Industrial Poverty Alleviation, Digital Innovation and Regional Economically Sustainable Growth: Empirical Evidence Based on Local State-Owned Enterprises in China

Abstract

This paper takes the industrial poverty alleviation of local state-owned enterprises in China as the research object, and takes the local state-owned enterprises listed in Shanghai and Shenzhen A shares in China from 2016 to 2020 as the sample to empirically test the impact of industrial poverty alleviation on the sustainable economic growth of the region and consider the regulatory effect of digital innovation. This study found that China’s local industrial poverty reduction behavior in state-owned enterprises can effectively promote regional economic growth. Moreover, the digital innovation produced a regulating effect; that is, if enterprises carry out digital innovation and have a higher degree of digital innovation, their industrial poverty alleviation behavior will have a stronger role in promoting regional economic growth. This conclusion still holds even after controlling for factors of robustness and endogeneity. In addition, the study of influence mechanisms found that the proportion of primary industry in GDP was the mediating effect of industrial poverty alleviation on regional economic growth, and the proportion of primary industry in GDP had a partial mediating effect. Further heterogeneous group testing shows that the impact of industrial poverty alleviation on regional economically sustainable growth is more obvious in agriculture-related, local state-owned enterprises; non-high-tech, local state-owned enterprises; and local state-owned enterprises with subsequent poverty alleviation plans. The empirical evidence in this paper verifies the role of local state-owned enterprises’ participation in industrial poverty alleviation in promoting regional economic growth. It is a useful supplement to the research literature on the economic consequences of Chinese enterprises’ participation in targeted poverty alleviation, which helps to better understand such a phenomenon and also provides a powerful explanation for China’s poverty alleviation achievements.

1. Introduction

Poverty alleviation is a great practice carried out by the Chinese government to target poor areas and poor people in the long term. It is an organic embodiment of China’s ‘common prosperity’ ideal, and an important component of China’s high-quality economic development [1,2]. On 13 November 2013, General Secretary Xi put forward the idea of ‘targeted poverty alleviation’ for the first time, pointing out that poverty alleviation work should be ‘practical and realistic, tailored to local conditions, guided by categories, and targeted’. On 23 November 2015, the Central Committee of the Communist Party of China (CPC) and the Central People’s Government of the People’s Republic of China jointly issued the Decision on Winning the Battle against Poverty, which further clarified China’s policy measures for implementing targeted poverty alleviation and accelerating poverty alleviation in the coming period of time. Subsequently, various parts of China, its provinces and cities have also issued policy documents and implementation plans for poverty alleviation, so as to cement poverty alleviation as important work from top to bottom and from inside to outside. With interaction and coordination in poverty alleviation policies from governments at all levels, China has gradually formed a poverty alleviation and development policy system with socialist characteristics, presenting an overall pattern of large-scale poverty alleviation, and making remarkable achievements in poverty reduction [3]. According to the World Bank’s international poverty standard of $1.90 per person per day, more than 800 million poor people in China have been lifted out of poverty, accounting for about two thirds of the world’s total population benefiting from poverty reduction [4]. By the end of 2015, there were 55.75 million poor people in 832 poverty-stricken counties in China. The number of people living in poverty and the incidence of poverty in China have been reduced year by year (see Figure 1). By 2020, all the rural poor living under the current standards will be lifted out of poverty, and the poverty reduction target set by the United Nations 2030 Agenda for Sustainable Development will be fulfilled 10 years ahead of schedule. This is also an organic reflection of the effect of the poverty alleviation policies of the Chinese government [5].

In the first stage of the Chinese government’s poverty alleviation, the income of poor residents is mainly increased through transfer payments, expenditure reduction, etc., such as increasing financial support for poor households, increasing the reimbursement rate of rural medical insurance, and exempting various insurance premiums. In the second stage, the Chinese government carried out targeted poverty alleviation mainly by improving the endogenous power of poor areas, via methods such as relocation, employment poverty alleviation, industrial poverty alleviation, photovoltaic poverty alleviation, tourism poverty alleviation, and e-commerce poverty alleviation, among which industrial poverty alleviation not only promoted the transformation of economic and industrial structures of poor areas in China, but also brought more employment opportunities for residents in poor areas; it also improved infrastructure construction in poor areas [6]. Enterprises are important participants in industrial poverty alleviation, helping poor areas realize the transformation from resource advantages to industrial advantages and economic advantages, as well as being the most active subjects to strengthen the awareness of poverty and stimulate the endogenous power of poverty alleviation in poor areas. They can also make up for the lack of financial funds invested by the government to fight poverty, so as to promote the sustainable development of the economy in poor areas [7]. Among Chinese enterprises involved in industrial poverty alleviation, Chinese state-owned enterprises have always been the main force in the fight against poverty because of their economic, political and social responsibilities for China’ economic and social development. By virtue of their advantages in capital, talent, technology, market, etc., SOEs fully tap the resource endowments of poor areas and regard the poor as an important link in the enterprise value chain, thus effectively promoting the benign development of industries and economies in poor areas [8]. During 2016–2020, China’s central state-owned enterprises helped 221 poverty-stricken counties lift themselves out of poverty, invested more than 98 billion yuan in total, and built more than 8000 industrial poverty alleviation projects in poverty-stricken areas, driving investment totaling 14.7 billion yuan, and helping 104,400 laborers in poverty-stricken areas achieve transfer employment. Compared with China’s central state-owned enterprises, China’s local state-owned enterprises have more initiative and enthusiasm to participate in poverty alleviation. On the one hand, local state-owned enterprises have a better understanding of the economic situation, industrial situation and the situations of poor areas, and can rely on their own resource advantages and professional expertise to develop targeted poverty alleviation strategies more ‘according to local conditions’. On the other hand, by participating in targeted poverty alleviation, local state-owned enterprises can drive their regions to overcome poverty and improve regional economic development, which is not only an important responsibility by which local state-owned enterprises fulfill their social responsibilities, but also can feed back the sustainable development of local state-owned enterprises. Therefore, it has important theoretical and practical significance to study the participation of local state-owned enterprises in poverty alleviation, effectively explore the mechanism and path of local state-owned enterprises in China to promote regional economic growth through industrial poverty alleviation, and provide empirical evidence for industrial poverty alleviation of local state-owned enterprises in China. China has actively promoted anti-poverty measures in recent years, and its achievements in anti-poverty action also provide a good realistic scenario for this study.

Numerous studies have explored the relationship between poverty reduction and economic growth. Some of them have found a negative correlation between poverty reduction and economic growth, and this negative relationship originates from numerous channels. For example, poor individuals have limited access to financial markets, which seals them off from potentially profitable investment opportunities, and they often suffer from poor health which affects their productivity; in turn, poor regions and countries have fewer individuals capable of adopting, managing and generating new technologies, they lack infrastructure and face much higher transaction costs [9,10,11]. In contrast, applying macroeconomic analysis to the linkages between poverty reduction and growth, some recent studies have provided evidence that interventions directly tackling the initial level of poverty could be beneficial in accelerating subsequent growth and making growth more effective in reducing poverty [12,13,14]. With international evidence, it is proven that faster poverty reduction is linked to faster growth in the entire developing world during 1981–2018 [15]. It can be seen that current literature ignores the important role of local state-owned enterprises, which are vital in the Chinese economy. Thus, this paper attempts to investigate the behavior of poverty reduction in local state-owned enterprises and its impact on economic growth. Meanwhile, as the world is entering a new era of digital economy, the impact of digital technology has also been ignored.

Digital economy has become a new driving force and engine for current global economic development [16,17]. In recent years, the Chinese government has actively promoted the development of the digital economy. By promoting the application of the Internet, big data, blockchain, artificial intelligence and other technologies in the real economy, enterprises are encouraged to actively start digital innovation, so as to realize the digital dividend of the development of the real economy and enterprises [18]. Through digital innovation, enterprises apply digital technology to product updates and iteration, marketing strategy innovation, production mode upgrade and business scope remodeling, management system change and organizational structure innovation, so as to realize leapfrog development and sustainable development of the enterprises. Similarly, digital innovation has also changed the original ecological chain and value chain of the industry in which the enterprise is located. It has not only changed the original boundary of the industry, but also brought new industrial participants to raise the level of competition within the industry, thus breaking the original industrial balance, continuing to drive the innovation and evolution of the business model within the industry, and promoting advancements in the industrial value chain. According to the survey data conducted by China Ai Media Consulting Company, in 2021, more than 80% of Chinese enterprise users used digital systems in their production, operation and management (see Figure 2).

Digital innovation is also widely used in the poverty alleviation work of the Chinese government. On the one hand, the digital village is an important path for the continuous integration and development of digital innovation and China’s agricultural and rural areas, as well as an important means for China to achieve rural modernization and win the battle against poverty. Since 2018, the Chinese government has vigorously promoted digital agriculture (see Figure 3) through a number of policies, increased the application of digital technology in the primary industry, determined the ideas, requirements and tasks of digital rural construction, and strengthened the construction of rural digital infrastructure. In 2020, China will build 13,700 4G base stations in poverty-stricken areas, with an optical fiber coverage of more than 95%. The total amount of network retail in poverty-stricken areas will exceed 300 billion yuan. Furthermore, 5G technology and the Internet of Things have also begun to spread in poverty-stricken areas, and the penetration rate of digital economy in the primary industry will reach 8.9% [19]. On the other hand, digital innovation is also an important starting point for Chinese enterprises to participate in industrial poverty alleviation. Through digital innovation, enterprises use satellite remote sensing, the Internet of Things, artificial intelligence and other digital technologies to promote agricultural standardized production, improve the allocation of industrial resources in poverty relief areas, and effectively open up the agricultural production and operation systems in poor areas, so as to reshape the industrial ecosystem in such areas and promote the high-quality development of industries in them [20]. In addition, digital innovation can also reduce the degree of information asymmetry between farmers in poor areas and the market, and provide more market opportunities for industries in poor areas by promoting rural e-commerce, live broadcast e-commerce and other digital marketing models [14].

Therefore, this paper takes Chinese local state-owned enterprises’ participation in industrial poverty alleviation as the research object, empirically tests the impact of enterprises’ industrial poverty alleviation behavior on the sustainable economic development of poverty-stricken areas, and considers the mediating effect of digital innovation. Compared with the existing literature, the marginal contribution of this paper is reflected as follows: first, this paper expands the research object of the effect of different targeted poverty alleviation models on Chinese enterprises. The existing studies on Chinese enterprises’ participation in targeted poverty alleviation mostly focus on their behavior in targeted poverty alleviation, and rarely distinguish the differences in participation modes in the process of targeted poverty alleviation. In addition, international evidence shows that subsidiaries of multinational companies (MNC) pay, on average, higher wages than local companies [21]; in developing countries, MNC subsidiaries have strong links backwards and forwards with domestic companies, including small and medium-sized ones, which favors employment and poverty reduction [22]. More specifically, the Multinational Enterprises of the United States (US-MNE) have a significant effect on poverty reduction of in a group of 18 developing countries [23]. This paper focuses on the industrial poverty alleviation behavior of Chinese enterprises and focuses on different models of targeted poverty alleviation, which is conducive to a clearer understanding of the effects of targeted poverty alleviation among Chinese enterprises. In addition, this paper discusses the impact of industrial poverty alleviation by Chinese enterprises on regional economic sustainable growth, which is also a useful supplement to the research on the economic consequences of targeted poverty alleviation by Chinese enterprises. Second, this paper complements the research on the mechanism of digital innovation in the targeted poverty alleviation process of Chinese enterprises. Although the existing literature has analyzed the promoting role of the development of the digital economy in China’s rural economy, it has not paid attention to the mechanism and path of promoting the sustainable development of the regional economy by enterprises using their own digital innovation in the process of targeted poverty alleviation. Therefore, the study of this paper is conducive to a clearer explanation of how Chinese enterprises can effectively promote regional economic growth in the process of industrial poverty alleviation. Thirdly, this paper enriches the research literature on the participation of local state-owned enterprises in targeted poverty alleviation in China. The existing literature discusses the targeted poverty alleviation behavior of China’s state-owned enterprises, but ignores the reality that local state-owned enterprises should become the main force of targeted poverty alleviation of China’s state-owned enterprises due to their better understanding of the regional economic development, industrial development and the actual situation of poor areas. Therefore, this paper, taking Chinese local state-owned enterprises as samples, not only enriches the literature on targeted poverty alleviation in Chinese state-owned enterprises, but also has important practical guiding significance for Chinese local state-owned enterprises to further actively participate in rural revitalization.

2. Theoretical Analysis and Research Hypothesis

2.1. Industrial Poverty Alleviation and Regional Economic Sustainable Growth

The participation of Chinese local state-owned enterprises in industrial poverty alleviation originates from the policy promotion and policy coordination of the Chinese government and local governments. The government has effectively stimulated the enthusiasm and initiative of local state-owned enterprises to participate in poverty alleviation through the supply, regulations, subsidies, support, etc. of relevant policies, and has also established a policy system for enterprises to participate in industrial poverty alleviation through systematic design, so as to provide support for industrial poverty alleviation and sustainable development in China’s poor areas in the form of policy supply [24]. As one of the multiple kinds of entities involved in poverty alleviation, local state-owned enterprises can effectively marshal the elemental resources of local economic development, achieve the integration of different resources, promote market competition in poverty alleviation areas, and promote the upgrading of industries in these areas, such as by promoting the ‘complementary chain’ of the industrial chain, strengthening the brand construction of regional industries, and improving the service system of the whole industrial chain to promote the construction of the industrialization joint system [25]. This means that the participation of local state-owned enterprises in industrial poverty alleviation can help local poor households gain benefits by participating in industries, so as to improve the self-development ability of poor households. Furthermore, local state-owned enterprises can also help local governments achieve the goals of economic development planning by meeting the needs of local governments, so as to promote the social and economic development of the region. Following current literature [26], this paper describes the impact mechanism of industrial poverty alleviation by local state-owned enterprises on regional economic growth as Figure 4.

From Figure 4, industrial poverty alleviation is a process in which local state-owned enterprises, by combining different actors (Government and Poor residents), allocate production factors according to the motives of these actors, so as to work together to build a poverty alleviation path. This is a process in which the links are complementary and integrated.

Directly speaking, the participation of local state-owned enterprises in industrial poverty alleviation can increase the income of poor residents, thus driving the economic growth of the region [15,27,28]. In the process of industrial poverty alleviation, local state-owned enterprises will formulate appropriate industrial poverty alleviation strategies for poor residents according to their poverty characteristics, development characteristics and industrial characteristics of agricultural production, so as to not only enable poor residents to obtain direct income, but also to drive and help them build corresponding industrial bases, and ensure the sustainability of poverty alleviation [22,23]. In addition, local state-owned enterprises will send staff to poor areas to help poor residents obtain more external resources, help them choose appropriate industrial projects and development forms based on their understanding of the actual situation of the poor, and mobilize the people in poor areas to actively participate in the work of industrial poverty alleviation and development, so as to drive the sustainable economic development of poor areas [29].

Indirectly, local state-owned enterprises have driven the industrial development of the regions through industrial poverty alleviation, meeting the industrial development needs of the local government, and can also drive the economic development level of the region. Local governments in China have a strong demand for industrial development, especially in the context of rapid changes in the current international economic form. Active industrial transformation and industrial upgrading can effectively drive the sustainable development of the local economy. However, being subject to practical problems such as financial pressure, industrial planning and infrastructure, many local governments are also faced with the dilemma of industrial upgrading. In particular, local governments in prefecture-level cities or county-level cities in western China are under great pressure in the process of industrial upgrading. Therefore, the industrial projects brought by local state-owned enterprises in the process of poverty alleviation can not only help the local government to complete the task of poverty alleviation, achieving the performance of the local government, but also help to drive the development of the regional economy [30] by cooperating with the industrial planning of the local government.

Therefore, we propose the following research hypothesis:

Hypothesis 1 (H1).

There is a positive relationship between industrial poverty alleviation and regional economic sustainable growth.

2.2. Industrial Poverty Alleviation, Digital Innovation and Regional Economically Sustainable Growth

As a result of the integration of modern information technology with global economic development, human production and lifestyle in recent years, the digital economy has been driven by digital innovation based on big data, cloud computing, blockchain, artificial intelligence and other technologies. It has become the commanding heights for all countries in the world to improve the quality of economic development and compete for the right to speak in the international economy [31]. The digital innovation of enterprises can strongly promote the economic development of the region. First, digital innovation itself can drive the development of regional economies. Enterprises, through digital innovation, improve the ability to handle non-standardized and unstructured data, impel the transformation of enterprise management modes and management systems, and promote production efficiency and industrial upgrading. To enhance the degree of specialization, enterprises in the same industry have brought about innovation and development in other enterprises, so as to promote performance in the enterprise itself. In this way, the economic growth of the region is promoted [32,33]. Second, there is integration between enterprise digital innovation and the high-quality growth of the real economy. Digital innovation can also promote the sustainable development of regional economies and society. Digital innovation plays an important role in demand, supply and market transactions matching supply and demand. It also improves the fairness of regional economic development, improves economic efficiency, and promotes regional economic efficiency [34]. Third, digital innovation promotes the development of the industrial factor market, which also allows the ownership of data to flow from manufacturers to consumers, maximizing the value of data, and becoming a powerful driver of economic development in the current digital economy era [35].

Digital innovation has also played a boosting role in the process of industrial poverty alleviation of local state-owned enterprises in China. On the one hand, digital innovation has promoted the flow of information and data in poor areas of China, creating more opportunities for industrial remodeling and upgrading. The development of the digital economy has been continuously penetrating into the development of Chinese agriculture and rural areas, and has also played a role in promoting the development of such areas. For example, the penetration rate of the digital economy in China’s primary industry was 8.2 percent in 2019 and had risen to 8.9 percent in 2020. The development and penetration of digital innovation in poverty-stricken areas in China improves the information transmission ability of poverty-stricken areas, reduces the information barrier in these areas, accelerates the transmission speed of data and information, and thus optimizes the flow channel of digital factor resources in urban and rural areas [36]. On the other hand, digital innovation also brings the application of digital technology into the poor areas of China, and expands the path of industrial development in these poor areas. Relying on the implementation of digital innovation, enterprises can apply more digital technologies to the process of industrial poverty alleviation, such as using big data technology to identify the industrial characteristics of poor areas, providing more industrial data, and using big data and cloud computing technology to expand the marketing path of industrial products in poor areas. In particular, when local state-owned enterprises in China participate in industrial poverty alleviation, many enterprises help poor areas to build on the model of ‘digital technology + e-commerce + industry’. By combining digital technology with e-commerce, they sell products from poor areas, or build the model of ‘digital technology + tourism + industry’. Digital leisure agriculture and other ways are used to drive the industrial transformation of poor areas so as to optimize the allocation of resources in these areas [37].

Digital innovation and industrial poverty alleviation have ‘three integrations’ of value, rule and mode in promoting local economic development (see Figure 5). In terms of value integration, industrial poverty alleviation recognizes the values of market orientation and profit creation, while digital innovation recognizes the values of innovation orientation and value reproduction in economic development and industrial reform [38]. In terms of law integration, industrial poverty alleviation proposes the law of poverty alleviation, while digital innovation proposes the law of new economic development [39]; In terms of mode integration, industrial poverty alleviation promotes the sustainable development of industries in poor areas, while digital innovation promotes the transformation and development of economy, industry and commerce in poor areas [40,41]. Therefore, under the effect of ‘three integrations’ and through the promotion of digital innovation, industrial poverty alleviation will play a stronger role in regional economic growth. This means that digital innovation will play a corresponding role between industrial poverty alleviation and regional economic growth. On the one hand, digital innovation reduces the cost of industrial poverty alleviation to promote regional economic growth, which will further increase the efforts of enterprises to strengthen industrial poverty alleviation [42]. Digital innovation reduces the degree of information asymmetry between local state-owned enterprises, local governments and poor residents involved in industrial poverty alleviation, which is conducive to better playing the role of market mechanism in poor areas, and also conducive to the formation of network effect and agglomeration effect in such areas, thus effectively boosting local economic development [43,44]. On the other hand, new industrial models and technologies brought about by digital innovation effectively boost regional economic growth by promoting industrial upgrading. By bringing digital technology to poverty-stricken areas, local state-owned enterprises have promoted e-commerce poverty alleviation, tourism poverty alleviation and other models, accelerated the formation of characteristic industries, and laid a good foundation for the development of these industries, which not only achieved the effect of poverty alleviation, but also provided opportunities for subsequent economic development in poverty-stricken areas [40,45].

Therefore, we propose the following research hypothesis:

Hypothesis 2 (H2).

Digital innovation has a moderating effect between industrial poverty alleviation and sustainable regional economic growth.

3. Study Design

3.1. Variable Design

3.1.1. Dependent Variable: Regional Economic Sustainable Growth

This paper measures the annual GDP growth rate (RESG) of the poverty alleviation regions where the sample companies are located.

3.1.2. Independent Variable: Industrial Poverty Alleviation

This paper measures the dummy variables and degree variables of sample companies’ participation in industrial poverty alleviation (IPA).

Participation in industrial poverty alleviation (WIPA): measured by whether the sample company participates in industrial poverty alleviation, that is, if the sample company participates in industrial poverty alleviation, WIPA = 1, otherwise WIPA = 0.

Industrial poverty alleviation degree (DIPA): measured by the natural logarithm of the industrial poverty alleviation investment amount of the sample company, namely DIPA = ln (amount of industrial poverty alleviation investment + 1).

3.1.3. Intervening Variable: Digital Innovation

This paper measures the dummy variable and degree variable of digital innovation (DI) of sample companies. For the definition of digital innovation, this paper refers to current measurement methods which measure the term frequency of digital innovation disclosed in the periodic financial reports of sample companies [46,47,48].

The first step is to define the dimensions of digital innovation words in this paper, including ‘artificial intelligence technology’, ‘blockchain technology’, ‘cloud computing technology’, ‘big data technology’ and ‘digital technology application’, and define relevant keywords under each word (see Figure 6). The second step is to use python to capture information from the regular financial reports of sample companies.

Digital innovation (WDI): measured by the digital innovation of the sample company, that is, if the frequency of digital innovation words appears in the annual financial report of the sample company, WDI = 1, otherwise WDI = 0.

Digital innovation degree (DDI): measured by the digital innovation degree of sample companies, namely DDI = ln (number of word frequency of ‘digital innovation’ keywords + 1).

3.1.4. Controlled Variables

The following control variables are added in this paper:

Total Assets (Size): measured by the natural logarithm of the total assets at the end of the year;

Asset-liability Ratio (Debt): measured by the ratio of the total liabilities at the end of the year to the total assets;

Return on Assets (Roa): measured by the ratio of year-end net profit to total assets;

Growth Rate (Growth): Measured by the growth rate of year-end operating income of sample enterprises;

Ownership Concentration (H10): measured by the squared sum of the shareholding proportion of the top 10 shareholders at the end of the year;

Management Shareholding (MS): measured by the enterprise shares held by the management of the sample enterprise;

Enterprise Area (Area): if the sample enterprise is located in western China, then Area = 1, otherwise Area = 0;

Institutional Environment (IET): measured by the index of marketization process in the region where the sample enterprises are located.

3.2. Empirical Model Design

3.2.1. Benchmark Regression Test Model Design

In order to test the impact of industrial poverty alleviation by local state-owned enterprises on sustainable regional economic growth, this paper constructs the following benchmark regression model:

$$RES{G}_{\mathrm{i},\mathrm{t}}={\alpha }_0+{\alpha }_{1}IP{A}_{\mathrm{i},\mathrm{t}}+{\alpha }_{\mathrm{i}}Control{s}_{\mathrm{i},\mathrm{t}}+Year+Industry+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(1)

To further test the moderating effect of digital innovation, this paper constructs the following regression model:

$$RES{G}_{\mathrm{i},\mathrm{t}}={\alpha }_0+{\alpha }_{1}IP{A}_{\mathrm{i},\mathrm{t}}+{\alpha }_2\left(IP{A}_{\mathrm{i},\mathrm{t}}\times D{I}_{\mathrm{i},\mathrm{t}}\right)+{\alpha }_{3}D{I}_{\mathrm{i},\mathrm{t}}+{\alpha }_{\mathrm{i}}Control{s}_{\mathrm{i},\mathrm{t}}+Year+Industry+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(2)

In Equations (1) and (2), the Year and Industry factors of the sample company are simultaneously controlled.

3.2.2. Endogeneity Test Model Design

Although control variables have been added to Equations (1) and (2) in this paper, the endogeneity problem may still exist in the empirical model. One possibility is that a local state-owned enterprise’s own industry attribute determines its poverty reduction efforts involved in industry, such as compared to the public class of state-owned enterprises, state-owned business enterprises with stronger industry attributes can bring about more possible ways to alleviate poverty through industry, and thus may accomplish industrial poverty alleviation with regional economic growth. Therefore, the two-stage least-squares method is used for the endogeneity test in this paper. In the first stage, instrumental variables are used to estimate the explanatory variable IPA, namely:

$$IP{A}_{\mathrm{i},\mathrm{t}}={\alpha }_0+{\alpha }_{1}CS{E}_{\mathrm{i},\mathrm{t}}+{\alpha }_{i}Control{s}_{\mathrm{i},\mathrm{t}}+Year+Industry+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(3)

In Equation (3), the variable CSE is a tool variable; that is, if the sample state-owned enterprises belong to commercial state-owned enterprises, CSE = 1, otherwise CSE = 0.

According to the Guiding Opinions on the Definition and Classification of State-owned Enterprises (SOES) Functions jointly issued by the State-owned Assets Supervision and Administration Commission of the State Council, the Ministry of Finance and the National Development and Reform Commission of China in 2015, China’s SOES are classified into public welfare SOES and commercial SOES. Among them, the main objectives of commercial state-owned enterprises are to enhance the vitality of the state-owned economy, enlarge the functions of state capital, maintain and increase the value of state assets, and carry out commercial operations in accordance with market requirements, while the main objectives of public welfare state-owned enterprises are to ensure people’s livelihoods, serve society, and provide public goods and services. Referring to current studies [49], sample companies belonging to A01–05, B10–12, C13–15, C17–24, C26–30, C33, C35, C39–43, E47, E49–50, F, H, K, L, O and R88–89 in the industry classification of the China Securities Regulatory Commission are defined as commercial state-owned enterprises. The reason why this variable is selected as an instrumental variable is that, on the one hand, commercial state-owned enterprises have a better industrial foundation and are more likely to carry out industrial poverty alleviation, which meets the requirement of the correlation of instrumental variables. On the other hand, the industry attributes of local state-owned enterprises do not affect the level of regional economic growth, which also meets the requirement of exogeneity of instrumental variables. Therefore, on the basis of fitting variable IPA through Equation (3), Equations (1) and (2) are further tested.

3.2.3. Intermediary Effect Test Model Design

Local state-owned enterprises promote regional economic growth through industrial poverty alleviation. In this process, the industries built by local state-owned enterprises are basically primary industries. In other words, the poverty alleviation of local state-owned industries promotes the sustainable growth of regional economy by promoting the growth of local primary industries. Therefore, this paper takes the ratio of primary industry to GDP in the region where sample enterprises are located as the mediating variable (PGDP) to construct the mediating effect model test, namely:

$$PGD{P}_{\mathrm{i},\mathrm{t}}=a_0+a_{1}IP{A}_{\mathrm{i},\mathrm{t}}+a_{\mathrm{i}}Control{s}_{\mathrm{i},\mathrm{t}}+Year+Industry+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(4)

$$RES{G}_{\mathrm{i},\mathrm{t}}=b_0+b_{1}IP{A}_{\mathrm{i},t}+b_{2}PGD{P}_{\mathrm{i},\mathrm{t}}+b_{\mathrm{i}}Control{s}_{\mathrm{i},t}+Year+Industry+{\epsilon }_{\mathrm{i},t}$$

(5)

In Equations (4) and (5), if there is a mediation effect, the coefficient value a₁ in Equation (4) and the coefficient value b₂ in Equation (5) are significant. If the coefficient value b₁ is not significant, it means that there is a complete intermediary effect, while if the coefficient value b₁ is significant, it means that there is a partial intermediary effect. However, if at least one of a₁ and b₂ is not significant, the Sobel test is required.

3.3. Data Selection and Description

The data are extracted from the China Stock Market and Accounting Research database (CSMAR). Starting from 2016, China’s Shenzhen Stock Exchange and Shanghai Stock Exchange have required listed companies to disclose information on targeted poverty alleviation. Therefore, our sample period is from 2016 to 2020, and local state-owned enterprises listed on China’s A-share stock market are selected in our sample. Based on the original samples, this paper conducts data deletion. The principles of deletion are as follows: first, the samples of enterprises in the financial, insurance and securities industries are removed; second, the sample of enterprises with special treatment is eliminated; third, the sample of enterprises with IPO is excluded; fourth, the enterprise samples with missing data which cannot be supplemented are removed. Finally, this paper obtains 2935 samples of 799 local state-owned enterprises in China from 2016 to 2020.

4. Results and Analysis of Empirical Tests

4.1. Descriptive Statistical Results and Analysis

Table 1. Descriptive statistical results of full sample.

Variable	Mean	Median	Standard Deviation	Maximum	Minimum	5%	25%	75%	95%
RESG	0.063	0.068	0.022	0.118	−0.231	0.019	0.050	0.078	0.088
WIPA	0.269	0.000	0.444	1.000	0.000	0.000	0.000	1.000	1.000
DIPA	3.006	0.000	5.589	20.832	0.000	0.000	0.000	0.000	14.235
WDI	0.701	1.000	0.458	1.000	0.000	0.000	0.000	1.000	1.000
DDI	1.316	1.099	1.197	5.684	0.000	0.000	0.000	2.079	3.611
Size	22.987	22.896	1.330	28.636	17.954	20.991	22.056	23.801	25.297
Debt	0.496	0.501	0.207	2.123	0.027	0.163	0.336	0.646	0.832
Roa	0.048	0.046	0.076	0.745	−1.495	−0.027	0.027	0.070	0.139
Growth	0.004	0.001	0.083	4.290	−0.010	−0.004	0.000	0.002	0.006
H10	0.185	0.153	0.130	0.753	0.000	0.035	0.086	0.261	0.453
MS	0.004	0.000	0.025	0.474	0.000	0.000	0.000	0.000	0.012
Area	0.173	0.000	0.378	1.000	0.000	0.000	0.000	0.000	1.000
IET	7.473	7.047	1.790	11.109	0.969	4.557	6.393	9.054	10.290

shows the descriptive statistical results of the full sample in this paper. The mean value of RESG is 0.063. On average, in recent years, China’s overall economic growth level has been maintained at about 6%. The mean value of WIPA of the variable is 0.269, indicating that 26.9% of the local state-owned enterprises in the sample have carried out industrial poverty alleviation. The mean value of variable DIPA is 3.006, indicating that the average investment in industrial poverty alleviation of local state-owned enterprises in the sample is about 200,000 yuan. However, from the perspective of sample companies participating in industrial poverty alleviation, the average investment in industrial poverty alleviation is about 707.21 million Yuan. The mean value of WDI is 0.701, indicating that about 70% of local state-owned enterprises in the sample have carried out digital innovation. The mean value of the variable DDI is 1.316, indicating that on average, there are three instances of words related to digital innovation in the annual financial reports of local state-owned enterprises, but from the perspective of the sample companies carrying out digital innovation, the number of words related to digital innovation in the financial reports is six instances on average.

Table 2. Descriptive Statistical Results of Groups.

Variable	WIPA = 1			WIPA = 0			t Test	Wilcoxon Z
	N	Mean	Median	N	Mean	Median
RESG	790	0.066	0.070	2145	0.061	0.068	5.160 ***	4.674 ***

Note: *** indicates that they have passed the significance test at the 1% confidence level.

shows the statistical results based on the description of variable WIPA grouping. Compared with the group with variable WIPA value of 0, the average value and median value of RESG in the group with variable WIPA value of 1 are larger, and both can pass the significance test of the conventional confidence level, which indicates that compared with the local state-owned enterprises that do not carry out industrial poverty alleviation, the economic growth of the regions where the local state-owned enterprises carry out industrial poverty alleviation is faster.

4.2. Correlation Test Results and Analysis

Table 3. Correlation Test Results.

Variable	RESG	WIPA	DIPA	WDI	DDI	Size	Debt	Roa	Growth	H10	MS	Area	IET
RESG	1
WIPA	0.095 ***	1
DIPA	0.083 ***	0.886 ***	1
WDI	0.057 ***	0.071 ***	0.063 ***	1
DDI	0.052 ***	0.035 **	0.019 ***	0.718 ***	1
Size	−0.017	0.172 ***	0.169 ***	0.148 ***	0.140 ***	1
Debt	−0.012	−0.012	−0.007	0.029 *	−0.006	0.394 ***	1
Roa	−0.008	0.003	−0.002	0.009	0.005	0.042 **	−0.021	1
Growth	0.016	0.023	0.028	0.010	0.027	0.021	0.023	−0.003	1
H10	−0.025	0.131 ***	0.131 ***	0.062 ***	0.007	0.243 ***	−0.089 ***	0.026	0.040 **	1
MS	−0.037 **	−0.073 ***	−0.062 ***	0.011	0.057 ***	−0.098 ***	−0.066 ***	−0.010	−0.005	−0.110 ***	1
Area	0.081 ***	0.163 ***	0.144 ***	0.019	−0.008	−0.060 ***	0.031 *	−0.010	−0.004	0.006	−0.052 ***	1
IET	−0.008	−0.176 ***	−0.168 ***	0.039 **	0.065 ***	0.014	−0.056 ***	0.019	−0.007	0.005	0.023	−0.433 ***	1

Note: ***, **, and * indicate that they have passed the significance test at the 1%, 5%, and 10% confidence levels, respectively.

shows the correlation values of the main variables in this paper. The correlation values of WIPA, DIPA and RESG are all significantly positive, indicating that there is a positive correlation between industrial poverty alleviation of local state-owned enterprises and regional economic growth. The correlation values of variables WDI, DDI and variable RESG are significantly positive, indicating that the digital transformation behavior of local state-owned enterprises is also positively correlated with the economic growth of their regions. The correlation values between variables WIPA and DIPA and variables WDI and DDI are also significantly positive, indicating that poverty alleviation in local state-owned industries will also have a positive correlation with digital innovation. In addition, the correlation value between variables in Table 3 is not high, indicating that there is no multicollinearity problem.

4.3. Empirical Results and Analysis

4.3.1. Benchmark Regression Test Results

Table 4. Benchmark Regression Test Results (1).

	(1)	(2)	(3)	(4)
WIPA	0.0048 *** (0.0009)		0.0048 *** (0.0010)
DIPA		0.0003 *** (0.0001)		0.0003 *** (0.0001)
Size			−0.0003	−0.0003
Debt			−0.0011	−0.0013
Roa			−0.0022	−0.0021
Growth			0.0045	0.0044
H10			−0.0067 **	−0.0066 **
MS			−0.0302 *	−0.0314 *
Area			0.0049 ***	0.0051 ***
IET			0.0006 **	0.0005 **
Year	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes
Constant	0.0614 ***	0.0617 ***	0.0654 ***	0.0648 ***
Adj R2	0.0087	0.0065	0.0153	0.013
F-statistics	26.6286 ***	20.1893 ***	60.7888 ***	54.458 ***

Note: ***, **, and * indicate that they have passed the significance test at the 1%, 5%, and 10% confidence levels, respectively.

shows the test results of the impact of industrial poverty alleviation of local state-owned enterprises on regional economic growth. Before the control variables are added, the regression results (1) and (2) have significantly positive WIPA and DIPA coefficients, indicating that compared with the regions where the local state-owned enterprises do not participate in industrial poverty alleviation, the local state-owned enterprises that participate in industrial poverty alleviation and the regions where the local state-owned enterprises with higher industrial poverty alleviation efforts are located have faster economic growth. After adding the control variables, the coefficients of WIPA and DIPA in the regression results (3) and (4) are still significantly positive, which also shows that the industrial poverty alleviation behavior of local state-owned enterprises can effectively promote the economic growth of the region. It can be seen that the industrial poverty alleviation behavior of local state-owned enterprises can bring more industrial development and industrial investment to poor areas, thus driving the local industrial reform and upgrading, and effectively promoting the local economic growth, which verifies the H1 above.

Table 5. Benchmark Regression Test Results (2).

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
WIPA					0.0055 *** (0.0018)	0.0047 *** (0.0014)
DIPA							0.0003 ** (0.0001)	0.0002 ** (0.0001)
WIPA × WDI					0.0076 *** (0.0021)
WIPA × DDI						0.0011 *** (0.0001)
DIPA × WDI							0.0075 *** (0.0001)
DIPA × DDI								0.0071 *** (0.0006)
WDI	0.0028 *** (0.0009)		0.0029 *** (0.009)		0.0029 *** (0.0010)		0.0032 *** (0.0010)
DDI		0.0010 *** (0.003)		0.0010 *** (0.003)		0.0011 *** (0.0004)		0.0012 *** (0.0004)
Size			0.0002	0.0002	−0.0001	−0.0001	−0.0001	−0.0001
Debt			−0.0022	−0.0025	−0.0012	−0.0015	−0.0014	−0.0017
Roa			−0.0023	−0.0024	−0.0021	−0.0023	−0.0021	−0.0022
Growth			0.0051	0.0053	0.0046	0.0049	0.0045	0.0045
H10			−0.0054 *	−0.0060 *	−0.0064 *	−0.0072 **	−0.0065 **	−0.0072 **
MS			−0.0324 *	−0.0307 *	−0.0285 *	−0.0268 *	−0.0298 *	−0.0283 *
Area			0.0058 ***	0.0057 ***	0.0051 ***	0.0050 ***	0.0053 ***	0.0051 ***
IET			0.0005 *	0.0005 *	0.0006 **	0.0006 **	0.0006 **	0.0006 **
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.0646 ***	0.0640 ***	0.0589 ***	0.0581 ***	0.0634 ***	0.0627 ***	0.0630 ***	0.0623 ***
Adj R2	0.0029	0.0024	0.0106	0.0100	0.0186	0.0177	0.0166	0.0160
F-statistics	3.5479 ***	3.0220 ***	4.4774 ***	4.2841 ***	16.0552 ***	17.949 ***	15.5010 ***	15.3338 ***

Note: ***, **, and * indicate that they have passed the significance test at the 1%, 5%, and 10% confidence levels, respectively.

shows the test results of industrial poverty alleviation and regional economic growth considering the moderating effect of digital innovation. In regression results (1) and (2), before adding control variables, the coefficient values of variables WDI and DDI are significantly positive, indicating that compared with local state-owned enterprises without digital innovation, local state-owned enterprises with digital innovation and higher degrees of digital innovation have faster economic growth in their regions. However, in the regression results (3) and (4), after adding the control variables, the coefficient values of variables WDI and DDI are still significantly positive, which also indicates that the digital innovation behavior of local state-owned enterprises can also promote the economic growth of their regions. Considering the moderating effect of digital innovation, in regression results (5) and (6), after adding the interaction term, the WIPA coefficient value of the variable is still significantly positive, and the interaction terms WIPA × WDI and WIPA × DDI are also significantly positive, indicating that under the conditions of considering the influence of digital innovation, the industrial poverty alleviation behavior of local state-owned enterprises plays a stronger role in promoting regional economic growth. The regression results (7) and (8) show a similar situation. The variable DIPA and the interaction terms DIPA × WDI and DIPA × DDI are all significantly positive, which also verifies that digital innovation has a moderating effect between industrial poverty alleviation and regional economic growth. It can be seen that local state-owned enterprises improve their digital capability through digital innovation, and also provide technical support for their participation in industrial poverty alleviation through digital technology, ensuring the full effect of industrial poverty alleviation, so as to be more conducive to promoting regional economic development, which verifies H2 above.

In order to verify the robustness of benchmark regression results, this paper conducts corresponding robustness tests. First, this paper tests the investment of local state-owned enterprises in industrial poverty alleviation as an explanatory variable. Furthermore, we use the number of projects of local state-owned enterprises in industrial poverty alleviation as an explanatory variable to conduct a new empirical test. Second, considering the lagging effect of industrial poverty alleviation on local economic growth, the next economic growth variable of the region where the sample local state-owned enterprises are located as the explained variable is taken to conduct an empirical test again. Third, considering the differences between administrative divisions and administrative levels in China, the sample of local state-owned enterprises located in municipalities directly under the Central Government (Beijing, Shanghai, Tianjin, Chongqing) is removed and a new empirical test is conducted. There is no substantial difference between the robustness test results and the benchmark regression test results above, which verifies the robustness of the regression results in this paper.

4.3.2. Endogeneity Test Results

Table 6. Endogeneity Test Results.

	1st Stage		2nd Stage
	(1)	(2)	(3)	(4)
	WIPA	DIPA	RESG	RESG
CSE	0.0144 *** (0.0016)	0.2736 *** (0.0205)
WIPA			0.0048 *** (0.0010)
DIPA				0.0003 *** (0.0001)
Controls	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes
Constant	−0.9842 ***	−12.2941 ***	0.0661 ***	0.0655 ***
Adj R2	0.0879	0.0792	0.0151	0.0132
F-statistics	32.4016 ***	29.0375 ***	55.0501 ***	49.3849 ***
J-statistics	—	—	0.2970	0.8240

Note: *** indicates that they have passed the significance test at the 1% confidence level.

shows the results of the endogenous test. In the 1st Stage regression results (1) and (2), the coefficient values of variable CSE are significantly positive, indicating that compared with public welfare local state-owned enterprises, competitive state-owned enterprises are more likely to carry out industrial poverty alleviation, and the industrial poverty alleviation efforts are also higher, which verifies the correlation of tool variables. In the 2nd Stage regression results (3) and (4), the WIPA and DIPA coefficients of the variables are also significantly positive, which indicates that after considering endogenous factors, the industrial poverty alleviation behavior of local state-owned enterprises can still promote local economic growth. In addition, the J statistic of the 2nd Stage regression results failed to pass the significance test of the conventional confidence level, which also verified the rationality of the selection of tool variables.

4.3.3. Intermediary Effect Test Results

Table 7. Intermediary Effect Test Results.

	(1)	(2)	(3)	(4)	(5)	(6)
	RESG	RESG	PGDP	PGDP	RESG	RESG
WIPA	0.0048 *** (0.0010)		0.0095 *** (0.0014)		0.0042 *** (0.0010)
DIPA		0.0003 *** (0.0001)		0.0006 *** (0.0001)		0.0003 *** (0.0001)
PGDP					0.0571 *** (0.0123)	0.0589 *** (0.0123)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.0654 ***	0.0648 ***	0.2241 ***	0.2221 ***	0.0526 ***	0.0517 ***
Adj R2	0.0153	0.013	0.4922	0.4896	0.0222	0.0208
F-statistics	60.7888 ***	54.458 ***	317.0343	313.7072 ***	76.6030 ***	72.4093 ***

Note: *** indicates that they have passed the significance test at the 1% confidence level.

shows the results of the intermediary effect test. The regression results (1) and (2) are the benchmark test results mentioned above. In the regression results (3) and (4), the WIPA and DIPA coefficients of variables are significantly positive, indicating that the greater the participation of local state-owned enterprises in industrial poverty alleviation and industrial poverty alleviation, the higher the proportion of primary industry in the GDP of the region. In the regression results (5) and (6), the PGDP coefficient of the variable is significantly positive, which verifies the existence of the intermediary effect, indicating that the increase in the proportion of primary industry in GDP effectively drives the sustainable economic growth of China’s poor areas. However, in the regression results (5) and (6), the WIPA and DIPA coefficients of the variables are also significantly positive, which indicates that the influence of the proportion of the primary industry in GDP only has a partial intermediary effect.

4.4. Heterogeneity Grouping Regression Test Results and Analysis

4.4.1. Grouping Test between Agriculture-Related Enterprises and Non-Agriculture-Related Enterprises

Whether in China or other countries, the main targets of poverty alleviation are poor rural areas, and the industrial base driven by industrial poverty alleviation is mostly agriculture and agriculture-related processing and manufacturing industries. This means that local state-owned enterprises engaged in agricultural production or related to agricultural production (such as rural tourism developed in rural China in recent years, processing and manufacturing industries based on rural expertise, etc.) may form a better driving force for industrial poverty alleviation in poor areas due to their own technological base, product base, market base, etc. Therefore, this paper further tests the samples belonging to agriculture related enterprises and non-agriculture related enterprises in groups. According to the industry classification standards of the CSRC, this paper classifies enterprises that belong to A, C13–16, C20, H, N as agriculture related enterprises, and others as non-agriculture related enterprises.

Table 8. Grouping test results for agricultural and non-agricultural enterprises.

	Sample of Agriculture-Related Enterprises		Sample of Non-Agriculture-Related Enterprises
	(1)	(2)	(3)	(4)
WIPA	0.0045 *** (0.0010)		0.038 ** (0.0017)
DIPA		0.0003 *** (0.0001)		0.0002 * (0.0001)
Controls	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes
Constant	0.0671 ***	0.0066 ***	0.0747 ***	0.0759 ***
Adj R2	0.0141	0.0126	0.0044	0.0410
F-statistics	15.1199 ***	14.6862 ***	12.717 ***	12.5814 ***

Note: ***, **, and * indicate that they have passed the significance test at the 1%, 5%, and 10% confidence levels, respectively.

shows the grouping test results of agricultural enterprises and non-agricultural enterprises. In the regression results (1) and (2), the WIPA and DIPA coefficient values of the variables are significantly positive, indicating that industrial poverty alleviation can effectively promote the economic growth of the regions where agricultural local state-owned enterprises are located. In the regression results (3) and (4), the WIPA and DIPA coefficient values of the variables are also significantly positive, indicating that industrial poverty alleviation can also promote the economic growth of the regions where non-agricultural local state-owned enterprises are located. However, comparing the test results of different types of enterprises, the regression results (1) and (2) have larger WIPA and DIPA coefficients and higher significance, which indicates that compared with non-agriculture related enterprises, the industry poverty alleviation of agriculture related enterprises has a stronger role in promoting economic growth in the region. It can be seen that the agricultural enterprises themselves have a better understanding of agriculture, rural areas and farmers, which is more conducive to linking their own business with the development of poor areas, and more convenient for enterprises to establish good industrial planning for the local area so as to effectively implement industrial poverty alleviation.

4.4.2. Grouping Test between High-Tech Enterprises and Non-High-Tech Enterprises

Enterprises with different scientific and technological attributes have differences in digital innovation ideas, tendencies, degrees, etc., which will make different local state-owned enterprises have different efforts in industrial poverty alleviation through digital innovation, and may also make the economic consequences of industrial poverty alleviation different. For example, high-tech enterprises contain more technology in their own industrial development, but non-high-tech enterprises also attach importance to the application of technology in recent years, and technology plays a stronger role in promoting their industrial development. Therefore, this paper further conducts a grouping test on the samples of high-tech enterprises and non-high-tech enterprises [50].

Table 9. Grouping test results for high-tech enterprises and non-high-tech enterprises.

	Sample of High-Tech Enterprises		Sample of Non-High-Tech Enterprises
	(1)	(2)	(3)	(4)
WIPA	0.0041 ** (0.0018)		0.0049 *** (0.0011)
DIPA		0.0002 * (0.0001)		0.0004 *** (0.0001)
Controls	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes
Constant	0.0712 ***	0.0698 ***	0.0629 ***	0.0626 ***
Adj R2	0.0103	0.0077	0.0140	0.0126
F-statistics	12.0060 ***	11.7447 ***	14.2689 ***	13.9315 ***

Note: ***, **, and * indicate that they have passed the significance test at the 1%, 5%, and 10% confidence levels, respectively.

shows the grouping test results for high-tech enterprises and non-high-tech enterprises. In regression results (1) and (2), the values of variables WIPA and DIPA coefficients are significantly positive, indicating that in the regions where high-tech local state-owned enterprises are located, the poverty alleviation behavior of enterprises can significantly promote regional economic growth, while in regression results (3) and (4), the values of variables WIPA and DIPA coefficients are also significantly positive, indicating that in regions where non-high-tech local state-owned enterprises are located, the poverty alleviation behavior of enterprises can also significantly promote economic growth. Comparing the regression results of different samples, the values of WIPA and DIPA coefficients in regression results (3) and (4) are higher and more significant. It can be seen that because non-high-tech local state-owned enterprises use less technology, technological innovation can play a stronger role in the primary stage, which is more conducive to industrial poverty alleviation, and thus plays a stronger role in regional economic growth.

4.4.3. Grouping Test between Follow-Up Poverty Alleviation Plan Enterprises and No-Follow-Up Poverty Alleviation Plan Enterprises

When enterprises publicly participate in poverty alleviation, they will simultaneously publish whether there is a follow-up poverty alleviation plan. On the one hand, the existence of follow-up poverty alleviation plans means that enterprises will continue to participate in poverty alleviation in addition to participating at present, which shows the sustainability of enterprises’ participation in poverty alleviation. On the other hand, releasing the follow-up poverty alleviation plan is also an effective means for enterprises to disclose information to market investors. Therefore, this paper further tests the samples of enterprises that have issued follow-up poverty alleviation plans and enterprises that have not issued follow-up poverty alleviation plans.

Table 10. Grouping test results for follow-up poverty alleviation plan enterprises and no-follow-up poverty alleviation plan enterprises.

	Sample of Follow-Up Poverty Alleviation Plan Enterprises		Sample of No-Follow-Up Poverty Alleviation Plan Enterprises
	(1)	(2)	(3)	(4)
WIPA	0.0062 *** (0.0024)		0.0057 *** (0.0013)
DIPA		0.0006 *** (0.0002)		0.0003 *** (0.0001)
Controls
Year
Industry
Constant	0.0686 ***	0.0687 ***	0.0574 ***	0.0585 ***
Adj R2	0.0113	0.0122	0.0182	0.0110
F-statistics	13.2046 ***	13.3708 ***	13.4711 ***	12.4831 ***

Note: *** indicates that they have passed the significance test at the 1% confidence level.

shows the group test results of enterprises with and without follow-up poverty alleviation plan. In the regression results (1) and (2), the coefficient values of variables WIPA and DIPA are significantly positive, indicating that in the regions where enterprises with subsequent poverty alleviation plans are located, the industrial poverty alleviation behavior of enterprises can significantly promote regional economic growth. In regression results (3) and (4), the coefficient values of variables WIPA and DIPA are also significantly positive. It shows that in the regions where the enterprises have no follow-up poverty alleviation plan, the poverty alleviation behavior of enterprises can also promote the economic growth of the regions. However, compared with the test results of different samples, the coefficient values of variables WIPA and DIPA in the sample of enterprises with subsequent poverty alleviation plan are significantly larger, indicating that the industrial poverty alleviation of enterprises with a subsequent poverty alleviation plan has a stronger role in promoting the economic growth of their regions. It can be seen that the release of follow-up poverty alleviation plans by enterprises is not only an attitude of their continuous participation in poverty alleviation, but also a reflection of their own real investment in industrial poverty alleviation, so as to exert a stronger role in promoting regional economic growth.

5. Conclusions

Chinese enterprises, especially the local state-owned enterprises, not only build a good industrial development plan for poor areas and solve the problems of poor regions in terms of public poverty, but also contribute to the sustainable development of the regional economy. Especially in the era of the digital economy, digital innovation not only provides good technical means for enterprises, but also helps enterprises to apply digital technology for industrial poverty alleviation. Therefore, taking the industrial poverty alleviation of local state-owned enterprises in China as the research object with a sample of the local state-owned enterprises listed in Shanghai and Shenzhen stock exchanges from 2016 to 2020, this paper empirically tests the impact of industrial poverty alleviation on the sustainable economic growth of the local region and the moderating effect of digital innovation. Our findings show that China’s local industrial poverty reduction behavior in state-owned enterprises can effectively promote regional economic growth. Moreover, if enterprises carry out digital innovation and have a higher degree of digital innovation, their industrial poverty alleviation behavior will have a stronger role in promoting regional economic growth. This conclusion still holds even after controlling for robustness and endogeneity factors. In addition, the study also finds that the mediation effect of the proportion of primary industry in GDP is statistically significant, and constitutes partial mediation. Heterogeneous analysis shows that the impact of industrial poverty alleviation on regional economic sustainable growth is more pronounced in agriculture-related local state-owned enterprises, non-high-tech local state-owned enterprises, and local state-owned enterprises with subsequent poverty alleviation plans

This study contributes to the current literature in the following aspects. Firstly, this paper focuses on the research of the behavior of Chinese enterprises participating in industrial poverty alleviation and the effect of different targeted poverty alleviation models on Chinese enterprises. Current studies have only focused on whether Chinese enterprises participate in poverty alleviation or how strongly they participate in poverty alleviation. Second, we further investigate the moderating effect of digital innovation. The existing literature only studied the digital economy in China’s rural economy; this paper explores the impact mechanism of digital innovation on sustainable regional growth in the process of targeted poverty alleviation. Third, this paper investigates the role of local state-owned enterprises on targeted poverty alleviation, which has been ignored in current literature. Local state-owned enterprises should become the main force to promote regional economic development because they are more familiar with their regions.

6. Discussion

In recent years, the Chinese government has actively promoted poverty alleviation, giving poverty alleviation a prominent position in its governance, and promoting common prosperity by eliminating poverty and improving people’s livelihood. This is not only the baseline task of the Chinese government, to build a moderately prosperous society in all respects, but also an essential requirement of the socialist political and economic system. According to the empirical evidence in this paper, as local state-owned enterprises are familiar with and make important contributions to regional economic development, their active participation in industrial poverty alleviation can effectively promote the economic growth of their regions. Therefore, combined with the research content and conclusion of this paper, this paper puts forward the following countermeasures and suggestions. First, we will give full importance to the role of local SOES in the follow-up work of poverty alleviation. Although the Chinese government has solved the problem of absolute poverty in China’s poor areas, it still faces the problem of relative poverty after absolute poverty, the problem of preventing the return to poverty after solving poverty, and the problem of effectively connecting poverty alleviation with rural revitalization. Therefore, local state-owned enterprises should continue to go into districts, counties and rural areas where poverty has been eliminated, and maintain continuous attention and support to these areas, so as to ensure sustainable economic development in these areas. For example, the model of resident village cadres already implemented by local state-owned enterprises can be sustained in the post-poverty period and the period of rural revitalization, so as to ensure sustained support for the regional economy. Second, it is necessary to strengthen local state-owned enterprises to participate in rural industrial construction. Since local state-owned enterprises are more familiar with their regions and have the ability and strength to participate in rural industrial construction, they should play a more prominent role in the process of rural revitalization by the Chinese government, so that local state-owned enterprises can participate more in rural construction. In the process of participating in industrial construction, local state-owned enterprises should give full importance to the cultivation of farmers’ abilities for self-development. Through the encouragement of local government, the management of local state-owned enterprises, and the participation of farmers and rural departments, they should promote the transformation of the concept of industrial development and industrial transformation in rural areas, so as to better promote regional economic development. Third, it is necessary to strengthen the application of digital technology in rural areas. The digital economy is the mainstream trend of economic development of all countries in the world. Digital technology is widely used not only in urban areas, but also in rural areas of China. On the one hand, in the process of agricultural mechanization production, the application of digital technology is beneficial to more efficient agricultural production; on the other hand, the development of digital technology can also be leveraged for the rural industry to provide a broader market space. For example, many rural areas in China are now actively promote the “e-commerce + industry” model, and have obtained very good results. Therefore, digital innovation and digital transformation in rural China should be actively promoted, and more digital technologies, such as big data and blockchain, should be used to provide new driving forces for the development of agricultural industry and rural economic growth.