1.482 (0.000 ***)	1.834 (0.009 ***)	1.166 (0.145)
X2	−0.133 (0.728)	−0.864 (0.116)	−0.814 (0.043 **)
X3	0.298 (0.000 ***)	0.308 (0.000 ***)	0.285 (0.000 ***)
X4	−0.062 (0.630)	0.279 (0.288)	0.057 (0.652)
X5	0.619 (0.000 ***)	0.401 (0.000 ***)	0.322 (0.000 ***)
X6	0.266 (0.256)	0.344 (0.000 ***)	0.473 (0.000 ***)
R2	0.941	0.901	0.923

Data sources: Qixinbao (https://www.qixin.com/, accessed on 1 March 2023.), the National Bureau of Statistics of China, Statistical Yearbooks of Each City. ***, **, denote statistical significance at the 1%, 5% level, respectively.

). The results show that the model R² values for 2016–2022 were 0.941, 0.901, and 0.923, indicating good model fit. The signs and significance levels of the effects of the explanatory variables on the spatial distribution of the digital creative industry were largely consistent with those in the baseline regressions, suggesting that the conclusions of this study are robust under different model specifications.

4.4.3. Lag Effect Test

Based on the OLS and quantile regression models, the contemporaneous policy effectiveness variable (X4) was replaced with its one-period lag (X4_lag1) to construct a lag effect model (

Table 10. Results of the lag effect test.

Time	EV	OLS	Quantile
			Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
2019	X1	2.245 (0.009 ***)	4.183 (0.004 ***)	2.359 (0.072 *)	2.507 (0.050 *)	2.322 (0.061 *)	1.981 (0.100 *)	2.347 (0.054 *)	2.256 (0.067 *)	1.654 (0.190)	2.014 (0.095 *)
	X2	−0.513 (0.519)	0.667 (0.746)	−0.756 (0.625)	−0.564 (0.710)	−0.637 (0.673)	−0.325 (0.828)	0.106 (0.945)	0.190 (0.904)	−0.486 (0.785)	−0.236 (0.914)
	X3	0.270 (0.016 ** )	0.021 (0.020 **)	0.133 (0.491)	0.234 (0.228)	0.223 (0.291)	0.208 (0.333)	0.166 (0.446)	0.153 (0.496)	0.258 (0.277)	0.217 (0.369)
	X4_lag1	1.676 (0.047 **)	2.715 (0.354)	0.935 (0.523)	0.885 (0.571)	0.611 (0.713)	1.849 (0.269)	2.084 (0.212)	2.800 (0.113)	1.340 (0.046 **)	0.442 (0.084 *)
	X5	0.325 (0.090 *)	0.307 (0.085 *)	0.629 (0.022 **)	0.482 (0.087 *)	0.486 (0.098 *)	0.374 (0.184)	0.161 (0.566)	0.138 (0.630)	0.331 (0.264)	0.277 (0.334)
	X6	0.436 (0.005 ***)	0.488 (0.041 ** )	0.371 (0.103)	0.423 (0.072 *)	0.477 (0.059 *)	0.513 (0.033 **)	0.651 (0.009 ***)	0.650 (0.007 ***)	0.372 (0.093 *)	0.340 (0.116)
	R2	0.903	0.671	0.706	0.718	0.717	0.728	0.745	0.757	0.771	0.811
Time	EV	OLS	Quantile
			Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
2022	X1	1.147 (0.110)	2.344 (0.100 *)	1.673 (0.196)	1.504 (0.228)	1.523 (0.208)	0.928 (0.461)	2.047 (0.132)	1.629 (0.273)	1.567 (0.313)	0.811 (0.595)
	X2	−1.903 (0.010 **)	−2.970 (0.077 *)	−1.680 (0.091 *)	−2.041 (0.161)	−2.340 (0.191)	−1.843 (0.194)	−2.249 (0.123)	−1.955 (0.248)	−2.351 (0.240)	−3.288 (0.185)
	X3	0.241 (0.023 **)	0.375 (0.044 **)	0.135 (0.415)	0.187 (0.219)	0.201 (0.254)	0.270 (0.186)	0.302 (0.180)	0.262 (0.307)	0.181 (0.542)	−0.018 (0.959)
	X4_lag1	0.055 (0.054 *)	1.191 (0.381)	0.533 (0.703)	1.079 (0.432)	1.170 (0.455)	0.442 (0.798)	0.241 (0.898)	1.500 (0.031 **)	0.795 (0.064 *)	0.288 (0.069 *)
	X5	0.338 (0.015 **)	0.420 (0.086 *)	0.452 (0.058 *)	0.446 (0.055 *)	0.412 (0.068 *)	0.405 (0.093 *)	0.236 (0.337)	0.277 (0.283)	0.187 (0.479)	0.221 (0.369)
	X6	0.610 (0.001 ***)	0.472 (0.060 *)	0.685 (0.007 ***)	0.705 (0.007 ***)	0.764 (0.006 ***)	0.631 (0.030 **)	0.588 (0.060 *)	0.479 (0.162)	0.639 (0.111)	0.955 (0.041 **)
	R2	0.915	0.748	0.731	0.736	0.737	0.734	0.747	0.757	0.776	0.812
Time	EV	OLS	Quantile
			Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
2019–2022 (pooled)	X1	1.891 (0.004 ***)	2.421 (0.006 ***)	2.433 (0.003 ***)	2.396 (0.004 ***)	1.856 (0.023 **)	2.057 (0.013 **)	1.408 (0.115)	1.551 (0.098 *)	1.626 (0.074 *)	1.189 (0.206)
	X2	−1.606 (0.013 **)	−1.641 (0.071 *)	−1.820 (0.132)	−1.281 (0.193)	−1.590 (0.101)	−1.514 (0.144)	−1.000 (0.346)	−1.914 (0.181)	−1.462 (0.207)	−1.571 (0.193)
	X3	0.253 (0.004 ***)	0.306 (0.013 **)	0.141 (0.170)	0.121 (0.314)	0.184 (0.185)	0.224 (0.165)	0.411 (0.018 **)	0.344 (0.072 *)	0.239 (0.239)	0.226 (0.254)
	X4_lag1	0.640 (0.045 **)	−0.831 (0.345)	−0.755 (0.339)	−0.857 (0.233)	−0.338 (0.645)	−0.364 (0.640)	−0.580 (0.490)	0.048 (0.958)	0.302 (0.061 *)	0.176 (0.038 **)
	X5	0.340 (0.008 ***)	0.465 (0.022 **)	0.541 (0.001 ***)	0.387 (0.017 **)	0.481 (0.005 ***)	0.428 (0.017 **)	0.338 (0.068 *)	0.349 (0.074 *)	0.234 (0.235)	0.387 (0.029 **)
	X6	0.544 (0.001 ***)	0.426 (0.004 ***)	0.472 (0.001 ***)	0.645 (0.001 ***)	0.589 (0.001 ***)	0.557 (0.001 ***)	0.462 (0.011 ** )	0.491 (0.014 **)	0.553 (0.006 ***)	0.462 (0.009 ***)
	R2	0.914	0.699	0.713	0.721	0.727	0.733	0.739	0.746	0.764	0.786

Data sources: Qixinbao (https://www.qixin.com/, accessed on 1 March 2023.), the National Bureau of Statistics of China, Statistical Yearbooks of Each City. Note: ***, **, and * denote statistical significance at the 1%, 5% and 10% level, respectively.

). The results showed that the model exhibited good fit. In the OLS model, the lagged variable (X4_lag1) was significantly and positively correlated with the spatial distribution of the digital creative industry during 2019–2022, confirming the existence of a lag in policy effectiveness. In the quantile regression results, X4_lag1 was significantly and positively correlated with the higher quantiles (Q80–Q90, Q70–Q90) during 2019–2022, indicating that policy measures require time to transmit their effects and exert a notable promoting influence in regions with high industrial levels. These findings, along with the pooled-sample results, were largely consistent with the baseline regression results.

5. Discussion

Since the reform and opening up, the YRD has gradually become the economic and financial center of China. However, as the economy and society have developed, the problems of unbalanced and inadequate regional development still exist [37,45], which not only hinder the optimization of regional spatial structures but also undermine efforts to achieve coordinated and sustainable development across regions. In particular, the outbreak of COVID-19 and the rise of geopolitical protectionism have exacerbated this contradiction [46]. In 2018, the Chinese Government proposed upgrading the integrated development strategy of the YRD into a national strategy, breaking regional administrative boundaries, promoting regional cooperation, and exploring new paths for transforming the national governance model. Driven by the integrated development strategy of the YRD, the dynamic role of government policies and markets in regional development is increasingly enhanced. Government-oriented and bottom-up regional market cooperation has changed the urban spatial structure and organizational form of the YRD. In this work, digital creative industries provide new theoretical and practical perspectives for the sustainable development of regional industries by means of spatial diffusion.

First, this study enriches the theoretical application of industrial diffusion theory. The spatial diffusion of the digital creative industry exhibits multiple modes across different periods, including contagious diffusion, hierarchical diffusion, corridor diffusion, and jump diffusion. In the case of the Yangtze River Delta (YRD), consistent with studies on traditional creative industries [55], we find that in the early stage, a “polarized” core of the digital creative industry emerged in large cities, dominated by contagious diffusion [26], whereby the industry expanded outward from central to peripheral cities along proximity-based pathways [29]. However, unlike cultural creativity, design services, film and television, and animation and gaming, which tend to form a single agglomeration core around higher-order cities [28,56,57], the digital creative industry in the YRD follows a hierarchical diffusion pattern [25] and has formed multiple agglomeration cores. Particularly under the impetus of YRD integration, the industry has accelerated its expansion toward inland cities, with hierarchical diffusion occurring from eastern core cities to western provincial capitals, forming large-scale agglomeration centers in core cities and provincial capitals. In the later stage, these industrial agglomeration centers gradually became interconnected through contagious and hierarchical diffusion, giving rise to a distinctive “Z-shaped corridor” pattern linking Hefei–Nanjing–Suzhou–Shanghai–Hangzhou–Ningbo. This “corridor diffusion” mode is not only a result of geographical proximity but is also closely related to regional integration strategies. Policies such as the G60 Science and Technology Innovation Corridor, supported by key transport arteries like the G60 Expressway and the Huzhou–Suzhou–Shanghai High-Speed Railway, have tightly connected core cities such as Shanghai, Hangzhou, and Hefei with small- and medium-sized cities along the route in terms of capital, talent, information, and technology. This has fostered an industrial division of labor along the corridor and accelerated the spatial restructuring of the digital creative industry. Moreover, similar to subsectors such as digital content and creative design that, under policy interventions, directly diffuse from urban centers to geographically peripheral areas [31], new industry agglomeration centers have emerged in cities such as Fuyang in the northwest and Wenzhou in the southeast of the YRD, driven primarily by factors such as population and location. Spatially, this reflects a “jump” diffusion pattern. Overall, the YRD has developed a mixed diffusion model combining hierarchical, contagious, corridor, and jump diffusion modes. Through multiple diffusion modes, the digital creative industries have promoted industrial agglomeration and relocation, facilitating the formation of a point-axis regional development structure. Evolving from a centralized “polar core” configuration to axis-based linkages and ultimately toward a networked pattern, they have reshaped regional industrial organization and offered a pathway toward a spatially balanced, functionally complementary, and sustainable urban system.

Second, the research provides valuable insights for developing countries to promote the sustainable development of digital industries in similar regions. In the early stages of regional development, government policies promoted the spatial agglomeration and site selection of digital creative enterprises [5,8,41,54] and played a crucial role in the spatial diffusion of digital creative enterprises. As the leading area of China’s reform and opening up, the YRD has the characteristics of significant economic volume, a high degree of industrial agglomeration, frequent factor flow, a relatively developed infrastructure, a dense urban distribution, and strong scientific and technological innovation capabilities. With the release of the Yangtze River Delta integration development strategy, led by market-oriented reforms, a series of policies, such as regional market integration, public service integration, development policy coordination, regional transportation integration, and cultural tourism integration, have been implemented. In particular, the continuous increase in government investments in science and technology, along with the accelerated construction of a collaborative innovation industrial system, has promoted the diffusion of industries such as the digital creative industry and supported the cross-regional development of enterprises. This has also driven the industrial development of peripheral cities in the region. The cases of Xuzhou, Suqian, and Fuyang in this study confirm this point. However, whether the complex and redundant policy system of industrial policies is effective remains unknown to us. Unlike the qualitative approach used to explore policies promoting industrial development in major cities such as Beijing and Shanghai [58,59], we found that the effectiveness of digital creative industry policies in the YRD has nonlinear effects and asymmetric characteristics on industry distribution. It affects enterprises differently at different stages of industrial development and exhibits significant lag effects. Policymakers and planners need to adopt a “systemic perspective,” focusing on market demand, strengthening overall planning, coordinating across multiple departments, scientifically assessing the stages of industrial development, enriching and improving the content and forms of policies, continuously diversifying policy measures, and optimizing the policy system for the digital creative industry. In the later stage of industrial development, innovation plays a critical role in the diffusion and regional response of the digital creative industry. For industries dominated by digital technologies, innovation capacity is essential [60]. The development of research institutions and the increase in patent numbers enhance enterprise innovation capacity, enabling firms to form new technological systems. Therefore, government decision makers and industry managers should place high importance on building innovation systems, including strengthening the layout of research platforms, improving intellectual property protection, and enhancing industry–university–research collaboration mechanisms, so as to comprehensively improve regional innovation capacity and provide sustained momentum for the sustainable development of the industry.

6. Conclusions

The research method herein was based on big data to obtain all the data of digital creative enterprises in the YRD, using the time the national major development strategy was introduced as time points. In addition, through geographic spatial analysis, the regional spatial diffusion model of digital creative industries was deconstructed in different periods; the different responses of digital creative industries to the national major strategies were analyzed from different perspectives, such as space, urban level, and segmented industries; and the OLS model and quantile regression model were used to explore the specific factors influencing spatial diffusion and the regional response formation of digital creative industries. The following conclusions are drawn.

First, the digital creative industries had different spatial diffusion modes in different periods. In the early stage, the digital creative industry showed a diffusion mode dominated by contagious diffusion, and the initial diffusion pattern with Shanghai as its core was formed spatially, with the spread carried out in Shanghai according to the law of geographical proximity. In the medium term, with the introduction of regional development strategies, the digital creative industries presented a diffusion mode dominated by hierarchical proliferation. The hierarchical diffusion mainly spread from Shanghai to Hangzhou, Nanjing, Hefei, and other provincial capitals, and, at the same time, the industrial agglomeration centers spread to the periphery via contagious diffusion. In the later stage, the digital creative industry showed domination via mixed diffusion. While hierarchical diffusion and contagious diffusion occurred, there was also “Z-shaped corridor” diffusion along the route from Hefei–Nanjing–Suzhou to Shanghai–Hangzhou–Ningbo. Additionally, in the northwestern region around Fuyang and southeastern region around Wenzhou, a “jump” diffusion pattern emerged, driven by different factors such as population and location.

Second, digital creative industries demonstrate differences in regional spatial responses to the implementation of a national strategy. Spatially, there is a high degree of coupling among the mean center of the digital creative industry, the SDE development direction, and the regional strategic direction. In terms of cities, the digital creative industries in regional core cities and provincial capital centers have developed rapidly and responded significantly to regional development strategies. Regionally, the riverside cities and southern coastal cities in central YRD have a more obvious response than northern cities. In different industries, CAI, DD, OCA, SE, and other industry enterprises are distributed in the YRD, and their response to regional development strategies is more obvious.

Third, the spatial diffusion and regional response of digital creative industries in the YRD are comprehensively affected by the institutional environment, urban economy, development, and innovation. Results from the OLS and quantile regression models indicate that government investments in science and technology and the number of universities have consistently been significant influencing factors, while policy effectiveness exhibits nonlinear effects and asymmetric characteristics in shaping the spatial distribution of the industry. From different stages of development, in the early stages, the institutional environment was the main factor affecting the clustering and diffusion of the digital creative industry. In the later stages, with the development of high-tech industries, the impact of innovation on the clustering and diffusion of the digital creative industry became increasingly significant. Additionally, the spatial distribution of the digital creative industry is negatively correlated with the urbanization rate.

The main innovation points of the research are as follows: First, the research expands the perspective of the digital creative industries. The existing research primarily analyzes the agglomeration of creative industries in cities, while this study explores the spatial diffusion of digital creative industries from a regional perspective. Second, the research enriches the relevant applications of diffusion theory. It is found that the digital creative industries in the YRD not only have a diffusion mode, such as hierarchical diffusion and contagious diffusion, but also involve “Z-type” corridor diffusion and factor-led jump diffusion. Third, this study uses the method of big data and takes full sample data to analyze the spatial diffusion of digital creative industries. Finally, existing research primarily analyzes the impact of policies on digital creative industries through qualitative research, while this study quantitatively analyzes the nonlinear impact and asymmetric characteristics of policies on the spatial diffusion of the digital creative industry.

At the same time, this study faces several problems and challenges. Regarding data sources, the Qixinbao platform was used. Although Qixinbao, as a well-known enterprise information retrieval platform in China, offers advantages in data accessibility and coverage, it still has limitations such as delayed updates, imprecise industry classification, and missing information for some enterprises, which may affect data accuracy. In terms of modeling assumptions, this study involves certain assumptions in variable selection and model functional forms, which may influence the robustness of the results. In addition, the analysis was conducted using the Yangtze River Delta in China as a case study, and the applicability of the findings to other regions (e.g., the European Union) requires further verification. Future research could integrate diverse statistical data with field surveys, incorporate spatial econometric models, and conduct international regional comparisons to perform comprehensive analyses of multiple influencing factors, thereby gaining deeper insights into the spatial diffusion patterns and regional response mechanisms of the digital creative industry.