Regional Innovation and Sustainable Development Interplay: Analyzing the Spatial Externalities of Domestic Demand in the New Development Paradigm

Abstract

Building a great modern socialist country in all respects requires enhancing innovation capacity and establishing a new development pattern, especially in the context of sustainable development. This paper begins by analyzing the theoretical relationship between innovation and the spatial externality of domestic demand, constructing a theoretical model, and then empirically testing this model using provincial panel data from 2012 to 2020 through the Spatial Durbin model. The study underscores the importance of innovation in promoting sustainable economic growth, highlighting how it expands domestic demand through both supply and demand sides and positively affects the domestic demand in surrounding areas through spatial spillover effects. The empirical results reveal that innovation significantly boosts the level of domestic demand in the region and its environs, with the spatial spillover effect of domestic demand constituting 66.92% of the total effect. This underscores the relevance of spatial externality in sustainable economic planning. Innovation mainly stimulates domestic demand through consumption, aligning with sustainable consumption patterns, while exerting a moderate inhibitory effect on investment demand. The spatial externality of investment demand appears less significant. Overall, innovation drives the spatial externality of China’s domestic demand and significantly contributes to establishing a new development pattern of “dual circulation”, primarily focusing on the domestic cycle, within a framework of sustainable development. The paper concludes with policy recommendations that align innovation strategies with sustainable development goals.

1. Introduction

As China’s economy enters a new development stage, the report of the 20th National Congress of the Communist Party of China sets out the overall goal of “achieving high-level technological self-reliance, entering the forefront of innovative countries; building a modern economic system and forming a new development pattern”. This goal not only emphasizes the importance of independent innovation but also proposes the strategic direction of constructing a new development pattern with domestic circulation as the main focus and the mutual promotion of domestic and international circulations. In light of this context, investigating how innovation can stimulate domestic demand and generate spatial externalities within the framework of innovation-driven development holds significant importance for China in realizing the new dual circulation development pattern.

The early discussions on the concept of “dual circulation” can be traced back to 2008, when perspectives suggested that China should reconsider its “catch-up” industrialization strategy. The proposal emphasized reshaping the industrial landscape, enhancing innovation capabilities, and utilizing the domestic market as a driving force for economic growth. It strategically advocated forming a mutually beneficial division of labor with resource-supplying countries. These viewpoints laid the foundation for the later development paradigm of “placing domestic circulation as the mainstay”. This strategic emphasis on high-level technological self-reliance and the development of a new dual circulation development pattern not only signifies a shift towards enhancing China’s innovation-driven economic growth but also aligns with the global imperative for sustainable development. By fostering an innovation-led economy, China aims to reduce its environmental footprint, promote green technologies, and ensure economic resilience, thus contributing to the environmental, cultural, and economic dimensions of sustainability. The pursuit of self-reliance in technology and innovation, coupled with the emphasis on domestic circulation, presents a unique opportunity to harmonize economic growth with environmental conservation and social well-being. This approach underscores the necessity of integrating sustainable practices into the heart of China’s development strategy, aiming to create a balanced and sustainable future that addresses the dual challenges of economic development and sustainability.

Against this development background, our research revolves around the following core question: how does innovation affect domestic demand within the large domestic cycle, and can this effect be transmitted to surrounding regions through spatial spillover effects, thereby promoting the overall uplift of regional economies? Based on this, we propose the following hypothesis: innovation activities can not only directly enhance the consumption and investment demand within a region but can also generate positive spatial spillover effects by improving economic links and interactions between regions, further promoting the expansion of domestic demand and comprehensive economic development. This study aims to verify the direct role of innovation in promoting domestic demand and its indirect role in inter-regional dissemination, which is of great theoretical and practical significance for understanding and optimizing the large domestic cycle and promoting the mutual reinforcement of domestic and international dual circulations.

2. Literature Review

(1)

Innovation’s Role in Economic and Social Advancement

Recently, with the increasing emphasis of the Chinese government on the role of innovation in economic and social development, innovation has become a hot topic in the academic community. It is well-known that innovation in knowledge and technology is a key driving force for economic development and is considered one of the core elements of national and corporate competitiveness [1,2]. At the national level, there are also diverse factors influencing innovation performance. Ding, using the NCA analysis method, reveals that research and development investment, globalization, and national wealth are necessary conditions for innovation performance [3]. Shi’s research focuses on seven fiscally decentralized countries, such as the United Kingdom and the United States, and finds that targeted fiscal policies, such as research and development expenditures and public debt management, have greatly changed the technological innovation landscape of these countries [4]. Mao’s research finds that industrial innovation affects carbon prices, which are directly related to socioeconomic development [5]. Appiah, using 29 emerging countries as samples, finds a negative correlation between a country’s innovation input and carbon emissions, highlighting the importance of innovation in environmental governance [6]. Xie’s research on coal mining enterprises shows that innovation can reduce the cost of safety rectification, which is of great significance for maintaining social stability [7]. Litina et al. investigates the impact of product market policies on innovation and concludes that regulations that clarify negative behaviors have a positive impact on innovation, while those that clarify negative behaviors have a negative impact on innovation [8]. Shi believes that the digital economy has enhanced the resilience of cities through technological innovation and has generated positive spatial spillover effects, bringing many benefits to society [9]. Through the analysis above, it can be seen that whether at the enterprise level or the national level, a deep understanding and effective management of innovation are crucial for promoting economic development and social progress. However, there is currently limited scholarly exploration into the relationship between innovation and the construction of a new development pattern. Exploring this field is essential for understanding and promoting the role of innovation in future socioeconomic development.

(2)

How Innovation Influences China’s Construction of a New Development Pattern

In exploring how innovation influences the topic of China constructing a new development pattern, Chinese scholars have conducted extensive and in-depth research. The prevailing view is that enhancing regional innovation capacity and achieving innovative development are the key paths to building the “dual-cycle” new development pattern. This perspective is based on the notion that the improvement in innovation capacity will establish a foundation for high-quality mutual promotion and a dynamic balance between domestic demand and supply, thereby enhancing the dynamism of China’s economic cycles. Jiang and Meng emphasize that accelerating independent innovation is a necessary path to drive high-quality domestic cycles [10]. They point out that this involves not only the enhancement of the demand side but also the strengthening of the real economy on the supply side through the cultivation of emerging industrial chains, thereby providing impetus for the sustained and healthy development of the socioeconomic system [11,12].

From the perspective of political economics, economic cycles are defined as the process of the movement of a country’s material and service products in the stages of production, distribution, circulation, and consumption [13] to achieve value increment and enhance the well-being of national consumption. In this framework, technological innovation is regarded as the foundation for achieving these goals [14,15]. Therefore, by enhancing regional innovation capacity and improving people’s well-being, it ultimately promotes the realization of the “dual-cycle” new development pattern. On this basis, another group of scholars believes that innovation plays a promoting role in domestic demand through enterprises providing new products to the market and raising the national income level. They explore the impact of innovation on the domestic cycle from the perspective of domestic demand. The overall idea is that the improvement in innovation capacity can lead to the expansion of domestic demand by improving the circulation effect of the internal cycle [16,17]. Qian and Xiang argue that innovation can stimulate the potential of domestic demand by creating high-quality products and services, thus expanding the domestic cycle [18]. Ren and Gong emphasize that core technological innovation is crucial for constructing a secure and independent domestic cycle [19]. Domestic demand, including consumption demand and investment demand, is a core element of this process [20]. From the perspective of consumption demand, Wang and Meng argue, from the perspective of supply-side reform, that technological innovation promotes cultural consumption in China [21], while Xu and Song find that corporate innovation behavior positively drives consumer behavior [22]. Wang also points out, from the perspective of rural service industries, that innovative service formats have a significant positive impact on consumption upgrading [23]. From the perspective of investment demand, Ye, starting from the micro perspective of manufacturing enterprises, finds that enterprises with higher levels of technological innovation bring about greater investment demand [24].

(3)

Innovation’s Spatial Externality on Domestic Demand

Although the mechanism by which innovation promotes domestic demand is widely recognized, there has not been in-depth research on the spatial externality of innovation in promoting domestic demand. This indicates that, despite the valuable insights provided by the existing literature on the impact of innovation on China’s new development pattern, there is still room for further exploration in this field. The existence of spatial externality is mainly due to imperfect competition, increasing returns, financial externality under market interaction, and market imperfections and coordination failures in spatial economics. The spatial externality of domestic demand refers to the influence of domestic demand in a local area on the domestic demand of surrounding areas during the expansion process. In the research field of the spatial externality of consumption demand, scholars have achieved a series of significant results covering various scenarios, demonstrating the impact and importance of the spatial externality of consumption demand in different contexts. Firstly, Burlig’s research focuses on the spatial externality of groundwater extraction in California agriculture, highlighting the critical role of externality in resource management. This study not only reveals the potential impact of groundwater extraction on surrounding regions but also provides empirical evidence for understanding spatial externality in natural resource utilization [25]. Next, Ten Kate proposes an innovative framework for understanding the dynamics of consumer demand in bilateral markets. By modeling the externality between customer groups and its impact on the demand and price sensitivity of both sides of the market, this framework provides a new theoretical perspective for analyzing consumer behavior in complex market structures [26]. Vitali’s research focuses on how consumer information friction drives businesses to choose locations within a city. This study combines various data from Uganda’s clothing industry, including transaction, customer, and mystery shopper data, revealing the significant role of information friction in influencing market positioning and consumer decision-making [27]. Additionally, Cohen provides a theoretical and empirical framework focusing on measuring and evaluating the impact of agglomeration economies and industry location decisions. In particular, this research focuses on spillover effects related to knowledge or other types of horizontal and vertical externality, providing profound insights into understanding industry agglomeration and its mechanisms’ impact on the market [28]. Zhang and Liu use the spatial Durbin model to study the impact of specialization and diversification agglomeration in the service industry on consumption structure. They find that both forms of agglomeration show a significant positive correlation with consumption structure in space, providing important empirical evidence for understanding the impact of service industry development on consumer behavior [29]. Finally, Qi and Zhang’s research focuses on the impact of the common prosperity goal on China, specifically the spatial spillover effects of consumption growth in the initially prosperous regions on the consumption growth rate in the later prosperous regions. This finding emphasizes the impact of regional development imbalances on consumption demand, providing an important perspective for understanding and addressing regional economic disparities [30]. Overall, these studies deepen our understanding of the spatial externality of consumption demand from multiple perspectives and offer valuable insights for policy-making and business strategies.

Regarding the research on the spatial externality of investment demand, another group of scholars has achieved certain research results. Bertinelli’s study tests the positive spatial autocorrelation of R&D investment and how the local environment influences firms’ decisions in R&D investment [31]. Häckner’s research extends the Hotelling spatial duopoly competition model to include R&D investment, analyzing how cost reduction through individual investment and spillover effects from competitors affect firms’ location choices and agglomeration [32]. Bode employs the System Generalized Method of Moments (GMM) estimation method to consider the potential endogeneity and spatial lags of foreign direct investment. Using data from U.S. states from 1977 to 2003, the results indicate positive external effects from foreign direct investment, while domestic firms’ external effects are negative [33]. In summary, the existing research results indicate the existence of spatial externality in China’s consumption and investment demands. However, the mechanisms through which innovation influences the spatial externality of domestic demand still need further clarification, and specific mechanisms need to be constructed. Additionally, the above-mentioned studies have not examined the spatial externality of China’s domestic demand from the perspective of the new development pattern. Based on this, building on existing research outcomes, this paper may contribute by: (1) examining the spatial externality of China’s domestic demand under innovation from a spatial econometrics perspective. Compared to existing research, this paper (2) enriches theoretical achievements and empirically verifies the promoting role of innovation in constructing a domestic large cycle when examining the spatial externality of domestic demand, further refining the concept of domestic demand, systematically explaining the promoting effect of innovation on consumption and investment demands, and constructing a theoretical model that expands the research perspectives and models of both, thus broadening the understanding of the interaction between innovation and domestic demand expansion.

3. Theoretical Modeling

The new double-cycle development pattern is an economic modernization strategy, characterized by innovation-driven unimpeded economic circulation, with scientific and technological self-reliance and independent innovation as its essential features. In August 2020, General Secretary Xi Jinping emphasized the urgent need to strengthen our capacity for independent innovation and achieve breakthroughs in key core technologies as soon as possible. This is a major issue related to the overall situation of our country’s development and it is also the key to the formation of a domestic macrocycle as the main body. The above discussion provides a theoretical basis for this paper to construct an evolutionary model of regional innovation and domestic general circulation. Regional innovation capacity serves as a basis for building a new development pattern. The specific theoretical model is shown in Figure 1.

The new economic geography believes that current regional economic activities have surpassed the traditional assumption in theories where interregional trade transportation costs are zero. It considers the existence of transportation costs and spatial spillovers in interregional economic activities. That is, the development of transportation infrastructure will not only affect the local economy but also have an impact on surrounding areas along the transportation infrastructure network. Therefore, only by comprehensively considering the effects of economic activities on both the local area and surrounding areas can the operational mode of economic activities be more fully revealed. As shown in Figure 1, within the theoretical framework of the impact of regional innovation on domestic demand, the operational mode of regional innovation includes the impact on both domestic demand in the local area and domestic demand in surrounding areas through spatial externalities. This impact is based on the high-level development of China’s transportation infrastructure. The impact of the improvement in regional innovation capacity on domestic demand in the local area can be examined from both the supply side and the demand side. On the supply side, innovation activities expand the market size by providing new products to the market, nurturing the industrial chain, and strengthening the real economy, laying the foundation for building a unified domestic market and providing impetus for promoting internal circulation. On the demand side, innovation increases residents’ income levels and utilizes China’s advantage of having the world’s largest middle-income group, unleashing the huge potential of China’s domestic demand market. Under the dual effects of the supply side and the demand side, consumption demand and investment demand in China are driven to expand, ultimately leading to the expansion of domestic demand in the local area. Meanwhile, due to the rapid development of China’s transportation infrastructure, the costs of the cross-regional flow of various production and innovation factors have been significantly reduced. Therefore, when domestic demand in surrounding areas cannot be met by the local market, it will spread to surrounding areas to seek release channels. The products provided by local innovation activities and the increased income not only meet the domestic demand of the local area but also provide direction for the release of domestic demand in surrounding areas. That is, the improvement in local innovation capacity spreads along interregional transportation networks to surrounding areas, generating externalities for domestic demand in surrounding areas. In summary, within the framework of new economic geography, considering the combined impact of regional innovation on domestic demand in both the local area and surrounding areas helps to more comprehensively reveal the spatial externalities of innovation demand.

4. Research Design

4.1. Selection of Variables

Explained variable: Domestic demand (DD) serves as the foundation of the domestic macrocycle, aiming to unleash the potential of China’s domestic demand and achieve high-quality economic development. During benchmark regression analysis, the level of regional GDP is utilized to represent the level of domestic demand. Furthermore, in the mechanism test, consumption demand and investment demand are examined separately to assess the mechanism of promoting domestic demand through innovation capacity. Consumption demand (cons) is quantified by per capita consumption expenditure, while investment demand (invest) is quantified by social investment in fixed assets.

Explanatory variable: This study focuses on regional innovation (inno) and its relationship with internal circulation. Within the framework of the new development pattern, innovation stimulates internal circulation by introducing new products to the market. As such, the sales of new products are used to assess regional innovation capacity, with technology market turnover serving as the measurement.

Control variables: Various factors influence the construction of the domestic macrocycle, including the impact of the foreign cycle. Import and export levels (inandout) are measured by the total import and export volumes of each province, while foreign direct investment (FDI) levels are determined by the amount of FDI in each province. Urbanization levels (urban) are quantified by the proportion of the urban population to the total population at the end of each province’s year, and government support levels (gov) are assessed by each province’s government’s financial expenditures.

4.2. Data Sources and Processing

A sample of 31 provinces (autonomous regions and municipalities directly under the central government) in China from 2011 to 2020 was selected, excluding Hong Kong, Macao, and Taiwan. Meanwhile, to ensure the authenticity of the data, the study adopts all the data from the China Statistical Yearbook of each year, and part of the missing data is made up by looking up the Provincial Statistical Yearbook of each province in that year, as well as by using the linear interpolation method. Given the varied sources of data acquisition in this paper, both the scale and significance differ. Therefore, to calculate the levels of common prosperity and high-quality economic development, normalization is necessary to eliminate differences in the quantitative representation of each indicator. The data normalization formula is presented in the following equation:

$$a_{xy}^{ \prime }=\frac{a_{xy}-\underset{x}{\mathit{min}}{a}_{xy}}{\underset{x}{\mathit{max}}{a}_{xy}-\underset{x}{\mathit{min}}{a}_{xy}}\times 99+1$$

(1)

4.3. Description of Spatial Matrix

The first theorem of geography posits a correlation among all phenomena, with proximity generally indicating stronger correlations. Therefore, in spatial measurement research, it is common practice to establish a spatial weight matrix to quantify the distances between phenomena. This paper adopts the adjacency matrix, geographic distance matrix, and economic distance matrix as the spatial weighting matrices for empirical research, respectively. The adjacency matrix considers that there is a connection between subjects with common boundaries or vertices, and there is no connection between subjects without common boundaries or vertices.

The specific elements of the matrix are set, as shown in

Table 1. Spatial weight matrix.

Matrix Name	Element Setting	Explanation
Adjacency Matri (w1)	If there is a common border between province i and province j, the element is set to 1; otherwise, it is set to 0.	When i = j, wij = 0
Geographic Distance Matrix (w2)	The reciprocal of the geographic distance between province i and province j.
Economic Distance Matrix (w3)	The reciprocal of the absolute difference in GDP between province i and province j in the current year.

below:

4.4. Spatial Correlation Test

To test the spatial autocorrelation of China’s domestic demand, this paper selects the domestic cycles of 31 provinces (cities and districts) in China from 2011 to 2020, which are discriminated by Moran’s I index with the following formula:

$${I_{ic}}^t=\frac{n\times \sum _{i=1}^n\sum _{j=1}^n{w}_{ij}({{ic}_i}^t-\overline{ic})({{ic}_j}^t-\overline{ic})}{\sum _{i=1}^n\sum _{j=1}^n{w}_{ij}{\sum }_{i=1}^n{({{ic}_i}^t-\overline{ic})}^2}$$

(2)

Formula, ${I_{ic}}^t$ represents the Moran’s I index of domestic demand in 31 provinces (cities and districts) of China in year t; t = 2011, 2012, …, 2020.

i, j are defined as above. $w_{ij}$ represents the elements of the spatial matrix for province i and province j. The element settings are shown in Table 1, and $\overline{ic}$ represents the average of the domestic demand in each province in year t. The results of the Moran’s I index test for domestic demand for each year are shown in

Table 2. Moran’s I index of domestic demand in China for each year.

Year	Adjacency Matrix w1			Geographic Distance Matrix w2			Economic Distance Matrix w3
	I	Z	p	I	Z	p	I	Z	p
2011	0.279	2.816	0.002	0.037	2.108	0.018	0.227	3.219	0.001
2012	0.275	2.776	0.003	0.036	2.075	0.019	0.227	3.210	0.001
2013	0.275	2.776	0.003	0.035	2.061	0.020	0.224	3.186	0.001
2014	0.275	2.783	0.003	0.035	2.060	0.020	0.215	3.078	0.001
2015	0.282	2.846	0.002	0.036	2.102	0.018	0.204	2.944	0.002
2016	0.296	2.985	0.001	0.041	2.241	0.013	0.146	2.232	0.013
2017	0.296	2.987	0.001	0.041	2.265	0.012	0.135	2.102	0.018
2018	0.295	2.976	0.001	0.043	2.306	0.011	0.312	2.062	0.020
2019	0.309	3.101	0.001	0.056	2.697	0.003	0.112	1.805	0.036
2020	0.312	3.135	0.001	0.055	2.685	0.004	0.110	1.791	0.037

below:

As shown in Table 2, Moran’s I index, calculated with three kinds of spatial weight matrices, is significantly positive, indicating that there is a positive spatial correlation in China’s domestic demand, i.e., the domestic demand of each region shows a “high-high” agglomeration or “low-low” agglomeration, and the gap in the domestic demand between the neighboring regions is relatively small. In this case, the Moran’s I index calculated by constructing a spatial matrix based on geographic location criteria increased over time, and the Moran’s I index calculated by constructing a spatial weighting matrix based on the level of economic development decreased slightly over time. This shows that in recent years, economic exchanges among regions in China have predominantly been influenced by geographic proximity. Closer geographic distances between regions correlate with more frequent exchanges, leading to a decreased significance of inter-regional economic disparities in these exchanges.

4.5. Model Setup

This paper sets up the spatial measurement model as follows:

$$ic={\alpha }_0+{\beta }_0\times {inno}_{it}+{\beta }_1\times w\times {inno}_{it}+{\delta }_0\times C_{it}+{\delta }_1\times w\times C_{it}+{\mu }_i+{\theta }_t+{\lambda }_{it}$$

(3)

In Equation (3), ic is the explanatory variable; inno is the explanatory variable, ${\beta }_0$ is the coefficient of the explanatory variable, and ${\beta }_1$ is the spatial coefficient of the explanatory variable; w is the spatial weight matrix; C is the control variable, ${\delta }_0$ is the control variable coefficient, and ${\delta }_1$ is the control variable spatial coefficient; ${\mu }_i$ stands for individual fixed effects; ${\theta }_t$ stands for time fixed effects; and ${\lambda }_{it}$ stands for double fixed effects. In practical tests, specific spatial econometric models also need to be identified through Wald tests and LM tests.

5. Model Regression and Analysis of Results

5.1. Model Selection

The presence of spatial effects was first determined by the LM test and Robust LM test, and the use of the spatial Durbin model was determined by the Wald test. The results of the tests are shown in

Table 3. Spatial econometric model selection results.

Test Statistic	W1		W2		W3
	Test Value	p-Value	Test Value	p-Value	Test Value	p-Value
LM Error	0.186	0.666	0.221	0.638	0.204	0.651
Robust LM Error	0.280	0.597	0.725	0.394	7.266	0.007
LM Lag	6.844	0.009	3.349	0.067	5.340	0.021
Robust LM Lag	6.937	0.008	3.853	0.050	12.402	0.000
Wald Spatial Lag	23.37	0.0015	47.52	0.0000	58.21	0.0000
Wald Spatial Error	124.78	0.0000	39.28	0.0000	37.66	0.0000

.

From Table 2 above, the comprehensive three spatial weight matrix results, the LM test, and the Robust LM test indicate that spatial econometric analysis should be carried out, and the Wald test results indicate that the test results of the spatial lag model and spatial error model satisfy the requirement of 1% significance, i.e., spatial Durbin model will not degenerate into the spatial lag model and spatial error model; therefore, this paper chooses the spatial Durbin model to test the data.

5.2. Baseline Regression Results

The results of benchmark regression are shown in

Table 4. Spatial Durbin model benchmark regression results.

Variables	W1	W2	W3
	SDM	SDM	SDM
	DD	DD	DD
Inno	0.1811 *** (0.0321)	0.1813 *** (0.0328)	0.1081*** (0.0333)
Inandout	0.1765 *** (0.0619)	0.1706 *** (0.0594)	0.2295 *** (0.0617)
Fdi	−0.0451 *** (0.0103)	−0.0442 *** (0.0104)	−0.0190 * (0.0109)
Urban	0.3334 *** (0.0680)	0.2409 *** (0.6677)	0.3604 *** (0.0709)
Gov	0.6064 *** (0.0313)	0.6519 *** (0.0298)	0.6585 *** (0.0322)
W×Inno	−0.1161 * (0.0631)	−0.4085 ** (0.2007)	0.2565 ** (0.1052)
W×Inandout	0.2389 ** (0.1138)	0.5501 * (0.3222)	0.2284 (0.1647)
W×Fdi	0.0678 *** (0.0207)	0.0660 (0.0669)	0.0602 ** (0.0248)
W×Urban	−0.2698 *** (0.0932)	0.0328 (0.3789)	−0.9321 *** (0.2109)
W×gov	0.2462 *** (0.0857)	1.5310 *** (0.2757)	0.0298 (0.1531)
ρ	0.1156 (0.0777)	−0.2808 (0.2232)	−0.2284 ** (0.1095)
σ2	3.4777 *** (0.2783)	3.3262 *** (0.2651)	3.5222 *** (0.2838)
Fixed/Random	Fixed	Fixed	Fixed
OBS	310	310	310
R2	0.7222	0.6299	0.6300

Note: *, **, and *** indicate that the variables are significant at the 10%, 5%, and 1% levels, respectively, with standard errors in parentheses. W1 is the adjacency matrix, W2 is the geographic distance matrix, and W3 is the economic distance matrix, as follows.

below.

As shown in Table 4, the regression results of the model using the adjacency matrix and the geographic distance matrix as the spatial weight matrix show that the coefficient of the impact of innovation on the promotion effect of domestic demand in the region is significantly positive at the 1% level. In contrast, the coefficient of the spatial spillover effect of innovation on domestic demand is negative.

The results show that innovation significantly contributes to the level of domestic demand in the region under the conditions of the adjacency matrix and the geographic distance matrix as a spatial weighting matrix, while innovative activities in the neighboring regions have a certain inhibiting effect on the domestic demand in the region. For the economic distance matrix, it can be seen that the promotion effect of innovation on domestic demand is significantly positive at the 1% level, and innovation significantly enhances the level of domestic demand in the region, while the coefficient of the spatial spillover effect of innovation on domestic demand meets the requirement of significance and is positive, which indicates that the innovation behavior of the surrounding areas also has a significant effect on the promotion of the level of domestic demand in the region.

The above results show that when the spatial externality of innovation on domestic demand is measured by geographic location, the spatial externality of innovation on domestic demand is negative under the adjacency matrix and the geographic distance matrix, because when the innovation results of the region have a positive effect on the domestic demand of the region, they will generate spatial spillover effects along the transportation network to the surrounding regions, attracting the domestic demand of the surrounding regions to be transferred to the region, thereby inhibiting the domestic demand of the surrounding regions, which mainly unfolds along the transportation infrastructure network.

When economic development levels serve as the spatial weighting matrix, inter-regional exchanges primarily depend on economic development levels. Closer exchanges occur between regions with similar economic development levels. Innovation drives domestic demand expansion within regions, attracting the introduction of innovative technologies from similarly developed regions, thus further stimulating domestic demand.

5.3. Decomposition of Spatial Effects

The analysis results indicate that innovation plays a significant role in promoting domestic demand within a region, and there is considerable spatial externality present. Here, partial differential equations are used to further calculate the direct, indirect, and total effects of innovation on domestic demand. Due to space constraints, only the decomposed effects under the economic distance matrix are listed, as shown in

Table 5. Spatial effects decomposition.

Variables	Direct Effect	Indirect Effect	Total Effect
	W3	W3	W3
Inno	0.1020 *** (0.0341)	0.2045 ** (0.0928)	0.3065 *** (0.1004)
Inandout	0.2212 *** (0.0596)	0.1399 (0.1398)	0.3611 ** (0.1516)
Fdi	−0.0198 * (0.0106)	0.0546 *** (0.0209)	0.0349 (0.0231)
Urban	0.3909 *** (0.0702)	−0.8455 *** (0.1819)	−0.4545 *** (0.1703)
Gov	0.6623 *** (0.0298)	−0.1033 (0.0951)	0.5590 *** (0.1060)

Note: *, **, and *** signify that the variables are significant at the 10%, 5%, and 1% levels, respectively, with standard errors in parentheses. W1 represents the adjacency matrix, W2 is the geographic distance matrix, and W3 is the economic distance matrix, as stated below.

.

From the perspective of the core explanatory variables, the direct effect coefficient of regional innovation is 0.1020, which is significant at the 1% confidence level, indicating that innovation significantly promotes domestic demand in the region. The indirect effect coefficient of regional innovation is 0.2045, which is significant at the 5% confidence level, suggesting that innovation also has significant spatial externality on the domestic demand of surrounding areas. Further analysis reveals that the total effect coefficient of regional innovation is 0.3065, with the spillover effect of innovation on domestic demand accounting for 66.92% of the total effect, indicating that the indirect effects are predominant. The spatial spillover effect of innovation on domestic demand should be taken into account.

In terms of control variables, the direct effects of the levels of import/export, urbanization, and government policy are 0.2212, 0.3909, and 0.6623, respectively, which are all significant at the 1% confidence level, indicating that all three also significantly promote domestic demand in the region. However, the direct effect coefficient of the level of external investment is −0.0198, which is significant at the 10% level, suggesting a significant inhibitory effect on domestic demand. The indirect effect coefficient for the level of external investment is 0.0546, which is significant at the 1% level, indicating significant spatial externality on the domestic demand of neighboring areas. The indirect effect coefficients for the levels of import/export and government policy are 0.1399 and −0.1033, respectively, with no significant indirect effects. The total effect coefficients for the levels of import/export and government policy are 0.3611 and 0.5590, respectively, with both predominantly contributed by direct effects, where the level of import/export accounts for 61.26% of the total effect, indicating that the direct impact of import/export levels and government policy on domestic demand should be considered. The total effect coefficient for urbanization level is −0.4545, with the indirect effects being predominant, suggesting that the inhibitory effect of urbanization level on the spatial spillover of domestic demand should be considered. The total effect coefficient for the level of external investment is 0.0349, but it not significant, indicating that the direct effects are predominant (Figure 2).

5.4. Mechanism Testing

To further illustrate the promotion mechanism of innovation on domestic demand, this paper refines the concept of domestic demand into consumption demand and investment demand and examines the promotion effect of innovation on consumption demand and investment demand, respectively.

The results of the mechanism test are presented in

Table 6. Spatial Durbin model mechanism test results.

Variables	W1	W2	W3	W1	W2	W3
	SDM	SDM	SDM	SDM	SDM	SDM
	Cons	Cons	Cons	Invest	Invest	Invest
Inno	0.1441 *** (0.0281)	0.1653 *** (0.0309)	0.0674 ** (0.0315)	−0.0670 (0.0610)	0.0150 (0.0629)	−0.1265 * (0.0651)
Inandout	0.13905 *** (0.0523)	0.1480 *** (0.0544)	0.2257 *** (0.0586)	−0.1583 (0.1123)	−0.1881 * (0.1077)	−0.1778 (0.1223)
Fdi	0.0128 (0.0089)	0.0228 ** (0.0097)	0.0161 (0.0102)	0.1338 *** (0.0191)	0.1407 *** (0.0190)	0.0916 *** (0.0211)
Urban	−0.6636 *** (0.0591)	−0.7631 *** (0.0613)	−0.6577 *** (0.0678)	1.2157 *** (0.1268)	1.2986 *** (0.1211)	0.6458 *** (0.1427)
Gov	0.0154 (0.0388)	−0.0017 (0.0444)	−0.0295 (0.0113)	0.5302 *** (0.0829)	0.5106 *** (0.0874)	0.3056 *** (0.0954)
W×nno	0.0009 (0.0568)	−0.1905 (0.1824)	0.3865 *** (0.0973)	0.7334 *** (0.1130)	2.4150 *** (0.3598)	−0.0376 (0.2033)
W×Inandout	−0.0577 (0.0984)	0.1970 (0.2990)	0.0680 (0.1550)	−0.6354 *** (0.2112)	−1.3379 ** (0.5913)	−0.2310 (0.3223)
W×Fdi	0.0345 ** (0.0177)	0.2591 *** (0.0616)	0.0368 (0.0232)	−0.0371 (0.0384)	−0.0517 (0.1208)	0.1186 ** (0.0512)
W×Urban	0.1641 ** (0.0765)	0.3549 (0.3575)	−0.4432 ** (0.2134)	−0.7182 *** (0.1644)	−0.4316 (0.6860)	−0.3242 (0.4619)
W×gov	−0.3964 *** (0.0872)	−0.8910 ** (0.3526)	−0.3665 ** (0.1712)	0.4044 ** (0.1864)	1.0570 (0.7304)	−0.5452 (0.3628)
ρ	0.4519 *** (0.0800)	0.3334 ** (0.1663)	0.2833 *** (0.0997)	−0.0320 (0.0817)	−0.6104 ** (0.2616)	−0.5151 *** (0.1197)
σ2	2.4034 *** (0.1982)	2.6842 *** (0.2175)	2.9521 *** (0.2416)	10.9567 *** 0.8813	10.4831 *** (0.8451)	12.8695 *** (1.0553)
Fixed/Random	Fixed	Fixed	Fixed	Fixed	Fixed	Fixed
OBS	310	310	310	310	310	310
R2	0.5468	0.4299	0.6920	0.7398	0.6621	0.7615

Note: *, **, and *** indicate that the variables are significant at the 10%, 5%, and 1% levels, respectively, with standard errors in parentheses, W1 is the adjacency matrix, W2 is the geographic distance matrix, and W3 is the economic distance matrix.

.

The results of the economic distance matrix regression are analyzed here, since the level of model fit is better under the economic distance matrix. The regression results with consumption demand as the explanatory variable show that the enhancement of innovation ability significantly promotes the level of consumption demand in the region, and from the perspective of spatial spillover effect, the spatial lag coefficient ρ passes the test of the 1% significance level and is positive, which indicates that there is significant positive spatial externality in inter-regional consumption demand, and there is a significant spatial spillover effect of innovation on consumption demand. The reason for this is that when innovative behavior in the region produces innovative results, on the one hand, it will provide high-quality products to the local market and stimulate the consumption demand of the region’s residents. On the other hand, the benefits brought by the output of innovation results will raise the income level of the residents and increase the consumption enthusiasm of the residents in the region. This leads to the expansion of consumer demand in the region, both in terms of product supply and market demand. The spatial externalities of innovation on consumer demand come mainly from the cross-regional sales of innovations; when innovations are sold in markets outside the region through transportation networks, they push to unleash the potential of domestic demand in neighboring regions, which, in turn, expands domestic demand in neighboring regions. With investment demand as the explanatory variable, innovation has a dampening effect on regional investment demand, and the spatial coefficient of innovation turns out to be insignificant. This may be because the innovation process requires a large amount of capital and manpower, among other innovative factors. Therefore, when the innovation capacity of the local area improves, it will obtain high-quality innovation resources from various aspects through China’s transportation infrastructure, rather than just obtaining single innovation resources from the local area. Therefore, innovation will, to some extent, suppress local investment demand while also generating negative spatial externalities for investment behavior in surrounding areas.

In summary, innovation not only significantly promotes local consumption demand but also generates significant spatial externalities for consumption demand in surrounding areas. As for investment demand, innovation suppresses local investment demand, failing to meet significant requirements for investment demand in surrounding areas.

6. Conclusions and Recommendations

6.1. Conclusions of the Study

The construction of a new development pattern of a double cycle, with “the domestic macrocycle as the main focus and the domestic and international double cycle promoting each other”, provides a suitable narrative background for the study of the influence mechanism and spatial externality between regional innovation and domestic demand. Based on absorbing the existing theoretical results on innovation and domestic demand, this paper elaborates on the interactive mechanism of innovation to promote the spatial externality of domestic demand from a macro perspective and constructs a theoretical model. At the same time, using the spatial panel data of 31 provinces (cities and districts) in China from 2011 to 2020, the spatial Durbin model is used to carry out empirical tests and decompose its spatial spillover effects. Finally, the mechanism of innovation to promote the expansion of domestic demand is tested.

The main conclusions are as follows: (1) theoretical research in new economic geography indicates that innovation plays a dual role in boosting China’s domestic demand. On the supply side, innovation drives internal circulation by introducing new products to the market. On the demand side, it stimulates domestic demand by increasing population income. Furthermore, innovation and domestic demand generate spatial externalities in neighboring regions through spatial spillovers; (2) the empirical results demonstrate that innovative behavior significantly stimulates domestic demand expansion within the region under all three spatial weight matrices. While the adjacency matrix and geographic distance matrix exhibit a certain inhibitory effect of innovation on neighboring regions’ domestic demand, the economic distance matrix shows a significant promotional effect of innovation on neighboring regions’ domestic demand; (3) analyzing the spatial spillover effect of innovation on domestic demand, the results reveal that the indirect effect, termed spatial externality, contributes 66.92% to the total effect. This suggests that spatial spillover plays a significant role in promoting domestic demand and warrants further emphasis in research; (4) through the refinement of domestic demand into consumption demand and investment demand and research on the mechanism of innovation to promote domestic demand, the results show that the promotion effect of innovation on domestic demand is mainly realized through consumption demand, and innovation has a significant promotion effect on the consumption demand of the region and the surrounding areas, while innovation has a certain inhibitory effect on the investment demand of the region, and the effect on the investment demand of the surrounding areas is not significant. Overall, innovation promotes the spatial externality of our domestic demand, which in turn has a positive effect on our domestic demand.

6.2. Policy Recommendations

• Adhering to the innovation-driven development strategy: Continuously shaping new momentum for development by adhering to the central position of innovation in the overall situation of China’s modernization, we will enhance people’s well-being through innovation, upgrade the industrial chain, ensure the stability of China’s supply chain, provide the sustained impetus for China’s economic development, promote the expansion of domestic demand through the enhancement of the capacity for innovation, and provide favorable support for the construction of a new development pattern of the “double cycle”.
• Continuing to actively build a new development pattern that “focuses on the domestic cycle”: Grasping the dominant position of the domestic cycle in the double cycle and insisting on the expansion of domestic demand as the basis for development. Governments at all levels and in all parts of the world, in the process of realizing the expansion of domestic demand, will not only have a positive impact on the region but also have a positive effect on neighboring regions. Specifically, governments should face up to the differences in development between regions and implement innovative strategies to promote the expansion of domestic demand in the region according to local conditions, such as for the eastern region, because the level of innovation and domestic demand have a great advantage over inland regions, so it is necessary to pay attention to the spatial externality that arises from the process of expanding the region’s domestic demand to promote the realization of the expansion of the inland region’s domestic demand through spatial spillover effects.
• Emphasize the centrality of consumer demand to domestic demand: Although domestic demand consists of consumption demand and investment demand, in practice, the expansion of domestic demand is mainly realized through consumption demand, so in the process of building the domestic cycle in the future, we should actively grasp the role of consumption demand for the domestic cycle, taking advantage of the fact that our country has the world’s largest middle-income group, and focus on unleashing China’s huge potential for domestic demand and realizing a two-way drive of the demand side and the supply side through consumption demand to ultimately help China’s domestic demand expansion.

7. Limitations of the Study and Future Research Directions

This study investigates the impact of regional innovation on China’s domestic demand, facing limitations due to data availability only up to 2020. This restricts our ability to analyze recent global economic shifts and their influences comprehensively. While employing provincial panel data and the spatial Durbin model offers a substantial overview, it may not entirely encapsulate the intricate dynamics and externalities associated with regional innovation. Future research should aim to broaden the temporal and spatial scope of analysis, integrate more varied datasets, and delve into the implications of digitalization and technological progress on innovation ecosystems. Furthermore, conducting comparative analyses across diverse economic and policy landscapes would greatly enhance our understanding of the universal relevance and consequences of our findings, facilitating a deeper and more detailed perspective on the role of regional innovation in economic development globally.