The Growth of Service Sectors, Institutional Environment and Quality Development in China’s Manufacturing

Abstract

The path of exploring the high-quality development of the manufacturing industry has become one of the focus issues in China. In terms of experience, the impact of service sector growth on the quality of China’s manufacturing industry is unclear. Therefore, institutional environment is incorporated in the research frame, and the Sys-GMM is used to estimate the relationship between them based on China’s provincial panel data from 2010 to 2020. The main results showed that: (1) Both the service sector growth and institutional environment can improve the quality of manufacturing development, and they have a significant interaction. (2) The impact of service sector growth on the quality of manufacturing industry decreases in the eastern, central and western regions of China. The expansion of foreign direct investment in services cannot significantly improve the quality development in China’s manufacturing, and there is no obvious interaction between SFDI and the institutional environment. However, the domestically funded service sector is the opposite. The main contribution of this paper is to empirically verify the impact of service sector growth on the quality of manufacturing development in China, and consider the differences in the impact of regional development of service sectors and the heterogeneity of investment subjects. In addition, from the perspective of system development, this article considers the influence of the institutional environment, so as to analyze the former’s impact on the latter in more detail.

1. Introduction

Undoubtedly, the manufacturing industry in China has made amazing achievements since the 1980s. From 2010 to 2020, the value-added of the manufacturing industry has increased from CNY 13 trillion to CNY 26.59 trillion, doubling in ten years. With relatively perfect supply chain and cost advantages, China has also shown strong competitiveness internationally. The export volume of industrial products in China has increased from CNY 13.15 trillion to CNY 17.13 trillion in the last five years, with an average annual growth rate of 6.8%, which is much higher than the average annual growth rate of 2% of world trade. However, the proportion of value-added of China’s manufacturing industry in the whole economy has decreased from 31.86% to 26.35% in the last ten years, showing a continuous downward trend. On the contrary, the proportion of service sectors increased from 44.2% to 54.5%. Compared with the continuous declining share, insufficient innovation investment, poor business environment and low development quality brought by the extensive business model are more serious challenges facing the manufacturing industry in China at this stage. In terms of innovation investment, the level of research and development on manufacturing is not only lower than the overall domestic level, but also has a greater gap compared with developed countries. In 2020, the R&D intensity of industrial enterprises above designated size in China is 1.41%, lower than 2.4% of that of the whole society, and even less than half of that of industrial powers such as Japan and Germany. Meanwhile, according to the statistics in 2018 from the EU on the R&D investment of 2500 industrial enterprises in the world, only Huawei ranked in the top 50 companies among them. From the perspective of energy consumption, although the standard coal consumed per CNY 10,000 of value-added by manufacturing in China decreased from 1.21 tons in 2015 to 1.02 tons in 2020, a decrease of 16%, this figure is still about twice that of the United States. In terms of labor efficiency, the value-added per capita of manufacturing in China has reached CNY 166,800 in 2020, an increase of 33.87% over 2015, but this is still less than 19% compared with the United States, and 30% compared with Germany and Japan. The historical experience of economic development, especially the lesson of “deindustrialization” in western developed countries, shows that even in the intelligent era, manufacturing is one of the most basic driving forces to promote economic and social development. In China, industrialization, especially modernization, is still an unfinished business, which determines that there is a huge potential for the development of manufacturing industry.

In recent years, the cost advantage of China’s manufacturing industry has been weakening compared with that of the United States and some developing countries due to the slowdown of economic growth and the rise of labor costs, and thus its competitiveness is facing severe challenges. Therefore, a key report delivered at the opening session of the 20th National Congress of the Communist Party of China stressed that high-quality development is the top priority of building a modern socialist country in all respects, while the high-quality development of manufacturing is the key and foundation of the high-quality development of the economy. It has become the national development strategy, and is also the threshold that China must cross to become a high-income country. So, how to improve the development quality of manufacturing industry, that is, to explore the path to promote the high-quality development of manufacturing industry, has become one of the focus issues among all walks of life, among which promoting the integration of the manufacturing and service industries is one of the important paths. Although it has been officially established for more than 30 years, China’s market system is not perfect and affects the development of economic society, which is a fact that cannot be ignored. Given such basic characteristics, can the growth of the service sector improve the quality of China’s manufacturing industry? What are the effects of institutional changes? Does the growth of different regions and different types of investment entity service sectors have heterogeneity on the quality of manufacturing development? There are at least no clear answers to these questions. Of course, these are also the core of this paper.

Although there is no consensus on how to achieve high-quality development, a large amount of relevant research has been accumulated. The research related to this paper is mainly carried out along the following two strands of the literature: One is how to assess high-quality development in manufacturing. It is the first problem to be solved in studying the impact of the growth of service sectors on the quality of manufacturing industry development. On the other hand, the path to promote the high-quality development of the manufacturing industry is explored from a different perspective.

There are two basic ways to evaluate the quality of manufacturing development. One way is known as the single indicator method, and the most commonly used indicators are total factor productivity (TFP) [1,2,3]. There is no doubt that it is relatively simple and requires a relatively small amount of data [4]. However, due to the different measurement methods of total factor productivity, the results of TFP are quite different, leading to the lack of robustness of the evaluation results [5]. For example, He and Shen [2] calculated TFP by the Solow residual method, Wang and Xu [6] used ML index and the Dagum Gini coefficient method to measure TFP, and so on. Of course, labor productivity [7], real GDP per capita [8,9], and carbon emission intensity [10] have also been used as single indicators to measure the high-quality development of the manufacturing industry by different researchers. Although Zhao and Gu [11] measure and compare the manufacturing development quality in China and the US from TFP in production processes, value acquisition ability and technical sophistication in export link, this method still belongs to a single indicator system, because they estimate the quality of manufacturing development from each angle. Nie and Jian hold that, with this method, it is often difficult to reflect the meaning and essence of high-quality development of the manufacturing industry [4], and it can also easily cause deviations in the analysis.

The other way is called an indicator system method. In order to make up for the deficiencies of the single index method, a multi-dimensional composite evaluation system has been constructed [12,13]. For example, Shi and Zhang [14] constructed a high-quality development index system of the Chinese urban economy from three dimensions of development fundamentals, social achievements and ecological achievements. However, the dispute over it mainly lies in that the indicator selection still pays more attention to the quality of economic growth than the quality of development, because the concept of “growth” is much smaller than that of “development” [15]. In fact, this is the same as the single indicator method, which tends to regard the quality of economic growth as economic development. The indicator selection should focus on how to reflect the development quality of manufacturing industry, especially the new development concept [5]. The new development concept proposed by President Xi Jinping, namely, the development concept of innovation, coordination, greenness, openness and sharing, is regarded as the embodiment of high-quality development economic and social practice. Therefore, the high-quality development indicator system based on this concept has been established. Zhan and Liang [16] believe that the characteristics of high-quality development of regional manufacturing can be preliminarily conceptualized into five dimensions, including “innovation, green, openness, sharing and efficiency”. Considering that the external risks facing the development of China’s manufacturing industry are increasing, Qu et al. incorporated risk control into the measurement system on the basis of the above research [17]. However, Guo, et al. [5] argued that innovation, coordination, green, open and sharing are normative value judgments, and their connotation and extension are not clearly defined, thus making them unworkable. In addition, they measured the quality of regional economic development from four dimensions: engine, structure, mode and achievement. Most scholars have ignored the basic principle and fact that only maintaining a moderate amount of growth can bring about qualitative improvement, which is particularly important in China with an economic downturn.

How to achieve the high-quality development of the manufacturing industry is another important issue that needs to be solved. This paper is related to at least two strands of the literature: institutional environment and the role of services. From a theoretical perspective, Kisha and Xiaoxiao [18] pointed out that the optimized business environment can stimulate enterprises to carry out technological innovation and achieve the upgrading of manufacturing and high-quality development, while Chuanlin [19] verified the above conclusions empirically. In fact, the basic foundation of the high-quality development system is that the market determines the resource allocation, including the market-oriented allocation of factors, the improvement of the property rights system with the principles of competitive neutrality, the system of market decision resource allocation, and the establishment of an open access order [20]. Fu and Zhang found that under the inspiration of the system, innovation reforms have significantly reduced institutional transaction costs, promoted corporate innovation, and coordinated the two to promote the high-quality development of enterprises [21]. Zhang and Wang also examine the impact of economic institutional changes and industrial structure evolution on the high-quality development of China’s economy [22]. They insist that the economic system indirectly promotes the high-quality development of China’s economy by correcting resource mismatches and promoting industrial structure optimization. However, Xiaoliang advocates that constraints on the institutional level and insufficient support of production factors are important obstacles to the high-quality development of the manufacturing industry in China [23]. Most of this research has investigated the influence of institutional environment on the development quality of the manufacturing industry from the perspective of technological innovation.

The role of service sectors on manufacturing industry development mainly focuses on the mechanism and path of the effect. Zhiyuan [24] indicates that the deep integration of the manufacturing industry and modern service sectors should be the basic path to promote the high-quality development of the manufacturing industry. Xianhai and Zhujun [25] elaborate the basic mechanism of producer services promoting the high-quality development of the manufacturing industry from four dimensions: cost saving effect, preference enhancement effect, industry competition effect and financing constraint effect, providing a theoretical basis for the study of their relationship. Fernandes [26] points out that the inflow of FDI in services (SFDI) helps to improve the total factor productivity of manufacturing industries in host countries. The production efficiency of enterprises will increase with the reduction and improvement of the input service cost [27]. Of course, the improvement effect is related to the nature and intensity of the input service [28]. Xiaojun and Dexue [29] proved that SFDI in China improves the development of the manufacturing industry by promoting technological progress rather than technological efficiency, and it can only be seen after enterprises have passed a certain technical threshold and gone through a certain adaptation period. Guo and Yuan [30] claims that the relationship between producer services agglomeration and manufacturing upgrade is not a simple linear relationship; instead it shows a U-shaped relationship. In addition, the high-quality development of producer services has a threshold effect on the upgrading of the manufacturing industry [31]. Although the upgrading of the manufacturing industry is not the only reason for the high-quality development, it shows there is certain relationship between them. However, not all research supports the above views. Guangmin [32] argues that China’s high-tech service cannot effectively improve manufacturing efficiency. The reason may be that the effective supply capacity of the service sectors is insufficient, and its efficiency is not high.

The research mentioned above mainly focuses on how to effectively measure the high-quality development of the manufacturing industry and explore the path to promote the high-quality development of the manufacturing industry. In some cases, the evaluation index system is constructed by using the new development concept. The impact of business environment optimization on the quality of manufacturing development is also analyzed from the perspective of technological innovation. The mechanism and experience of manufacturing production efficiency or structure upgrading are also sought from the perspective of service industry development. However, the evaluation index system puts too much emphasis on qualitative dimensions at the expense of other aspects. This may cause the evaluation results to not reflect the quality development of the manufacturing industry effectively. At the same time, few scholars directly analyze the impact of the growth of service industry on the development quality of the manufacturing industry, especially lacking corresponding evidence from experience, and do not consider how the role between institutional environment and service industry affects the high-quality development of the manufacturing industry. In addition, in the face of huge regional development differences and unique manufacturing structures in China, do the growth of service industry and institutional environment have the same impact on the quality of manufacturing development? These issues are rarely addressed. Therefore, to close these research gaps, this study explored the following research questions:

(1) How to establish an effective evaluation index system for the development quality of China’s manufacturing industry?

(2) Can the growth of service sectors effectively improve the development quality of China’s manufacturing industry? Suppose the answer is yes, how effective is it? What is the role of the institutional environment?

(3) Is the impact of service sector growth and institutional environment on the development quality of China’s manufacturing industry heterogeneous?

To answer these questions, institutional environment, an important feature of China’s transition period, is incorporated into the research framework. The Sys-GMM method is employed to evaluate the impact of service sector growth and institutional environment on the development quality of the manufacturing industry by using the provincial-level panel data from 2010–2019.

Based on observations from the existing literature, the marginal contributions of this paper are mainly as follows: (1) Based on the new development concept, a more scientific and reasonable manufacturing development quality evaluation index system is established under the premise of the unity of quality development and quantity growth. (2) Taking full account of China’s basic national conditions of institutional transformation, and incorporating institutional variables into the analysis framework, this paper not only theoretically and empirically, but also from a systematic perspective, more accurately analyzes the impact of service industry growth on the quality of manufacturing development. (3) Considering the regional differences of China’s economic development and the different investment subjects of the service sector, the heterogeneity of the growth of service sectors on the development quality of China’s manufacturing industry is confirmed.

The remainder of this paper is organized as follows. Section 2 is theoretical analysis and some research hypotheses have been proposed. Section 2 also includes research design, including model establishment, variables and data description. Our empirical study and results analysis are presented in Section 4. Robustness test and heterogeneity analysis are completed in Section 5. Ultimately, Section 6 concludes the paper with conclusion remarks, policy implications and future research directions.

2. Theoretical Analysis and Research Assumptions

It is widely recognized that there are extensive, complex and close technical and economic links between industries in a complex economic system. With the deepening of social division of labor, the production system of enterprises is becoming more and more complex, and the production chain is getting longer and longer, which will promote the improvement of production efficiency. The improvement of the detour degree of manufacturing production requires a large number of intermediate inputs as the premise, and service sector growth provides it with the possibility. In modern manufacturing production, the proportion of intermediate input represented by technical services, business services and financial services has increased significantly. In fact, as the carrier of human capital and knowledge capital, they are embedded in the production process, which can obtain more output and improve the value of products [33]. In a developed financial market system, it is easier for the capital-intensive manufacturing industry to enhance its competitiveness and expand its share in the international market [34]. The rapid growth of the service sectors promotes the rapid expansion of scale and the use of more advanced technological achievements. While improving the quality of service, the service cost is constantly reduced, which brings about spillovers to downstream users of the services and the improvement of the benefits of the manufacturing industry. Francois and Woerz [35] agree that the growth of business service outsourcing will significantly improve the productivity and competitiveness of technology-intensive industries in OECD countries. In addition, technological progress and the improvement of production efficiency mean that the manufacturing industry develops high-quality development.

From a micro perspective, it is assumed that a service supplier and a manufacturer belong to the service sectors and the manufacturing industry, respectively, and that there is an input–output relationship between them, that is, the former provides various services for the latter to meet the needs of the production process. In other words, they are supply chain partners. In order to cope with the uncertainty of the market, both of them need to manage the supply chain, and strive to reduce costs and achieve business objectives while meeting customer needs. Due to the differences in industries and development stages of different companies, and the lack of a unified centralized center to make effective coordination and arrangements, it is difficult for its operating goals to be completely consistent.

With the upgrading of consumption and the increase in demand, downstream manufacturers continue to upgrade their technological level and expand their production capacity, thus requiring more and higher-quality services. If the service supplier fails to provide relevant services on time, the manufacturer will not be able to complete the established production of goods, meet the needs of downstream customers or consumers in a timely and effective manner, meaning that it cannot expand production capacity, increase profits and provide sufficient support for subsequent innovation investment. Supply shortages will bring market punishment costs for manufacturers. If the cost per unit shortage is high enough, it will make the manufacturer more inclined to maintain production instead of being punished [36], which will greatly reduce innovation activities, resulting in slow technological progress and decline of enterprise competitiveness. If this phenomenon spreads to the industry level, it will curb the improvement of the quality of the manufacturing industry. In order to effectively solve this problem, there are generally two strategies: First, manufacturers can share market information with service suppliers, design a centralized plan coordination mechanism to promote the increase in service supply, avoid supply chain disturbance and ensure the interests of all parties involved. In other words, a decentralized decision-making mechanism of the supply chain will be established [37]. Second, the manufacturer will transfer the cost of punishment to the upstream of the supply chain, and the service supplier will face the danger of being removed from the supply chain system. Manufacturers will need to choose new suppliers, but establishing a new supply chain system may face new transaction costs such as search and negotiation costs. When the cost is too high, the former will be the optimal strategy. Of course, if the supply chain disturbance comes from the increase in service supply of service providers, it will bring about cost reduction and more high-quality services for manufacturers. The reduction of the relative price of production factors will also encourage manufacturers to invest in more services and improve the competitiveness of manufacturers. In short, regardless of whether the service supply represents passive or active expansion, it promotes the improvement of the quality of the manufacturer’s development, and then improves the situation of the whole industry. Therefore, we put forward the following hypothesis:

Hypothesis 1 (H1).

Growth in the service sectors can improve the quality of manufacturing development.

Institutional economists agree that the performance of economic and social development ultimately depends on the quality of the institutions, because the incentives provided by the institutions control the direction and effectiveness of the efforts of micro entities, and then affects the quality of manufacturing development. For more than 40 years, building and continuously improving the socialist market economy system has resulted in the most important and far-reaching institutional environment in China. At the same time, the efficient execution of Chinese government also provides the necessary guarantee conditions for the institutions to play their role [38].

With the introduction of the market mechanism, entrepreneurship has been fully released, and the innovation vitality of the micro entities has also been stimulated. The enterprises continue to increase their investment and expand their scale rapidly. In order to cope with more fierce competition, enterprises put more resources into innovation activities, and their innovation ability is constantly enhanced. The development of the manufacturing industry has begun to rely more on scientific and technological progress rather than purely quantitative expansion, and the competitiveness of enterprises continues to improve, the industrial structure tends to be reasonable, the energy consumption is constantly reduced, green and sustainable development is realized, and the development quality can be improved. Promulgation of a series of national policies, which encourage innovation and the construction of social culture and infrastructure, not only further strengthens the innovation will of micro entities, but also provides good external conditions for innovation activities. The transformation of government functions and the improvement of efficiency, the continuous optimization of the business environment, the gradual improvement of the rule of law system and the strengthening of the environmental protection system have brought a more stable and predictable development environment for micro entities and reduced various institutional transaction costs, achieved sustainable development and improved the quality of development [39]. In addition, with the development and improvement of the factor market, resources continue to flow to industries and fields with higher marginal income. The efficiency of factor allocation is improved, which promotes the optimization and adjustment of the manufacturing industry structure and promotes the manufacturing industry to climb to a higher value chain. In other words, institutional reform is conducive to improving the quality of manufacturing development.

On the other hand, the institutional quality of a country or region is significantly and positively correlated with the proportion of its service sectors [40]. Due to the intangibility, the difference and individuation of demand, and the immediacy of production and consumption in service sectors, the development of service sectors is becoming more sensitive to various institutional measures, including legal institutions and transaction institutions. The improvement of the market economy system in China has promoted the improvement of development policies on service sectors, reduced the uncertainty and cost of transactions, and provided a fertile “soil” for the development of the service sectors. It is true that the development of the service sectors has put forward higher requirements for establishing institutions, provided impetus for the further improvement of the institutions and accelerated the pace of institution change. As a result, the second hypothesis is as follows:

Hypothesis 2 (H2).

The optimization of the institutional environment is conducive to the improvement of the quality of the development of the manufacturing industry, and the interaction with the service industry, magnifying each other’s influence.

Generally speaking, the factor endowment structure of an economy determines its industrial structure. Labor and natural resources are more abundant, and labor-intensive or resource-intensive types have become its leading industries for some economies in primary development. Knowledge and capital have become the core elements of developed economies, determining that technology-intensive or capital-intensive types have become advantageous industries [41]. Since the reform and opening-up, China has implemented the unbalanced growth policy in favor of the coastal areas [42]. Relying on the preferential policy of taking the lead in the development of the east of the country, and the advantageous geographical conditions and good development foundation, the eastern regions have vigorously developed the manufacturing industry of fine processing, deep processing and high value-added, with relatively advanced production technology and gradually improved cultural, educational, scientific and technological levels [43]. In addition, the typical regional differences in technical efficiency of equipment manufacturing industry has been formed, which are sequentially decreasing in the order of the Pearl River Delta, the Bohai Rim, the Yangtze River Delta, the midwest and the northeast [44]. More importantly, industrial evolution and development depend on initial conditions. Once the development advantage is formed, the increasing income mechanism will further strengthen its market position [45], thereby forming path dependence and path locking. The industry linkage effects between service sectors and manufacturing industry determine that the impact of the former on the latter may vary with regional economic development. Based on this, we proposed the following third hypothesis:

Hypothesis 3 (H3).

The influence of the growth of service sectors on the development quality of manufacturing industry is heterogeneous.

3. Research Design

3.1. Establishing the Model

In order to test the specific effects of the growth of service sectors and institutional environment on the development quality of manufacturing industry, this paper constructs two main models. The first model is the benchmark model, which mainly examines the impact of service sectors growth on the quality of manufacturing development, and constructs the following regression model:

$$m{i}_t=c+{\beta }^{\prime }\times s{d}_{i,t}++{\varphi }^{\prime }\times X_{i,t}+{\mu }_{i,t}$$

(1)

where $mi$ denotes the development quality index of the manufacturing industry, which measures its development quality; $sd$ refers to the growth of the service sectors; $X$ is a number of control variables; $C$ denotes the constant term; ${\beta }^{\prime }, {\varphi }^{\prime }$ are coefficients; $\mu$ denotes the random disturbance term; and $t$ represents the period.

As mentioned above, service sector growth is also influenced and regulated by the institutional environment. In order to test this role and the effect of institutional environment, Equation (2) is constructed. Considering the huge inertia of the development mode of the manufacturing industry, the lag term of the development quality index of the manufacturing industry is introduced into the model, and the following regulation effect model is constructed:

$$m{i}_{i,t}=c_{i,t}+\alpha \times m{i}_{i,t-j}+\beta \times s{d}_{i,t}+\lambda \times in{s}_{i,t}+\pi \times s{d}_{i,t}\times in{s}_{i,t}+\varphi \times X_{i,t}+{\mu }_{i,t}$$

(2)

In Equation (2), ${mi}_{i,t-j}$ is lag term of $mi$ in the order of $j$, which is used as the explanatory variable of the model. ${\mathit{ins}}_{i,t}$ represents the environmental conditions of institutions of region $i$ in the $t$ period; $s{d}_{i,t}\times in{s}_{i,t}$ is the interaction item of service sectors growth and institutional environments; $\alpha ,\beta ,\lambda ,\pi ,\varphi$ are coefficients, $u_{i,t}$ is a random error term.

The degree of capital deepening will affect the direction, level and quality of manufacturing development [46]. Knowledge flow can reduce the cost of enterprises to acquire new technologies, accelerate the spillover effect of knowledge and help enterprises improve their innovation ability and development quality [47]. Generally speaking, the degree of openness determines the degree of competition in the manufacturing industry and affects the innovation willingness and motivation of micro entities [48]. In addition, opening up and cooperation can bring spillover effects to the development of the domestic manufacturing industry to a certain extent, which can promote the improvement of the quality of manufacturing development. The development of infrastructure is an indispensable external condition for the development of the manufacturing industry, which includes not only traditional infrastructure such as road and transportation [49], but also digital infrastructure that plays an important role in driving the high-quality development of the manufacturing industry [50,51]. Therefore, this paper selects the degree of capital deepening, knowledge flow, openness and infrastructure development as control variables.

3.2. Variable Selection

3.2.1. Development Quality Index in Manufacturing

The differences in research perspectives and understanding of connotation lead to differences in the measurement of the development quality in manufacturing among researchers. The indicator system evaluation method measures the development quality in the manufacturing industry more comprehensively from a multi-dimensional perspective [52]. Therefore, this idea is followed to build a development quality evaluation system of manufacturing industry in China. There is no doubt that high-quality development emphasizes the quality dimension in the development process. However, this does not mean that the growth of quantity can be ignored. On the contrary, the development of the manufacturing industry cannot be considered as high-quality without a reasonable change in quantity. In other words, the high-quality development of the manufacturing industry is the unity of quality and quantity, which complement each other. Mlachila [53] believes that the greater growth rate and longer social-friendly growth is a high-quality type of growth for developing countries. However, most scholars do not include the growth of manufacturing quantity into the development quality evaluation system according to the existing research results. Therefore, these factors are incorporated into the evaluation system in this paper.

The development of the manufacturing industry is a dynamic process. Therefore, evaluating the quality of manufacturing development should not only reflect the existing development achievements, but also pay attention to the sustainability of future development. In terms of quantity, high-quality development means that the manufacturing industry maintains a reasonable development scale. In terms of quality, it mainly lies in improving its performance, innovation, optimizing its internal structure and realizing the sustainability of development. On the basis of existing studies, this article has established an indicator system for high-quality development of the manufacturing industry from five dimensions, including development scale [54], economic performance [55], structural optimization [54], innovation capacity building [56], and green development [14,56], as shown in

Table 1. Development Quality Evaluation Indicator System of Manufacturing Industry.

First-Level Indicators	Second-Level Indicators	Third-Level Indicators	Basic Indicators
Quality indicators of manufacturing development	Development scale [54]	Production capacity [14]	Value-added of manufacturing industry per capita [14]
		Market share [57]	The proportion of value-added of manufacturing industry in GDP [57]
		Employment scale	The proportion of annual average employees in manufacturing industry to number of employed persons
	Economic performance [55]	Labor productivity [4,10]	Per capita gross output value of employed persons in manufacturing industry [4,10]
		Beneficial result [58]	Profits rate of industrial enterprises above designated size [58]
			Return on assets of industrial enterprises above designated size [58]
	Structural optimization [54]	Enterprise quantity Structure	The share of number of high-tech manufacturing enterprises in industrial enterprises
		Income structure [55]	The share of revenue of high-tech enterprises in industrial enterprises
			The share of profits of high-tech enterprises in industrial enterprises [55]
		Employment structure	The share of annual average employees of high-tech enterprises in industrial enterprises
	Innovation ability [56]	R&D investment [58]	The share of R&D expenditure in industrial enterprises within the whole society [58]
		Human capital investment [58]	The share of R&D expenditure in industrial enterprises within the whole society [58]
		Investment in new product development [57]	The share of expenditures for new products of industrial enterprises within revenue [57]
		Patent application [55]	Number of invention patent applications of industrial enterprises per capita [55]
	Green development [14,56]	Exhaust emission [59]	The nitrogen oxide emissions per CNY 100 million of GDP [59]
		Waste discharge [59]	Common industrial solid wastes generated per CNY 100 million of manufacturing GDP [59]
		Energy consumption [58]	Energy consumption per CNY 100 million of manufacturing GDP [58]
		Environmental remediation	Investment completed in the treatment of industrial pollution per CNY 100 million of manufacturing GDP

. The indicator system includes 5 s-level indicators, 16 third-level indicators and 18 basic indicators.

Except for the development scale, the rest measure the development of manufacturing quality from different perspectives in the second-level indicators. The measurement indicators of development scale generally include production capacity [14], market share [57] and employment scale. Considering the difference of economic scale in different regions, per capita value-added of manufacturing industry [14], the proportion of value-added of manufacturing industry in GDP [57] and the share of employment in industrial enterprises in the whole society are used as specific measurement indicators. The measurement indicators of economic benefit mainly include labor productivity [4,10] and beneficial result [58], in which the rate of return is measured by two second-level indicators, namely, the profit rate and the return on assets of industrial enterprises above a designated size [58]. Structural optimization is mainly measured by income structure [55], but the changes in the number of high-tech enterprises and employment structure also reflect the upgrading of manufacturing industry. Therefore, we incorporate them into the measure of structural optimization. The amount of R&D investment and R&D personnel determines the innovation ability of an enterprise in the future [58], and is also an important guarantee for high-quality development. The investment in new products [57] and patent application [55] mainly reflects the existing innovative ability and achievements. Therefore, this paper measures the innovation ability of the manufacturing industry from these aspects. Green development is one of the important features of the high-quality development of the manufacturing industry, which is generally measured by exhaust emission, waste discharge [59] and energy consumption [58]. The governance of environmental pollution is one of the important ways to maintain the sustainable development of the economy and society. Therefore, we take environmental remediation as one of the measurement indicators of green development. Starting from the authority and availability of data, nitrogen oxide emissions, common industrial solid wastes generated and energy consumption and investment completed in the treatment of industrial pollution per CNY 100 million of manufacturing GDP are selected as basic indicators in this paper.

Among the basic indicators, nitrogen oxide emissions, common industrial solid wastes generated and energy consumption per CNY 100 million of manufacturing GDP are inverse indicators, while the rest are positive indicators. The original data of all the basic indicators are from different kinds of statistical yearbooks or calculated from them. Due to the different meanings and dimensions of various indicators, there is a lack of comparability. Therefore, all indicators need to be dimensionless. In addition, this paper adopts the method of deviation standardization. The reverse index is processed in the forward direction using the transformation threshold method. The linear weighting method is used to measure the development quality index of China’s manufacturing industry in this paper. Following Xiaolu’s research method, the arithmetic average method is applied to calculate indicators at all levels [60]. That is, each index is given the same weight, expressed in $mi$.

3.2.2. Growth of Service Sectors

In different studies, scholars have slightly different indicators for measuring service sector growth. Rajan and Zingales use the share of the sum of private bank credit and stock market capitalization in GDP as a substitute variable for service sector development [61]. Liu and Matto used the value-added per capita of financial and business services to measure the development of the U.S. service sectors [62]. Although financial and business services are the core of modern service sectors which occupy a dominant position especially for developed economies, the traditional services represented by wholesale, retail and transportation are the intermediate services with the largest investment in manufacturing industry in China. If the fact is ignored, it is difficult to effectively reflect the impact of service sector growth on the quality of manufacturing development in China. Meanwhile, considering the differences in economic scale, the value-added per capita of service sectors is used to measure its growth in this paper, and is standardized and expressed as $sd$.

3.2.3. Institutional Environment

Due to the universality and complexity of the institution itself, and the diversity of research purposes, scholars are far from reaching consensus on the measurement of institutions. Different understandings of the extension of the institution lead to different measurement methods. The commonly used alternative variables include the business environment, the strength of intellectual property protection, the status of law and order, and the development of non-state-owned economy. These indicators reflect the status of a country’s institution construction from a certain perspective. In order to more fully reflect the institutional environment of an economy, Lijun [63] has constructed an indicator of China’s economic freedom from four dimensions of property rights, commerce, investment and financial freedom to measure institutional environment. On this basis, Fulin [64] includes the government efficiency into the measurement range. Xiaolu [60] regards the relationship between the government and the market, the development of non-state-owned economy, the development of product and factor markets and the status of legal institutional construction as the main contents of institutional construction in China, and compiles a continuous marketization index. In view of the content involved in the indicator, which better reflects the connotation and composition of China’s market economy system and the direction of future changes, this paper uses Wei [65] practice to measure the degree of institutional environment with the marketization index of each region, which is expressed as $ins$. Up to now, the marketization index only provides panel data of each region from 2008 to 2019; the linear extrapolation method was used to estimate the marketization index of each region in 2020.

3.2.4. Other Variables

In this paper, capital deepening is defined as per capita fixed assets of industrial enterprises, expressed by $\mathit{capd}$, and the ratio of total industrial fixed assets of the region in the period to industrial employees is used as a substitute variable [46]. Industrial fixed assets are measured using the commonly used perpetual inventory method. Haojie’s research and methods [66] are used for the initial capital estimation, that is, the initial capital stock of each region is calculated by dividing the amount of fixed assets formed in 1980 (2010 as the base year) by the weighted depreciation rate. The calculation method of weighted depreciation rate is: the sum of depreciation rate and geometric average growth rate of fixed asset investment over five consecutive years, with the depreciation rate of 10%, which assumes that the depreciation rates are equal in all regions.

The flow of knowledge among different industries or enterprises is often reflected in the transaction of the technology market. Using Dechao’s research for reference [47], the ratio of the turnover of the technology market to the total industrial output value in each region is used as a substitute variable for the flow of knowledge, represented as $\mathit{knf}$. The greater or more frequent the knowledge transfer, the more conducive it is to the improvement of enterprise technology level, thus promoting the improvement of the development quality in the manufacturing industry. In addition, this paper defines the degree of opening to the outside world as the proportion of total imports and exports of goods in GDP of a region [67], denoted as $\mathit{od}$. The development of the manufacturing industry cannot be separated from the improvement of infrastructure. The quality of infrastructure in a country or region not only affects the speed and quality of manufacturing development, but also is an indispensable external condition for the transformation and upgrading of the manufacturing industry. Considering the availability of data, this paper uses the infrastructure development indicator to measure the infrastructure development status of a region, denoted as inf. The specific calculation process is as follows: the railway operating mileage, highway mileage, number of internet broadband access users and mobile telephone switch capacity in each region are indexed, and then the weighted average is performed.

3.3. Data Description

This paper selects the panel data of 30 provinces, autonomous regions and municipalities in China from 2010 to 2020 for research. Considering the particularity of Tibet’s industrial structure, it is not included in the sample, nor does the paper consider Hong Kong, Macao and Taiwan. Due to the adjustment of administrative divisions in China, in order to maintain the consistency and consistency of the data, when calculating the fixed assets of various regions, the data of Chongqing before 1997 are not included in the statistical scope of Sichuan. Similarly, the data of Hainan before 1988 are excluded from the relevant data of Guangdong Province. All data are obtained from China Industrial Statistics Yearbook, China Statistics Yearbook, China Fixed Assets Investment Statistics Yearbook, the Tertiary Industry Statistics Yearbook, China Provincial Marketization Index Report and the statistical yearbook of provinces, autonomous regions and municipalities. The total import and export, fixed capital investment and GDP are deflated by the import and export commodity price deflator and GDP deflator, respectively, and the value-added in service sectors is deflated by the tertiary industry value-added deflator, with 2010 as the base period.

4. Empirical Study

4.1. Endogenous Problems

An endogeneity problem refers to one or more explanatory variables that are related to unknown random disturbance terms. It is impossible to list all explanatory variables in Equations (1) and (2), which will cause the impact of omitted variables to be included in the random disturbance item, thus causing an endogeneity problem. Although adding control variables can alleviate the problem caused by omitted variables to some extent, there may still be endogeneity problems caused by two-way causality and other problems. The improvement of the quality of manufacturing development not only opens up a broad market for the growth of the service sectors, but also encourages more manufacturing enterprises to participate in international competition and promote the expansion of opening up. All this means that endogeneity problem is inevitable. In order to solve it, the following processing ideas are adopted: First, the instrumental variable method is used for testing in the benchmarking model, taking the lagged capital deepening, knowledge flow and openness as instrumental variables, because they are highly related to their current explanatory variables and not related to current error terms. Secondly, the dynamic panel method is used to test in the regulatory impact analysis model. In addition, the dynamic panel model is formed while the model is endogenous because the lagged terms of development index in the manufacturing industry are introduced into the model. The system generalized method of moments (Sys-GMM) itself has an instrumental variable to solve the endogeneity problem of dynamic panel model.

4.2. Empirical Results and Analysis

4.2.1. Regression Results of Benchmark Model

Columns (1), (2) and (3) in

Table 2. Regression Results of the Impact of the growth of Service Sectors on the Quality of Manufacturing.

Explanatory Variable	Equation (1)			Equation (2)
	(1)	(2)	(3)	(4)	(5)	(6)
	OLS	OLS	IV-2SLS	SYS-GMM	SYS-GMM	SYS-GMM
${mi}_{i,t-1}$	——	——	——	0.791 ***	0.679 ***	0.554 ***
				(0.001)	(0.000)	(0.000)
${mi}_{i,t-2}$	——	——	——	0.071 **	0.024 *	0.016 **
				(0.019)	(0.058)	(0.037)
${sd}_{i,t}$	0.497 ***	0.209 ***	0.227 ***	0.154 **	0.127 **	0.126 *
	(0.000)	(0.000)	(0.000)	(0.014)	(0.040)	(0.052)
${\mathit{ins}}_{i,t}$	——	——	——	——	0.082 ***	0.097 ***
					(0.000)	(0.000)
${sd}_{i,t}{*\mathit{ins}}_{i,t}$	——	——	——	——	——	0.077 **
						(0.007)
${\mathit{capd}}_{i,t}$	——	0.572 ***	0.594 ***	0.408 **	0.334 **	0.318 **
		(0.000)	(0.000)	(0.024)	(0.045)	(0.024)
${\mathit{knf}}_{i,t}$	——	0.307 ***	0.382 ***	−0.193	0.223 **	0.276 **
		(0.000)	(0.000)	(0.108)	(0.017)	(0.032)
${\mathit{od}}_{i,t}$	——	0.868 **	0.903 **	0.679 *	0.449 *	0.397 *
		(0.035)	(0.028)	(0.081)	(0.073)	(0.088)
${\mathit{inf}}_{i,t}$	——	0.114 **	0.169 **	0.042 **	0.052 ***	0.057 ***
		(0.012)	(0.017)	(0.017)	(0.000)	(0.000)
$C$	−0.209 **	−0.172 **	−0.202 **	0.791 **	0.680 ***	0.830 ***
	(0.028)	(0.019)	(0.037)	(0.021)	(0.000)	(0.000)
R2	0.608	——	——	——	——	——
Hausman Test	——	62.694 ***	——	——	——	——
		(0.000)
Kleibergen–Paap rk LM statistics	——	——	54.344 ***	——	——	——
			(0.000)
First stage regression F value	——	——	36.273 ***	——	——	——
			(0.000)
AR(1)	——	——	——	−2.104 **	−2.092 **	−2.572 **
				(0.031)	(0.039)	(0.017)
AR(2)	——	——	——	−0.558	−0.901	−0.562
				(0.582)	(0.473)	(0.835)
Sargan Test	——	——	——	22.906	23.704	25.9285
				(0.626)	(0.870)	(0.787)
Observations	330	330	330	330	330	330

The data in parentheses represent the p-values: ***, ** and * indicate 1%, 5% and 10% p-values, respectively.

are the regression results of the benchmarking model. In column (2), control variables are included, and Equation (1) is tested. The endogeneity problems are considered and estimated by IV-2SLS method in column (3). Table 2 reports the test results of the effectiveness of the instrumental variables. The results show that the selection of the instrumental variables is appropriate and effective.

According to column (2) of Table 2, the coefficient of $sd$ is 0.209, which is significant at the level of 1%, after effective control of individual differences and inclusion of control variables. This shows that service growth can significantly improve the quality of manufacturing development. Compared with column (1), the impact of service sector growth may be overestimated without adding control variables. However, compared with column (3), the impact effect is underestimated without considering endogenous factors. However, from the results of the first three columns, although the coefficients of $sd$ are different, their coefficients are all greater than zero, which preliminarily verifies Hypothesis 1.

In fact, the impact of the growth of the service sectors on the development of the manufacturing industry generally works through two opposite forces: On the one hand, as an intermediate input of the manufacturing industry, the service sector can improve the innovation ability of enterprises, increase the value-added of industrial products, enhance profitability and improve the quality of the manufacturing development. In addition, it is called the “intermediate effect” of growth of service sectors. On the other hand, the development of any industry needs to input various factors such as capital, labor and technology, and form a competitive relationship with others under the circumstances of limited resources. Therefore, the growth of the service sectors may squeeze out all kinds of resources needed for the development of the manufacturing industry, which hinders the improvement of its development quality. Here, it is defined as the “competitive effect” of the growth of service sectors [68]. Whether growth in service sectors can improve the quality of manufacturing development depends on the strength of the above two forces. In other words, when the intermediate effect is stronger than the competition effect, there is a positive correlation between them; when the two forces are equal, the growth of the service sector cannot promote the improvement of the quality of the manufacturing development. However, when the former is weaker than the latter, they are negatively correlated, which means the growth of the service sectors inhibits the improvement of the quality of the manufacturing development. The empirical results show that in China, the “intermediate effect” of the growth of the service sectors is stronger than the “competitive effect”, which means the growth of the service sectors improves the quality of the manufacturing development, which tests H1. However, the huge inertia of the manufacturing development model and the possible role of institutional environment are ignored in Equation (1), which may expand the impact of the growth of the service sector and cause errors in the estimation coefficient. Therefore, Equation (2) needs to be checked.

4.2.2. Further Analysis

The premise that the Sys-GMM can establish is that there is no autocorrelation of the disturbance term. The test results in Table 2 show that the first-order autocorrelation of the disturbance term is significant at least at the level of 5%; the original hypothesis is rejected. However, the second-order autocorrelation test is not significant at least under 10%, that is, there is no second-order autocorrelation. In other words, the Sys-GMM is valid. In addition, due to the use of instrumental variables in this method, it is necessary to conduct an over-identification test of instrumental variables. The test results show that, at the level of 10% significance, the null hypothesis is acceptable, that is, all instrumental variables in Equation (2) are valid. This shows that Equation (2) can be used to estimate the Sys-GMM.

In order to more effectively reflect the regulatory impact of the system, the lagged terms of $mi$, $ins$ and its interaction with $sd$ are introduced into Equation (1). The results of columns (4) to (6) in Table 2 show that the regression coefficients of the lagged terms of $mi$ in two periods are greater than zero and significant at the level of 10%. This means that the setting of the regulatory effect model is reasonable. The regression coefficients of $sd$ are 0.154, 0.127 and 0.136, respectively, and are significant at least at the level of 10%, which further validates Hypothesis 1.

The results In column (5) of Table 2 show that the regression coefficient of $ins$ is 0.082, which is significant at the level of 1%. It means that the optimization of the institutional environment can significantly improve the quality of manufacturing development. However, in column (6), the regression coefficient of the interaction variable between $ins$ and $sd$ is 0.077, and is significant at the level of 1%, which means that there is a significant interaction between them. The impact of the growth of service sectors on the quality of manufacturing development is expressed as:

$$\frac{\partial m{i}_{i,t}}{\partial s{d}_{i,t}}=0.126+0.077in{s}_{i,t}$$

(3)

Equation (3) shows that each unit of increase in the value-added per capita of the service sectors will promote the quality of the manufacturing development by 0.126 percentage when $ins$ is the lowest in the country. However, when $ins$ is one unit higher than the minimum value, $sd$ will promote $mi$ by 0.203 percentage, which is 0.077 percentage higher than the minimum value of $\mathit{ins}$. This shows that not only the growth of service sectors will help improve the quality of manufacturing development, but also that the promotion and improvement of the market system reform will further enhance this role, that is, the institutional variables have a regulatory effect. This is because the “intermediate effect” of the service sectors can reduce the cost of the manufacturing industry, intensify the competition in downstream industries and alleviate financing constraints, thus promoting the improvement of the quality of the development of the manufacturing industry. The optimization of the institutional environment marked by the improvement of market economic system speeds up the pace of service sector development, thereby amplifying the impact of service sectors. It is similar to the change in service sector growth. The impact of the improvement of the institutional environment on the quality of the manufacturing development increases with the rapid growth of the service sectors. This may be caused by the fact that the growth of the service sector requires a higher institutional environment. In order to achieve economic growth, governments at all levels are forced to speed up the improvement of the market economic system and optimize the development environment. In short, not only does the improvement of institutional environment contribute to the improvement of the quality of manufacturing development, but it also has a significant interaction with the growth of service sectors, which verifies Hypothesis 2.

In addition, it can be seen from Table 2 that the regression coefficients of ${\mathit{capd}}_{i,t}$, ${\mathit{od}}_{i,t}$ and ${\mathit{inf}}_{i,t}$ are positive and significant at least at the level of 10%, that is, the capital deepening, the expansion of openness and the improvement of infrastructure are positively correlated with the quality of manufacturing development. It is not difficult to explain that technological progress is one of the essential methods of high-quality development of the manufacturing industry. The application of new technologies often requires enterprises to invest in higher-value machinery and equipment, which leads to the improvement of capital per capita. The expansion of the degree of opening to the outside world can not only expand the market of manufacturing products and reduce their costs, but also obtain more advanced technology, improve the management level and promote the improvement of the quality of manufacturing development through foreign exchanges. The impact of infrastructure improvement on the quality of manufacturing development mainly comes from two aspects: on the one hand, the improvement of traditional infrastructure can effectively reduce the intermediate input cost of the manufacturing industry, and help to better realize its value, improve its income and development quality. On the other hand, digitalization and intellectualization provide a new path for technological innovation and green development of the manufacturing industry, and the improvement of information infrastructure investment will become the basic support and guarantee for intelligent development of manufacturing industry. However, when the institutional effect is ignored in Equation (2), the regression coefficient of ${\mathit{knf}}_{i,t}$ is −0.193, which is not significant at the level of 10%; this is contrary to the theory. However, after introducing institutional variables, their regression coefficients are 0.2237 and 0.276, respectively, which are significant under 5%, in line with expectations. The possible explanation is that the data quality is not high. These facts are consistent with the conclusions of existing studies.

5. Robustness Test and Heterogeneity Analysis

5.1. Robustness Test

In order to ensure the effectiveness and reliability of the estimation results, the following two aspects are mainly considered. The first method is the explained variable replacement method. Yongze [69] concludes that total factor productivity (TFP) is an important variable to understand and measure high-quality economic development. Following this conclusion, this paper selects industrial TFP as a substitute variable for high-quality manufacturing development to test Equation (2).

Data envelopment analysis (DEA) does not need to set specific input–output production function shape and weight, so it is not affected by model setting bias and human subjective factors. Ma and Suri use the DEA method to measure TFP in different regions of China in view of the bias in the setting of production function [70]. This paper uses this method for reference to measure industrial total factor productivity. With the gross industrial output value as the output variable, and the net value of fixed assets and the average annual employees as the capital and labor input factors, the growth rate of industrial total factor productivity is calculated using the output-oriented DEA model under the condition of constant returns to scale.

The net capital is calculated using the perpetual inventory method, and the specific calculation method is based on the practice of Haojie [66]. In order to eliminate the impact of price, we have made price adjustment to the gross industrial output value and the net value of fixed assets. The gross industrial output value is adjusted by the industrial GDP deflator, and the net capital value is adjusted by the fixed asset investment price index, with 2010 as the base period.

Secondly, the explanatory variable replacement method is adopted. The share of value-added of service sectors in GDP is used as the measurement indicator of the growth of service sectors to test the model renewal in Equation (2).

Table 3. The results of our robustness checks.

Variables	Replace the Explained Variable	Replace the Explanatory Variable
${sd}_{i,t}$	0.091 **	0.115 **
	(0.014)	(0.019)
${\mathit{ins}}_{i,t}$	0.825 ***	0.091 ***
	(0.000)	(0.000)
${sd}_{i,t}{*\mathit{ins}}_{i,t}$	0.066 **	0.062 ***
	(0.023)	(0.000)
AR(1)	−3.241 ***	−3.983 ***
	(0.000)	(0.000)
AR(2)	−1.072	−0.941
	(0.335)	(0.236)
Sargan Test	39.31	46.05
	(0.767)	(0.857)
Observations	330	330

The data in parentheses represent the p-values: *** and ** indicate 1% and 5% p-values, respectively.

shows the results of the robustness test. Under the two test methods, the growth of the service sectors and the optimization of the institutional environment still have a significant positive correlation with the development quality of the manufacturing industry. The coefficient of their interaction variables is still greater than zero and is significant at the level of 1%. The interaction variables of these two variables still influence each other and promote the high-quality development of the manufacturing industry. It can be seen that the test model constructed by the above two methods is consistent with the conclusions of the benchmarking model and the regulatory impact model. Therefore, the constructed model has a certain robustness.

5.2. Heterogeneity Analysis

The foregoing empirical research reveals the impact of institutional environment and the growth of service sectors growth on the quality of manufacturing development. However, will the imbalance of regional economic development and the different investment purposes of the service sector lead to the heterogeneity of the impact of service industry? This is the question that needs to be answered in this section.

5.2.1. Regional Heterogeneity Analysis

In China, the imbalance of regional economic development is widespread and difficult to eliminate in the short term. The value-added per capita of service sectors in the eastern, central and western regions are CNY 188,300, 126,200 and 119,800, respectively, in 2020. In addition, those of the eastern regions are 1.49 times those of central regions and 1.57 times those of western regions. In the same period, the average number of effective invention patents per 10,000 people of industrial enterprises in the three regions are 17.76, 5.71 and 3.40, respectively. In addition, the pace of marketization reform varies from region to region. The average scores of the total marketization index in the eastern, central and western regions are 7.05, 5.83 and 4.64, respectively, in 2019. This means that the impact of service sectors and institutional environment on the quality of manufacturing development may be different in different regions. Therefore, all samples are divided into three regions: the eastern, the central and the western regions, according to the classification method commonly used in economic analysis by National Bureau of Statistics of China. The results are shown in columns (1), (2) and (3) of

Table 4. Heterogeneity Analysis of Service Sector Growth.

Variables	Regional Heterogeneity			Heterogeneity of Investors
	(1)	(2)	(3)	(4)	(5)
	Eastern Regions	Central Regions	Western Regions	SFDI	Domestic-Funded Services
${sd}_{i,t}$	0.341 **	0.076 *	0.052 **	0.117	0.159 ***
	(0.039)	(0.064)	(0.035)	(0.240)	(0.000)
${\mathit{ins}}_{i,t}$	0.213 ***	0.101 ***	0.023 ***	0.108 ***	0.098 ***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
${sd}_{i,t}{*\mathit{ins}}_{i,t}$	0.194 ***	0.081 *	0.037 **	0.065	0.082 ***
	(0.003)	(0.061)	(0.017)	(0.307)	(0. 000)
AR(1)	−2.91 ***	−3.07 ***	−2.618 ***	−2.989 ***	−3.186 ***
	(0.000)	(0.000)	(0.008)	(0.003)	(0.001)
AR(2)	−0.399	−0.323	−0.245	−0.342	−0.240
	(0.655)	(0.741)	(0.802)	(0.733)	(0.810)
Sargan Test	22.475	24.67	21.31	23.653	20.426
	(0.489)	(0.502)	(0.421)	(0.579)	(0.469)
Observations	121	88	121	297	330

Note: The data in parentheses represent the p-values: ***, ** and * indicate 1%, 5% and 10% p-values, respectively. Division of east, central and west: the eastern regions include Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong and Hainan; The central regions include Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei and Hunan; The western regions include Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia and Xinjiang.

.

It can be seen from Table 4 that the regression coefficients of $sd$, $ins$ and their interaction variables in the eastern, central and western regions are greater than zero, and are significant at least at the level of 10%, which further verifies the above conclusions. However, the coefficients of the three variables show a downward trend in the eastern, central and western regions, which means that there are significant differences in their impact on the quality of manufacturing development in different regions. Taking the value-added per capita of service sectors as an example, when the institutional variables are at the lowest level in the eastern, central and western regions, the regression coefficients of the variables are 0.341, 0.076 and 0.052, respectively. This shows that although the impact of the growth of service sectors on the manufacturing industry is slightly stronger in the central regions than in the western regions, it is far lower than in the eastern regions, and other variables show a similar trend of change. The possible reasons are mainly that the rapid development of the service sectors in the eastern regions, especially the producer services industry, has provided the manufacturing industry with rich and high-quality advanced production factors, and promoted the improvement of the quality of the manufacturing development. In contrast, the service sectors in the central and western regions have lagged behind, reducing their impact.

With the fierce competition and the upgrading of technology, the manufacturing industry has stepped into a stage where both qualitative expansion and quantitative growth are emphasized. The healthy environment formed by the rapid market-oriented reform has promoted the growth of the service sectors, which will provide strong support for the improvement of the quality of manufacturing development. However, the relatively lagging process of market economy reform and the poor business environment in the central and western regions have greatly weakened the role of the institution and spillover effects. In short, the quality of the innovation ecological chain in the eastern regions is far better than that in the central and western regions [71], resulting in differences in the impact of the growth of service sectors on the development of manufacturing in different regions.

5.2.2. Analysis on the Heterogeneity of Investors

Considering that the purpose of foreign direct investment enterprises in service sectors entering the host country is different from that in local service sectors, the impact of the development of different types of service sectors on the quality of manufacturing development may be different. Therefore, from the perspective of investors, this paper distinguishes the service sectors from domestic-funded and foreign direct investment in service sectors (SFDI), and examines the difference of their impact on the quality of manufacturing development based on Equation (2). Among them, the growth index of domestic service sectors multiplies the share of value-added of service sectors in GDP by the proportion of the number of domestic service enterprises in the service sectors. The growth index of foreign direct investment in service sectors is divided by the total foreign direct investment in service sectors actually used by each region. Since the data of FDI industry segments in Jilin Province, Sichuan Province and Hainan Province are unavailable, these three provinces were not included in the sample when investigating the impact of FDI in the service sectors. The regression results are reported in columns (4) and (5) of Table 4.

The results in column (4) of Table 4 show that the regression coefficient of $ins$ is 0.108, which is significant at the level of 1%. However, although the growth index of the service sectors and the coefficient of interaction variables with the institution are greater than zero, they are not significant at the level of 10%. This means that the entry of foreign direct investment in service sectors has not significantly improved the quality of manufacturing development in China. This is not difficult to explain because having an appropriate transmission path is a prerequisite for the service sector to play its role in the “intermediate effect”. When the services provided by foreign direct investment enterprises are just the intermediate inputs required by the manufacturing industry, the former can affect the development of the latter. Once the channel connecting the two does not exist or is cut off, foreign direct investment enterprises in the service sectors cannot have an impact on the quality of manufacturing development.

In fact, the manufacturing industry in China still has not gotten rid of the extensive development mode dominated by primary production factors. The technology level is relatively low, and the servitization level of the manufacturing industry lags far behind that of developed countries [72], resulting in insufficient demand for modern service industries such as information technology and business services. In 2020, the three sectors with the largest investment in foreign direct investment in service sectors are real estate, leasing and business services, and information transmission and computer services, accounting for 22.842%, 20.061% and 13.022% of foreign investment, respectively. However, the top two services in China’s manufacturing industry are wholesale and retail and transportation, accounting for 10.948% and 7.901%, respectively.

The services provided by multinational corporations may enter China more to meet the needs of downstream foreign-invested manufacturing enterprises. As a result, this does not match the services invested by domestic manufacturing industry in structure, so it is difficult to promote the improvement of its development quality. According to the results in column (5), we find that both domestic service sector growth and institutional environment and their interaction variables are positive and significant at the level of 1%, which also verifies the robustness of the above conclusions from another aspect. In summary, these results confirm Hypothesis 3.

6. Conclusion Remarks, Policy Implications and Future Research Directions

6.1. Conclusion Remarks

Based on the basic characteristics of China in the period of institutional transition, this paper uses the panel data of 30 provinces, autonomous regions and municipalities from 2010 to 2020 to examine the impact of service sector growth and institutional environment on the quality of manufacturing development, and analyze the heterogeneity of regions and investors. The main findings are as follows:

(1)

Based on the new development concept, and focusing on the unity of quality and quantity in the development process, this paper constructs a more scientific and reasonable manufacturing development quality evaluation indicator system and calculates it. Existing literature shows that most scholars take the whole society or region as the research object to measure the quality of economic and social development, and few scholars directly measure the development quality of manufacturing industry. Compared with Yulin’s research [11], the indicator system method in this paper is adopted to overcome the defect that a single indicator is difficult to use to measure the development quality of the manufacturing industry comprehensively. Although Hongxiang [57] adopted the indicator system method, it was composed of three aspects of economic development quality, efficiency and power. It paid more attention to the allocation of resources and the realization of value, but also ignored green development and could not reflect the new development concept. Once the new development concept rises to the level of national strategy, it will become the guide of China’s economic and social practice and influence the development direction of the manufacturing industry.

(2)

The result proves that both the growth of service sectors and the optimization of institutional environment can significantly improve the development quality of China’s manufacturing industry, and that there is a significant interaction between them, with the interaction coefficient greater than zero. In other words, the growth of service sectors has a positive effect on the improvement of the development quality of manufacturing industry in China, and this effect will be enhanced with the optimization of the institutional environment. Of course, the optimization of the institutional environment has the same function. Different from the existing studies, this article emphasizes the impact of the growth of service sectors on the quality of manufacturing industry development, rather than the effect on the structural upgrading, production efficiency and export competitiveness of the manufacturing industry, which is the concern of most scholars. Although these factors reflect the quality of the development of the manufacturing industry from a certain aspect, they cannot cover the full picture. In addition, although some scholars have analyzed the impact of institutional reform on the quality of manufacturing development qualitatively in theory, and even verified it empirically, they examine the impact of institutional reform from the perspective of structural change. The role of institutional environment in this paper is embedded into the impact of service sector growth on the quality of manufacturing development.

(3)

Heterogeneity analysis shows that, from the perspective of regional development, the growth of service sectors and institutional environment are significantly positively correlated with the development quality of manufacturing industry in the eastern, central and western regions of China, and the effect in the eastern regions is much higher than that in the central and western regions. However, there is little difference in the central and western regions. From the perspective of investors, the growth of domestic-funded service sectors and the optimization of institutional environment have significantly improved the development quality of China’s manufacturing industry. However, the expansion of foreign direct investment in services cannot significantly improve the quality of China’s manufacturing development. Although the optimization of the institutional environment can still promote the improvement of the development quality of the manufacturing industry, its interaction with the growth of the service sectors is still not significant. Due to differences in the perspective and purpose of research, it has not been found that scholars have analyzed the impact of service sector growth and institutional environment on the development quality of manufacturing industry in different regions of China, as well as the impact of different service sector investors on the quality of manufacturing development.

6.2. Policy Implications

The policy implications of this paper are clear:

(1) Since the growth of service sectors can improve the quality of manufacturing development in China, it is necessary to vigorously promote the rapid development of service sectors, especially in areas where service growth is relatively lagging behind. This can not only reverse the imbalance of the industrial structure, but also promote the high-quality development of the manufacturing industry, realize the benign interactive development of the industry, and promote the healthy and rapid development of the economy. The institutional environment with the improvement of the market system mechanism as the core can not only directly improve the quality of the manufacturing development, but also further promote the high-quality development of the manufacturing industry by promoting the rapid development of the service sectors. Therefore, we should continue to promote in-depth market-oriented reform, force reform by opening up, implement various institutional measures to accelerate the improvement of market systems in various regions, establish and improve a more inclusive market economy system, and provide a sound institutional mechanism and a healthy environment for economic development.

(2) Compared with the eastern regions, the quality of manufacturing development in the central and western regions is relatively slow to improve, which may further strengthen the imbalance of regional economic development and form a “Matthew effect”. It is necessary to adopt appropriate administrative measures through top-level design to form a cohesive national development and introduce policies and measures to promote balanced regional economic development. At the same time, the central and western regions need to create a healthy business environment, effectively promote market system reform, optimize the internal structure of the service industry and encourage enterprises to engage in innovative activities.

Although the increase in foreign direct investment in services does not significantly promote the high-quality development of manufacturing in China, it does not mean that multinational enterprises do not need to invest in services. With the transformation and upgrading of manufacturing in China and the continuous increase in the value-added of products, it is bound to need to invest more in modern service sectors, especially in the southeast coastal areas and other developed regions that are beginning to export more and more high-tech products, which is the advantage of multinational service enterprises. The introduction of service multinationals, on the one hand, can meet the needs of different types of domestic enterprises for modern services, especially the needs of multinational companies or China’s high-tech enterprises for high-end productive services abroad. On the other hand, this can bring a “spillover effect” to local producer service enterprises, promote their growth and lay a foundation for high-quality development of manufacturing industry. Indeed, when formulating policies and measures, local governments need to combine the structural characteristics and technical conditions of local manufacturing industry, and introduce foreign direct investment enterprises in the service sectors according to local conditions and priorities, so as to more effectively promote the high-quality development of the manufacturing industry.

(3) It is important to strengthen the protection of intellectual property rights, improve intellectual property transactions, accelerate the flow of technology in different fields and promote the technological level of enterprises. This can effectively improve the quality of manufacturing development. Governments at all levels should unswervingly increase the opening-up, enhance the vitality and competitiveness of enterprises and effectively promote the high-quality development of our manufacturing industry. It is suggested to optimize the business environment, stimulate the innovation drive of enterprises, accelerate the reform of the education system, enhance the quality of workers, expand human capital investment, promote the improvement of manufacturing productivity and provide basic conditions for the transformation of new and old drivers of manufacturing. This paper also suggests improving infrastructure, stimulating enterprises to expand investment and promoting the transformation of manufacturing from labor-intensive to capital-intensive and knowledge-intensive types by focusing on digitalization and intelligence, so as to improve the quality of development.

6.3. Future Research Directions

An interesting direction for the future is to explore the impact of FDI in service sectors on the development quality of manufacturing industry in more detail, such as verifying the transmission path or mechanism. Facing an uncertain environment, how FDI in service sectors will change in China, the size of spillover effect on local service sectors, and the impact on the quality of manufacturing development in different regions of China will also be some key issues of study. In addition, it remains to be solved whether the influence of FDI in service sectors has a threshold effect on the development quality of China’s manufacturing industry.