The Interaction of Cultural and Creative Industries Clusters and Regional Economic Resilience from the Perspective of Spatial Analysis

Abstract

Cultural and creative industries (CCIs) clusters are essential in increasing regional economic resilience, and regions with solid economic resilience can also promote the formation of CCIs clusters. However, there is little research on the two-way effect of cultural and creative industries on regional economic resilience. This study explores the relationship between the clusters of CCIs clusters and regional economic resilience from a spatial perspective. This paper takes 31 provinces and cities in China as the research object, uses panel data from 1997 to 2020, and divides the CCIs clusters into specialization and diversification. The spatial simultaneous equations model was used to verify the two-way influence relationship and the spatial interaction between the specialization and diversification of CCIs clusters and the regional economic resilience, respectively. The study found a two-way positive relationship between specialized CCIs clusters and regional economic resilience, while the opposite relationship existed between diversified CCIs clusters and regional economic resilience. There were positive spatial spillovers between CCIs clusters and regional economic resilience. The enhancement of local economic resilience comes from the specialization of local CCIs clusters and the diversification of CCIs clusters in neighboring regions.

1. Introduction

Cultural and creative industries (CCIs) are essential to achieve economic recovery. Their role in regional economic structure transformation can benefit any community. The proportion of added value to China’s GDP is rising and was 4.43% in 2020 [1]. The 2020 COVID-19 crisis hit the world’s economies hard, and, in terms of the GDP growth rate of each country, the United States (−3.5%), the United Kingdom (−9.8%), Germany (−4.9%), France (−8.1%) and other countries have had negative growth. China is one of the few countries with a positive growth rate, with a GDP growth rate of 2.2% [2]. COVID-19 has created an enormous crisis for the cultural and creative industries. According to an OECD study, the value of cultural and creative industries in the EU27 and the UK fell by more than 30% in 2020, yet the global home/mobile entertainment market grew by 23% in 2020 [3]. As far as China is concerned, the value added by the CCIs decreased by 0.07% of the GDP compared to 2019, while the growth rate of the output value of the cultural entertainment and leisure industry decreased by 19.8%. However, the growth rate of the output value of the content creation and production industry to which the game industry belongs reached 11.1%, and it was the industry with the highest growth rate among the core areas and related fields of the entire cultural and creative industry [4].

Spatially, China’s CCIs are mainly kept together by developed clusters. With the ongoing promotion of strategic goals, such as cultural power and a sound cultural industry system, the knowledge spillover brought by the CCIs clusters has an irreplaceable role in promoting the optimization and upgrading of the regional economic structure, as well as enhancing the ability of the region to resist shocks and risks. At the same time, the economically developed regions in eastern China have solid economic resilience, and most of China’s mature cultural industry clusters are clustered in these regions. However, the current economic situation in China is severe, and the CCIs clusters suffer from problems such as weighty policy reliance and failure to give full play to the economy of scale effect. Therefore, identifying the relationship between CCIs clusters and regional economic resilience is a practical issue. Moreover, it is unclear whether there is a spatial interaction between CCIs clusters and regional economic resilience.

Relevant theories in evolutionary economics provide a good explanatory tool for the relationship between cultural and creative industries and economic resilience. CCIs can be combined with intra-regional economic development and local regeneration, which in turn can benefit CCIs [5]. Specialization in the field of creative innovation is considered a means to enhance regional resilience [6] while simultaneously making regions more attractive to CCIs [7].

Culture and creativity have received considerable attention as drivers of local development, with studies showing that regions with cultural and creative resources are better able to cope with natural disasters, structural change, and economic crises [8,9,10]. CCIs development can also provide a potential buffer against economic downturns [11]. CCIs have been known to increase resilience, stimulate investment, and serve as an essential source of improvement in the competitive advantages of cities and regions. Moreover, these achievements are often made possible by promoting sustainable economic endogenous growth that supports upgradable pathways [12,13,14], and there is a significant role in employment and long-term growth [15,16]. The OECD report shows that cultural and creative industries have significant spillover effects on the broader economy through forward–backward solid linkages supporting health and well-being, significant social impacts, and their roles as essential drivers of resilient economic recovery [3].

Cellini and Cuccia (2019), through a study of 20 Italian regions after a financial crisis, found that, the higher the supply and demand of cultural goods in a region was, the higher the economic resilience of the region [17]. However, some scholars argue that developing a cultural and creative economy suffers from sacrificing a community’s cultural policy in favor of a broader outreach to establish economic growth—often at the cost of forcing the impoverished out of their homes in favor of a better financed vision. Oakley (2015) gave several examples of this type of behavior on a global scale. He also noted a decline in creative entrepreneurship as perceptions of technology are downgraded and managerial expectations change, thus causing former solid positions to become unstable [18]. Other commentators on the development of CCIs indicate that it is taken for granted that the cultural and creative economy is “inherently clean”, but this assumption does not take into account the significant energy consumption and environmental pollution that are emitted from large centers and production sites in many CCIs sectors worldwide [19,20].

China’s CCIs tend to form clusters of cultural industry parks, and extensive literature confirms the critical role of CCIs on economic resilience. However, there is less discussion on CCIs’ specialized and diversified differences based on their roles, location, and mutual capabilities, which all contribute to economic resilience. From the perspective of industrial agglomeration, the roles of specialization and diversification in economic resilience are not the same. Studies have shown that specialization in the pursuit of economy of scale effects (producing more to increase profits) may bring about functional, cognitive, and political path-locking in a region where policies fail, thus resulting in less economic resilience [21]. Sector-specific specialization is not only vulnerable to shocks, but it is less able to change course and redirect economic recovery [22]. At the same time, the scale effects from a high degree of specialization implies higher economic efficiency and higher competitiveness, which may be less affected by recessions or unexpected downturns in the economy [23].

In contrast, diversification is seen as a risk-spreading portfolio strategy to protect regions from sudden shocks based on demand in specific sectors. This strategy has led to new growth theories [24]. Davies and Tonts (2010) argue that, the more diversified a region’s economy is, the more resilient its economy is [25]. However, diversification between sectors may lead to ineffective knowledge spillovers due to an excessively different knowledge base, which does not spread risk well. The debate between specialization and diversification has a long history, and the introduction of the related diversification concept has reconciled the differences [6,26]. In addition, relevant studies have shown that spatial spillover affects the clusters of CCIs and economic resilience [27,28,29], but the spatial interaction between the two is unclear.

Although the research on the influence of CCIs and industrial agglomeration on regional economic resilience has achieved fruitful results, and the influence of specialization and diversification within China’s clusters regarding economic resilience has provided both a theory and method for this study. However, some evidence of the shortcomings can be found in current research. First, the impact of regional economic resilience on CCIs clusters is not clear. It is also unclear whether CCI specialization or diversification is favored when the economic resilience of a region is enhanced. Second, existing CCIs studies focus on examples of regional economic resilience enhancement. Few studies can be found on CCIs’ two-way influence, spatial spillover, and interaction between CCIs clusters in regard to regional economic resilience from a spatial perspective, and there are fewer corresponding empirical tests. In this context, two main research questions are raised: (1) Is there a two-way influence relationship between CCIs clusters and regional economic resilience? (2) What is the spatial spillover of CCIs clusters and regional economic resilience, and what is the spatial interaction between them?

The synergism brought about by communities intricately connected to one another, both geographically and culturally, results in an agglomeration of artists, musicians, sculptors, and others who work with a variety of different media. They form clusters focusing on complementary cooperation with government, organizations or businesses, and that relationship nurtures the financial and cultural interests of all concerned. The clusters represented in this article achieved different levels of success (or possibly failure in some instances) based on the panel data of 31 provinces and cities in China (excluding Hong Kong, Macao, and Taiwan) from 1997 to 2020. This study analyzes bidirectional influence relationships, spatial spillover, and interaction effects between the clusters of CCIs and regional economic resilience with the help of a generalized, spatial, three-stage least-square (GS3SLS) estimation method, using a spatial simultaneous equation model.

The marginal contributions of this study are: First, it analyzes the bidirectional influence of CCIs clusters and regional economic resilience and verifies the inverse influence of regional economic resilience on these CCIs clusters. Second, this study analyzes the spatial spillover effect and regional interaction influence of CCIs clusters and regional economic resilience. The difference lies in the fact that the previous study was a spatial spillover test of CCIs clusters and regional economic elasticity, which were explained separately without emphasizing the regional interaction effects between them. The research results have reference significance for countries that want to enhance regional economic resilience through the development of CCIs clusters while ensuring the sustainable and healthy development of CCIs clusters.

The remainder of this study is organized as follows: Section 2 is the theoretical analysis and research framework. Section 3 defines the research design. Section 4 presents empirical results and discussion. Section 5 presents the conclusions or key takeaways, as well as policy inspiration.

2. Theoretical Analysis

2.1. Two-Way Impact Analysis of CCIs Clusters and Regional Economic Resilience

2.1.1. Two-Way Impact Analysis of Specialized CCIs Clusters and Regional Economic Resilience

From the perspective of specialized of CCIs clusters, firstly, specialized CCIs clusters help the survival and development of enterprises with a high degree of specialization, such as leading enterprises, enterprises that function above the economy of scale, and unicorn enterprises, including economies of scale and a work scope for each entity, which can reduce production costs, improve efficiency, and, thus, play a role in resisting risks and shocks [30]. Secondly, cultural and creative enterprises with a well-developed specialized labor system division can not only obtain more capital and financial support, but can also accumulate experience in “learning by doing” and carrying out technological research and development, as well as scientific and technological innovation to thereby achieve path breakthroughs, which enhance regional economic resilience. Finally, shocks can cause labor to leave, but specialized skills and human capital can shorten the unemployment period and contribute to more economic recovery [31].

From the perspective of regional economic resilience, a region with sufficient economic resilience can promote intra-regional factor correlation and economic efficiency [32]. First, the availability of production factors required for specialized CCIs clusters increases. Second, resilience is embedded in regional evolution as an inherent property of the region. It is a historical path dependence that is continuously self-reinforced by the influence of regional history and the external environment [32], i.e., a region with robust economic resilience can, to a certain extent, maintain its specialized CCIs clusters. Finally, regions with sufficient economic resilience can attract more labor [33], thus enabling a perfect division of the labor system, as specialized CCIs clusters reduce the training cost of labor.

Hypothesis 1a (H1a).

There is a two-way positive influence relationship between specialized clusters of CCIs and regional economic resilience.

2.1.2. Two-Way Impact Analysis of Diversification in CCIs Clusters and Regional Economic Resilience

From the perspective of specialized CCIs clusters, the diversity within CCIs clusters helps the development of gazelle enterprises and other types of enterprises with strong innovation ability, high development potential, and a fast growth rate, which nurtures a highly creative atmosphere and diversified CCIs business modes, which is not only conducive to spreading risks, but is also conducive to quickly finding regional development direction and achieving path breakthroughs. Secondly, diversified CCIs clusters can increase the interaction opportunities between the same industry and non-industry; promote the collision and overflow of ideas, technologies, and knowledge; turn shocks into opportunities for regional path breakthroughs; and show stronger resilience [34]. In contrast, the high value-added feature can promote the transformation and upgrading of other industries in the region [14] and improve the ability and resilience of the industry and the whole region to resist risks. Finally, the diversified clustering of CCIs can reduce the possibility of urban industrial rigidity by reducing transaction costs while increasing innovation opportunities [35,36].

From the perspective of regional economic resilience, regions must first have solid economic resilience, which can absorb more creative talents and create a work atmosphere that is conducive to developing diversified modes within the CCIs. The entrepreneurial spirit in the region can also promote diversified clustering of the CCIs [37]. Secondly, the concentration of creative talents produces a creative atmosphere, and the generation of local “buzz” will promote an increase in public demand for cultural consumption and enjoyment, making the development of diversified CCIs clusters a path that is essential for success. At the same time, creative knowledge is not entirely explicit, and regions with high economic resilience will significantly promote the exchange and dissemination of tacit knowledge, thereby promoting the diversified clustering of CCIs. However, during an economic recession, when regional economic resilience is not possible, it may bring permanent unemployment to a part of the labor force made up of creative talents, along with a decline in the level of R&D and investment in new technologies, and a decrease in productivity [31], which is not conducive to the development of CCIs and the diversified CCIs clusters.

Hypothesis 1b (H1b).

There is a two-way positive influence relationship between diversified CCIs clusters and regional economic resilience.

2.2. Spatial Spillover and Interaction between CCIs Clusters and Regional Economic Resilience

2.2.1. Analysis of Spatial Spillover Effects

No matter how specialized or diversified CCIs clusters are, the externality effect generates a spillover of transformative information and technology. An increase in networking and exchange platforms creates channels between regions that spreads knowledge, technology, and cooperative efforts to spread to other places [34]. In addition, in the process of establishing CCIs clusters, the depth of the specialized divisions of labor can promote both forward and backward extensions of the industrial chain, as well as the formation of industrial associations between regions [37]. In the CCIs clustering process of diversification, local governments may argue as they focus on growth figures featuring the added value of CCIs, which emphasizes the pursuit of quantity rather than quality, thereby crippling the diversified agglomeration of talent and innovation with forced uniformity. Finally, both specialized and diversified CCIs clusters can produce spillover effects and conflicts due to the cultural background of a spillover place [32]. Without relevancy or connections to the original industry and culture in a spillover place, forced uniformity may not be possible.

Some relevant studies have shown that economic resilience has a spatial spillover effects and can positively affect the economic resilience of neighboring regions [29]. Brad et al. [31] indicated that being surrounded by resilient regions is beneficial. The spatial spillover of economic resilience can occur through two primary channels. The first channel centers around interactions between regions. In a common economic context, competition in government planning and decision making between regions can promote the flow of production factors on a spatial scale. This type of integrated development allows regional economies to form larger-scale economies [38]. The second benefit of spatial spillover is due to the cross-regional and cross-sectoral flow of production factors. The flow of essential factors of production, such as talents, capital, and information between regions, is constantly being reorganized or optimized to form a region’s economic structure in a dynamic way, which continuously accumulates regional competitive advantages. In addition, “the first rich lead connects to the last rich lead” to create the spatial spillover effect of economic resilience. Thus, regions with solid economic resilience can drive their neighboring regions to improve their production efficiency [39], thereby creating a rapid economic development through a factor flow, such as human capital, which improves the economic resilience of neighboring regions.

Hypothesis 2 (H2).

There is a positive spatial spillover effect between CCIs clusters and regional economic resilience.

2.2.2. Analysis of Spatial Interaction

CCIs clusters’ spatial interactions and their effect on regional economic resilience includes three aspects. First, both specialized and diversified CCIs clusters can influence new CCIs cluster growth in neighboring areas through knowledge spillover, learning and imitation, technology support, and experience exchanges. This type of growth with industrial chain extensions affects the economic resilience of neighboring areas. However, the agglomerative development of local CCIs implies, to a certain extent, competition for production factors such as talent, technology, and the capital needed for the development of new CCIs in neighboring places. However, the outward flow of production factors in those places may not be conducive to risk resistance and economic recovery. Secondly, CCIs pay attention to intellectual property and copyrights, regardless of whether they are a part of a specialized or diversified CCIs clusters. The existence of plagiarism and incidents drawing attention to the problem has not only shocked local CCIs [34] but also made them pay more attention to the protection of intellectual property rights, which, to a certain extent, significantly reduces knowledge spillover. In other words, this type of protection can both increase and limit the use of innovative knowledge and economic resilience, depending on the desire of the innovators to commercialize their products. Finally, the third aspect is knowledge capacity. Some CCIs sub-sectors must have proper training resources to meet the needs of knowledge complexity and to train employees to reach certain learning thresholds. However, other neighboring places may not have the resources needed to bear the risk of industrial development based on the specialized training and facilities they cannot support [30,32]. In this case, the new knowledge and technology cannot be fully landed, applied, or even put on hold, thereby cancelling plans for startups and slowing down potential economic resilience. This type of disruption can also affect the economic resilience of neighboring areas if they put forth significant financial support prior to realizing the primary source of funding could not finance the specialized training and technology needed to set up an infrastructure compatible with the other area’s needs.

The spatial interaction of regional economic resilience on CCIs clusters has three problems. First, when the local economy is in recession or regional lockdown, neighboring places’ economic resilience may also decline under the same economic background, which only intensifies the spread of crisis and obstructs CCIs cluster development in neighboring areas. Secondly, from the perspective of factor flow, the enhancement of local economic resilience can accelerate the spillover effect of economic resilience, which also accelerates the flow of factors to neighboring places, thus promoting the agglomeration or formation of CCI clusters in cities or region in other towns or nearby counties. However, at the same time, factor flow may also increase the attractiveness of creative talents and capital to neighboring areas, thereby concentrating the production factors required for the CCIs cluster development throughout the region. However, some types of expansion may not be well received in some areas. Finally, the richness of local economic resilience may be attributed to the local government’s economic structure under good governance. Through learning and imitating the governance model, neighboring governments can also promote economic resilience by promoting new CCIs clusters in their spheres of influence.

Hypothesis 3 (H3).

Some local CCIs clusters may seem detrimental to economic resilience in adjoining areas; however, local economic resilience can develop a positive atmosphere that can help CCIs cluster expansion.

A research framework based on the hypotheses presented in this section is shown in Figure 1.

3. Materials and Methods

3.1. Study Area and Data Sources

The study area covered 31 provinces and cities in China (excluding Hong Kong, Macao, and Taiwan) from 1997 to 2020. The data were mainly obtained from the China Statistical Yearbook, the China Statistical Yearbook of Cultural Relics and Tourism, and the China Culture and Related Industries Statistical Yearbook. It should be noted that, in calculating economic resilience, based on the development trend of China’s economy, the research results of Li et al. (2019) [40] were borrowed and divided into five periods—all of which are contraction periods except for the expansion period from 1999 to 2007. The contraction periods were from 1992 to1999, 2007 to 2010, 2010 to 2019, and 2019 to 2020.

3.2. Variable Design

3.2.1. Regional Economic Resilience

The regional economic resilience measure under a major recessionary shock proposed by Martin (2019) [41] was borrowed and estimated for 31 provinces and cities in China (excluding Hong Kong, Macao, and Taiwan) by referring to the regional economic resilience measure by Liu et al. (2020) [42]. The method constructs a counterfactual assumption in the context of a national recession, i.e., each city or region is affected in the same way as the macro overall, and the national response is a comparative benchmark for each city or region. The measurement Equation (1) is expressed as:

$${res}_i^t=\frac{∆Y_i-∆E}{\left|∆E\right|}$$

(1)

$$∆Y_i=Y_i^t-Y_i^{t-k}$$

(2)

$$∆E=\left(\frac{Y_r^t-Y_r^{t-k}}{Y_r^{t-k}}\right)\ast Y_i^{t-k}$$

(3)

where ${res}_i^t$ denotes the economic resilience of region $i$ in year $t$, ${∆Y}_i$ is the actual economic performance of region $i$, and $∆E$ is the national economic performance. $∆Y_i$ is calculated by Equation (2), where $Y_i^t$ and $Y_i^{t-k}$ are the quantitative indicators of region $i$ in year $t$ and year $t–k$. $∆E$ is calculated in Equation (3), where $Y_r^t$ and $Y_r^{t-k}$ are nationwide volume indicators in year $t$ and year $t–k$. Both are measured using GDP.

3.2.2. CCI Clusters

The regional entropy is used to measure the degree of specialization of CCIs clusters and is calculated as [43]:

$$ccilq=\frac{x_{ij}/x_j}{\sum _j{x}_{ij}/\sum _j{x}_j}$$

(4)

where $x_{ij}$ denotes the employees of CCIs in region $j$, and $x_j$ denotes the total number of employees in all industries in region $j$.

The inverse of the Hirschman–Herfindahl index (HHI), ccivar, is used to measure the degree of diversity of CCIs clusters [44], which is calculated as follows with the same symbolic meaning as above.

$$ccivar=1/\sum x_{ij}^2$$

(5)

3.2.3. Control Variables

In addition to the above endogenous core explanatory variables, the following control variables were set. (1) The output value of primary and secondary industries is $fs$: The CCIs mainly belong to the tertiary industry, so the primary and secondary industries are controlled, and the output value of the primary and secondary industries is used for calculation [45]. (2) Degree of economic openness ($open$): The economy is not a closed system, and the degree of openness affects the strength of economic resilience, so the total import and export value is used for calculation [46]. (3) Human capital ($humcap$): Adequate human capital helps economic recovery and the clustering of CCIs; humcap is calculated by the number of college students [47]. (4) Financial support ($finance$): Government financial support is conducive to economic development and recovery and contributes to CCIs clusters; thus, finance is measured by total regional financial expenditure [48]. (5) Informatization ($info$): The development of regional informatization has a positive effect on economic growth. It also contributes to CCIs clusters; thus, info is measured by the total regional postal and telecommunications services [49]. (6) Economic development level ($pergdp$): The level of economic development represents the region’s economic base, and different economic bases reflect different responses to risks and shocks that are measured by per capita GDP [50]. (7) The level of urbanization ($city$): Urbanization can drive economic growth and is calculated using the urbanization rate [51]. (8) Foreign direct investment ($fdi$): This variable is said to lead to a higher degree of urban inclusion and is conducive to the clustering of CCIs [52]. (9) Investment ($invest$): This variable is focused on infrastructure construction. Perfect infrastructure helps the agglomeration and development of CCIs and is measured by the amount of fixed asset investment [53]. The variables are described as shown in

Table 1. Description of variables.

Symbol	Description	Mean	Standard Deviation	Min	Max
res	Economic resilience	0.118	0.366	−2.742	3.040
ccilq	The degree of specialization of CCIs clusters; calculated by regional entropy	1.112	0.753	0.225	6.223
ccivar	The degree of diversity of CCIs clusters; calculated by HHI	10.52	2.08	4.63	14.94
fs	The output value of primary and secondary industries	7.521	8.583	0.0461	48.76
open	Economic openness; calculated by total import and export values	0.811	1.706	0.0009	11.30
humcap	Human capital; calculated by the number of college students	5.942	5.020	0.032	24.92
finance	Government financial support; measured by total regional financial expenditures	2.572	2.803	0.0336	17.43
info	Regional informatization; measured by the total regional postal and telecommunications services	0.920	1.736	0.0011	20.83
pergdp	The economic development level; calculated by GDP per capita	3.242	2.838	0.2215	16.49
city	The level of urbanization; calculated by the urbanization rate	4.637	1.856	1.380	8.960
fdi	Amount of foreign direct investment	1.063	2.377	0.0011	27.45
invest	Investment in infrastructure construction; measured by the amount of fixed asset investment	8.959	11.84	0.0302	59.25

.

3.3. Model Setting

After defining the above theoretical analysis, a possible causal relationship was found between CCIs clusters and regional economic resilience. To solve the endogeneity problem and to consider the spatial spillover and interaction between endogenous variables, the generalized spatial three-stage least squares (GS3SLS) was used to construct a spatial simultaneous equations model for estimation. In addition, to enhance the credibility of the results, three-stage least squares (3SLS), excluding spatial interactions, was used as a robustness test.

A total of two sets of spatial simultaneous equations models were created to represent specialized CCI clusters. The following set is (6) and (7), which represent regional economic resilience and specialized CCI clusters; they are respectively given as:

$${res}_{it}={\alpha }_0+{\alpha }_1\sum _{j\ne 1}^n{\omega }_{ij}{res}_{jt}+{\alpha }_2\sum _{j\ne 1}^n{\omega }_{ij}{ccilq}_{jt}+{\alpha }_3{ccilq}_{jt}+\alpha X_{it}+{\epsilon }_{it}$$

(6)

$${ccilq}_{it}={\beta }_0+{\beta }_1\sum _{j\ne 1}^n{\omega }_{ij}{ccilq}_{jt}+{\beta }_2\sum _{j\ne 1}^n{\omega }_{ij}{res}_{jt}+{\beta }_3{res}_{jt}+\beta Z_{it}+{\eta }_{it}$$

(7)

where ${\alpha }_0$ and ${\beta }_0$ are constants, ${res}_{it}$ denotes the economic resilience of region $i$ in year $t$, and ${ccilq}_{it}$ denotes the degree of specialization of CCI clusters in region $i$ in year $t$. ${\alpha }_1$, ${\alpha }_2$, ${\beta }_1$ and ${\beta }_2$ are spatial correlation coefficients given to measure spatial spillover effects. ${\omega }_{ij}$ is the spatial relationship between regions. $X_{it}$ and $Z_{it}$ are control variables. $X_{it}$ includes the output value of primary and secondary industries, economic openness, human capital, financial support, informatization, economic development level, and urbanization level; $Z_{it}$ includes human capital, financial support, informatization, urban inclusion, and infrastructure construction. ${\epsilon }_{it}$ and ${\eta }_{it}$ are random disturbance terms.

The second set (8) and (9) represent the spatial simultaneous equations models of regional economic resilience and CCIs diversified clusters, which are respectively given as:

$${res}_{it}={\theta }_0+{\theta }_1\sum _{j\ne 1}^n{\omega }_{ij}{res}_{jt}+{\theta }_2\sum _{j\ne 1}^n{\omega }_{ij}{ccivar}_{jt}+{\theta }_3{ccivar}_{jt}+\theta X_{it}+{\mu }_{it}$$

(8)

$${ccivar}_{it}={\lambda }_0+{\lambda }_1\sum _{j\ne 1}^n{\omega }_{ij}{ccivar}_{jt}+{\lambda }_2\sum _{j\ne 1}^n{\omega }_{ij}{res}_{jt}+{\lambda }_3{res}_{jt}+\lambda Z_{it}+{\delta }_{it}$$

(9)

where ${\theta }_0$ and ${\lambda }_0$ are constants, ${res}_{it}$ denotes the economic resilience of region $i$ in year $t$, and ${ccivar}_{it}$ denotes the degree of diversity of CCIs clusters in region $i$ in year $t$. ${\theta }_1$, ${\theta }_2$, ${\lambda }_1$ and ${\lambda }_2$ are spatial correlation coefficients used to measure spatial spillover effects. ${\omega }_{ij}$ is the spatial relationship between regions. $X_{it}$ and $Z_{it}$ are control variables and contain the same variables as above. ${\mu }_{it}$ and ${\delta }_{it}$ are random perturbation terms.

Finally, considering the possible endogeneity problem of spatial weights, a normalized neighboring spatial weight matrix was selected in this study, i.e., ${\omega }_{ij}=1$ when region $i$ is adjacent to region $j,$ and ${\omega }_{ij}=0$ when region $i$ is not adjacent to region $j$ or $i=j$.

4. Results and Discussions

4.1. Results

Before conducting the spatial econometric analysis, the core explanatory variables needed to be tested for spatial autocorrelation. Spatial autocorrelation of the variables was measured using Moran scatter plots of panel data, and the results are shown in Figure 2, Figure 3 and Figure 4. Both CCIs clusters and regional economic resilience have significant spatial characteristics, which allows the subsequent spatial simultaneous equations model estimation to be carried out.

The results of the spatial simultaneous equations model of specialized CCIs clusters and regional economic resilience are shown in

Table 2. Estimation results of the association equation between specialized CCIs clusters and regional economic resilience.

	G3SLS		3SLS
	(1)	(2)	(3)	(4)
Variables	res	ccilq	res	ccilq
w.res	0.6753 ***	−6.8852 ***
	(8.99)	(−3.06)
w.ccilq	−0.3748 ***	1.3531 ***
	(−3.07)	(4.19)
ccilq	0.4394 ***		0.2842 *
	(2.77)		(1.82)
res		9.3040 ***		12.4176 ***
		(3.14)		(3.27)
fs	0.0127 **		0.0058
	(2.29)		(1.60)
open	−0.0558 *		−0.0388
	(−1.78)		(−1.24)
humcap	0.0025	0.0804	0.0057	0.1301 *
	(0.33)	(1.34)	(0.87)	(1.86)
Finance	−0.0175	0.3842 ***	−0.0481 ***	0.6864 ***
	(−1.32)	(2.98)	(−2.73)	(2.92)
Info	0.0026	−0.0817	0.0126	−0.1716 *
	(0.21)	(−1.09)	(0.85)	(−1.82)
pergdp	−0.0143		0.0209 **
	(−1.33)		(2.01)
city	−0.0268		−0.0247
	(−1.04)		(−0.97)
fdi		−0.0113		0.0027
		(−0.33)		(0.08)
Invest		−0.0831 **		−0.1221 **
		(−2.17)		(−2.36)
Constant	0.1032	−1.2245 **	−0.0852	−1.6379 *
	(1.43)	(−2.03)	(−0.82)	(−1.79)
Observations	744	744	744	744

Note: * p < 0.10, ** p < 0.05, *** p < 0.01.

. The results of Model 1 in Table 2 show that the estimated coefficient of the impact of specialized CCIs clusters on regional economic resilience was significantly positive, and the regional economic resilience will increase by 0.43% for every 1% point increase in specialized CCIs clusters. The estimated coefficient of the spatial lag in regional economic resilience was positive and significant at the 1% level. However, there was a significant negative spatial spillover of the specialized CCIs clusters in the neighboring regions’ economic resilience, and the estimated coefficient was significantly negative at the 1% level.

Model 2 results show that the effect of the regional economic resilience estimated coefficient on the specialized CCIs clusters was significantly positive, and for every 1% increase in regional economic resilience, the specialized CCIs clusters will increase by 9.3%. The coefficient of the spatial lag based on diversified CCIs clusters was significantly positive at the 1% level, whereas the coefficient of the spatial lag based on regional economic resilience was significantly harmful. Finally, the results of the generalized, spatial, three-stage least-square (GS3SLS) estimation method shows that the regression coefficients of specialized CCIs clusters and regional economic resilience were all significantly positive, which again indicated a two-way promotion effect between specialized CCIs clusters and regional economic resilience.

The results of the diversified CCIs clusters’ spatial simultaneous equations model and regional economic resilience model are shown in

Table 3. Estimated results of equation showing associations between diversified CCI clusters and regional economic resilience.

	G3SLS		3SLS
	(1)	(2)	(3)	(4)
Variables	res	ccivar	res	ccivar
w.res	0.740 ***	11.359 **
	(10.95)	(2.57)
w.ccivar	0.145 ***	0.596 ***
	(3.44)	(4.71)
ccivar	−0.234 ***		−0.1039 **
	(−4.10)		(−2.45)
res		−14.999 ***		16.6275
		(−2.58)		(1.23)
fs	0.011 ***		0.0146 ***
	(2.81)		(3.56)
open	−0.033 ***		−0.0097
	(−2.70)		(−0.89)
humcap	0.006	−0.125	−0.0066	0.5076 **
	(0.79)	(−1.06)	(−0.91)	(2.13)
fiance	−0.080 ***	−0.915 ***	−0.0877 ***	0.0607
	(−4.08)	(−3.59)	(−2.81)	(0.07)
info	0.029 **	0.240	0.0362 **	0.1275
	(2.21)	(1.54)	(2.22)	(0.40)
pergdp	−0.020 *		−0.0359 **
	(−1.95)		(−2.20)
city	−0.021		−0.0326 ***
	(−1.48)		(−2.63)
fdi		0.027		−0.0693
		(0.51)		(−0.80)
invest		0.144 **		−0.2455
		(2.04)		(−1.41)
Constant	1.222 ***	6.113 ***	1.6088 ***	7.5574 **
	(4.25)	(3.61)	(2.95)	(2.34)
Observations	744	744	744	744

Note: * p < 0.10, ** p < 0.05, *** p < 0.01.

. Model 1 in Table 3 shows that the estimated coefficient of the diversified CCIs clusters impact on regional economic resilience was significantly negative at the 1% level, and for every 1% increase in diversified CCIs clusters, regional economic resilience will decrease by 0.234%. The spatial lag term of regional economic resilience was significantly positive at the 1% level, and the spatial lag term coefficient of diversified CCIs clusters was significantly positive at the spatial lag term of regional economic resilience at the 1% level. The spatial lag diversified CCIs cluster coefficient was significantly positive at the 1% level. Model 2 shows that the estimated regional economic resilience coefficient’s impact on the diversified CCIs clusters was significantly negative at the 1% level. The spatial lag term of diversified CCIs clusters was significantly positive at the 1% level, and the regional economic resilience spatial lag term was significantly positive at the 5% level. Finally, the results of the 3SLS also show a significant negative effect of diversified CCIs clusters on regional economic resilience. However, the results of Model 4 compared with Model 2 indicate an endogeneity problem between regional economic resilience and diversified CCIs clusters. When the spatial factor and endogeneity problem are considered, regional economic resilience significantly affects diversified CCIs clusters.

4.2. Discussions

The research results presented herein show a positive two-way influence of specialized CCIs clusters and regional economic resilience. At the same time, our research also shows a negative two-way influence of diversified CCIs clusters and regional economic resilience. The findings support Hypothesis 1a (H1a) which affirms a two-way positive influence relationship between specialized CCIs clusters and regional economic resilience. However, the findings could not fully support the feasibility of H1b, which favored a two-way positive influence relationship between diversified CCIs clusters and regional economic resilience.

Our support for H1a is consistent with the conclusion that specialized CCIs clusters can generate scale effects, which effectively helps them withstand external shocks [23]. The lack of support for H1b is contrary to the conclusion that a diversified industrial structure was determined to be capable of absorbing, adjusting, responding to, and, ultimately, resolving external and internal shocks. (Cooke et al. [14], Bristow and Healy [30]). The possible reasons for this are: firstly, the development mode of China’s CCIs clusters takes park-based development as the main grip, which is, to a considerable extent, a planning and policy-dependent development. It causes specialized divisions of labor within the CCIs clusters to deepen, again and again, as the shared engrained standards of operation seamlessly strengthen the enterprises, thus causing production efficiency to continuously increase. However, specific technical barriers still exist, which can put cognitive distances between the diversified industries within the cultural industry of the park, thereby keeping the knowledge obtained by communication from being absorbed internally with high efficiency and preventing the advantages of the diversified CCIs clusters from being fully developed. Secondly, from a micro perspective, many small and micro enterprises in China’s cultural enterprises are weaker in attracting capital and talent, which naturally weakens their competitive status compared to the larger leading cultural enterprises. When risks and shocks come, small and micro enterprises are the first to be hit, and the progress in obtaining government financial support is slow compared with that of leading enterprises, which verifies that the diversification externalities of CCIs clusters in an underdeveloped region may not be able to deal with risks as well as the larger enterprises. Wang et al. [54] showed that leading firms have helped regions to survive this crisis by strengthening diversity when possible.

Thirdly, CCIs form an industry where artistic, technical, and economic creative capabilities work together. This industry also attaches great importance to intellectual property rights. Cultural enterprises with a well-developed specialized division of labor system can attract the creative class needed to develop cultural and creative industries. At the same time, with the formation of a creative atmosphere and an informal network among enterprises in the region, diversified communication and exchange are also made possible for individual enterprises, thus creating an overflow of specialized knowledge based on the predominant industry. Thus, when a research perspective embodied in new knowledge or specialized skills becomes part of the whole region, the scale effect generated by specialization is more prominent. Finally, the ongoing advancements in Internet communication or remote technology and the emergence of the attention economy has made the production of cultural and creative products more industrialized and standardized. Meanwhile, the pursuit of broadcasting volume and traffic flow seems to continually stream the concept of cultural and creative industries. This has helped establish the “cultural industry” from which the concept of cultural and creative industries was born. At the same time, the short video platform lowers the creation threshold of cultural products as the channels for disseminating cultural and creative products increase. However, even though cultural and creative products have become more abundant, the proportion of quality content seems to be decreasing. This decrease in quality reflects the effect of diversification overflow in cultural and creative industries in general, thereby creating the possible emergence of the locking effect brought on by poorly managed specialization.

The results show that the CCIs in regions with higher economic resilience tended to specialize in clustering, while those with lower economic resilience tended to diversify in clustering. For regions with higher economic resilience, the conditions varied when they were less affected by macro shocks and, thus, higher economic resilience eventually became evident. Such regions may be geographically isolated, and when developing CCIs, the government may focus on cultivating a single type of CCIs cluster or continue to support the traditional cultural industries, thereby forming specialized CCIs clusters that are customized to fit the talent and skills in the region. Other regions with diversified industrial structures and a relaxed policy based on the institutional environment have solid economic resilience; thus, the production factors required for CCIs clusters development will be greatly satisfied when the cultural market is mature and cultural demand is strong. This type of stabilized environment makes CCIs more inclined to specialize using their agglomeration of cluster-supported synergistic connections to gain competitive advantages.

Regions with low economic resilience are unlikely to provide the environment needed to create new companies and jobs or productivity-enhancing investments [55]. However, interventions are equally likely to exist. For one, resource-based provinces are less resilient, such as those in Northeast China, where industrial manufacturing is the dominant industry. Using the high value-added characteristics of CCIs clusters to integrate with traditional primary and secondary industries can achieve industrial transformation; moreover, upgrading is a common way to develop CCIs clusters in such regions. In these instances, cross-industry exchanges and knowledge flow promote the diversification of CCI clusters. Secondly, when developing CCIs clusters, planners must still recognize the fact that regions with weaker economic foundations cannot always provide the essential production factors needed to develop CCIs clusters. CCIs planners must also recognize that industrial transplants from other regions may not succeed if they are not sufficiently connected to the local cultural background and are not prepared to supplement the technological accumulation needed to make the collaboration function [30]. Finally, some CCIs planners fail to see that imitating or copying other strategies can lead to blind diversification, which leads to misconceptions that prevent the success they hoped for.

The study also found a spatial spillover effect between CCIs clusters and economic resilience, as well as a spatial interaction between the two, thus supporting the second hypothesis (H2) and verifying the conclusion of Li et al. (2022) that economic resilience has a spatial spillover [29]. The same is true of CCIs clusters [27]. In the context of economic integration, the economic resilience of a region is affected by changes in the economic resilience of the surrounding regions. The mutual flow of elements and information, the integrated development strategy of city clusters, and the synergistic cooperation between provinces and regions all promote the spatial spillover of regional economic resilience. In addition, both the specialized and diversified agglomeration of CCIs can have a positive diffusion effect on the clustering of CCIs in neighboring places. This is because the ways and means of developing cultural and creative industries are similar among Chinese provinces and cities, and the results of typical cases will serve as a model for other regions.

Finally, the authors found that the bidirectional spatial spillover effect of local CCIs diversified clusters combined with the regional economic resilience of neighboring regions was positive. In contrast, the bidirectional spatial spillover effect of specialized local CCIs clusters and regional economic resilience in neighboring regions was negative, which partially supports H3. The results of this study illustrate that the sources for enhancing local economic resilience are a specialized clustering of local CCIs and diversified CCIs clusters in neighboring regions, which is consistent with the fact that the current agglomeration-based development of China’s CCIs presents a development pattern of intra-regional intensive specialization and inter-regional complementary advantages, as well as synergistic innovation. However, after comparing the magnitude of regression coefficients, the authors found that the positive influence of local economic resilience on the specialized of local CCIs clusters was more significant than the negative spillover of economic resilience in neighboring regions. In other words, the development of specialized local CCIs clusters needs to enhance local economic resilience. At the same time, the negative influence of local economic resilience on the diversified local CCIs clusters is more significant than the positive spillover of local economic resilience; that is, the development of diversified local CCIs clusters needs to continuously amplify the knowledge spillover effect of diversified clusters, which can promote the diversified agglomeration of local CCIs clusters while enhancing the economic resilience of neighboring places.

5. Conclusions and Policy Implications

This study adapted new economic geography and evolutionary economics into the research perspective. The authors used the spatial simultaneous equations model to study the two-way influence between CCIs clusters and the regional economic resilience of 31 provinces and cities of China from 1997 to 2020. This article explicitly verifies the influential effects between the two CCIs externalities of specialization and diversification, and presents regional economic resilience from a spatial perspective. The following main conclusions were obtained: (1) There is a positive two-way influence between CCIs clusters specialization and regional economic resilience and a negative two-way influence between the diversification of CCIs clusters and regional economic resilience. (2) There is a positive spatial spillover effect of regional economic resilience and CCIs clusters. (3) Local specialized CCIs clusters and diversified CCIs clusters in neighboring places can enhance local economic resilience.

The study has specific policy implications for current CCIs clusters development in China and the enhancement of regional economic resilience. First, from the perspective of CCIs clusters, the authors promote the synergistic regional development of CCIs clusters, thereby forming a network pattern with a high level of intra-regional intensification and specialization, inter-regional complementary advantages and synergistic innovation, and a more resilient economic structure. On the one hand, the synergetic development channels favoring inter-regional CCIs clusters should be unblocked so that cultural resources and production factors can flow and integrate freely between regions to realize the optimal allocation of resources in the cultural market on a macro scale and the smoothing of domestic circulation. On the other hand, it is also necessary to promote the integrated development and cooperation of regional CCIs while following the economy of scale. It is also necessary to promote the integrated development and cooperation of regional CCIs based on the CCIs regional characteristics while following the economies of scale. At the same time, a spatial pattern of dislocation and agglomeration becomes obvious, as it enhances industrial resilience and improves regional economic resilience through its spatial spillover.

Secondly, from the perspective of regional economic resilience, under the strategic guidance of “more resilience”, the goal is to promote the deep integration of CCIs with regional development, thereby giving full play to the overflow advantages of CCIs clustering and exploring the establishment of CCIs virtual clusters that are not restricted by geographical distance. By promoting knowledge diffusion and CCIs cluster technological innovation, a knowledge base with regional cultural characteristics is formed, and the enhancement of regional economic resilience drives the enhancement of CCIs cluster resilience. Meanwhile, with the help of big data and cloud computing, regional knowledge networks and CCIs value networks are constructed to promote close connections, efficient communication, and a perfect specialized division of the labor system. A regional knowledge network and a CCIs clusters value network can be established with the help of the latest technology, including data and cloud computing, to afford all the benefits just mentioned, along with the establishment of virtual CCIs clusters with high resource utilization, strong risk resistance, and green/clean sustainability.

Finally, the research results also have certain limitations, and a follow-up study will be carried out by completing three tasks: (1) refine the granularity of the research objects, (2) expand from provincial data to municipal data, and (3) construct the spatial weight matrix of the knowledge base and creative network to more accurately portray the influence relationship between clustering CCIs and regional economic resilience.