Research on the Influence of Economic Development Quality on Regional Employment Quality: Evidence from the Provincial Panel Data in China

Abstract

Economic development plays an important role in regional employment quality. Taking provincial panel data in China during 2010–2019 as a sample, we constructed an employment quality and economic development quality indicator system. The system conducted a comprehensive evaluation model with entropy weight to measure provincial employment and economic development quality indexes. The system also applied fixed-effect panel regression models to research the influence of economic development quality on regional employment quality. The results showed that employment structure affected the regional employment quality index, followed by social security. There were regional differences in the employment quality index, among which the employment quality index in the eastern area was higher, and it was at its lowest in the western region. The sharing economy was a critical factor in economic development quality, which had significant regional heterogeneity. Economic growth had the most significant and positive influence on employment quality, followed by the sharing economy and economic structure. In addition, the consumer price index and the urban–rural gap substantially affected employment quality. However, education expenditure played a significant inhibitory role in employment quality. Economic development quality had the most significant effect on eastern employment quality, followed by the central region, and it had the least effect in the west.

1. Introduction

As economic development becomes the new normal in China, the economic growth rate has shifted from high speed to medium-high speed. The economic growth model has shifted from extensive growth to quality, efficiency, and intensive growth. According to modern economic theory, economic development generates labor demand through the activities of various sectors of the national economy and drives employment growth. With the continuous increase in China’s total employment and the continuous optimization of employment structure, employment quality has attracted more and more attention. The report to the 19th CPC National Congress pointed out that we should adhere to the strategy of giving top priority to employment and to positive employment policies, to achieve higher quality and full employment [1]. To follow the trend of the new scientific, technological, and industrial revolution, China has accelerated the strategic adjustment of economic structure and of economic transformation and upgrading. Great changes have happened on both the labor supply side and the demand side. The pressure created by this economic shift may transform into a massive employment development potential. Promoting economic growth is the fundamental premise for solving the employment problem. The academic community has been concerned about the relationship between economic development and employment [2]. However, there is no clear understanding of how economic development quality affects the quality and amount of regional employment. Therefore, against a background of the continuous upgrading and transformation of China’s population and economic structure, research on regional economic development quality and employment quality is of vital practical significance.

For related research on employment quality, the International Labor Organization (1999) was first to reflect employment quality, through an indicator system of decent work, including 6 dimensions and 11 indicators. In 1996, the Organisation for Economic Co-operation and Development (OECD) released the OECD employment outlook for the first time. In 2017, the OECD Framework [3] defined employment quality on the basis of earnings quality, labor market security, and the quality of the working environment. Hijzen and Menyhert [4] provided an in-depth discussion of the definition and measurement of labor market security. Anker et al. [5] used this indicator system to analyze UK employment quality. Given the long-term high unemployment rate and employment difficulties, the European Union (2001) selected employment and psychological satisfaction to reflect the quality of work from multiple factors. The employment quality indicator system of small enterprises, proposed by Morton [6], introduced employee safety and human resources. Kalleberg and Vaisey [7] applied the heterogeneity of job satisfaction to the measurement of employment quality. An indicator system designed by Liu [8] included 4 dimensions and 17 indicators. Erhel and Guergoat-Larivière [9] selected working hours, security, collective representative rights, and other indicators to construct an employment quality evaluation system. Lai et al. [10] applied the principal component analysis method to build an evaluation index system of employment quality from 6 dimensions, such as employment service and employment ability, including 50 three-level indicators: the results showed that, aside from provinces with rapid economic development levels, regional employment quality was low, and the difference in employment quality in the different areas was noticeable. Deguilhem and Frontenaud [11] selected six dimensions to measure employment quality, including employment status and synthetic remuneration. From the perspective of government employment public services in 2017, Kong et al. [12] constructed an employment quality indicator system with 6 dimensions and 22 indicators, including employability, employment status, and labor remuneration.

Economic development quality reflects a process of continuous coordination and optimization within the national or regional economic structure, and between the economy and society. János [13] analyzed the economic growth rate from the perspective of balanced growth. Webster and Muller [14] argued that economic structure, human resources, social environment, and resource endowment are all internal factors influencing the quality of urban economic development. In [15], a comprehensive evaluation system of the national economy consisted of 3 dimensions and 15 three-level indicators by the AHP method. Leng [16] measured economic development quality from seven aspects: effectiveness, adequacy, coordination, sustainability, innovation, stability, and sharing. Zanakis and Becerra-Fernandez [17] constructed a quality-indicator system of urban economic development from a non-economic dimension and analyzed the effect of non-economic factors on the quality of urban development. Zhu and Tan [18] constructed an economic quality system with three dimensions: economic growth, structure, and quality; based on the relevant data from Beijing during 2001–2009, it was proved to conform to Beijing’s economic development law. Xu [19] highlighted scientific and technological innovation, total social demand, and people’s livelihood improvement to construct an indicator system of economic development quality, and measured economic development quality in Hebei Province. Yao and Zhang [20] proposed an indicator system with 4 factors and 17 indicator layers. Wang and Wang [21] selected people’s livelihood, social development, and environmental quality to construct a regional economic development quality system and adopted the factor analysis method for measurement. Based on Beijing’s time-series data during 2009–2018, Wang and Yao [22] constructed an indicator system and concluded that: Beijing’s high-quality development achieved good results; the innovation drive, ecological civilization, and infrastructure gradually improved; and social people’s livelihood needed to be improved. Yang et al. [23] held that high-quality economic development should include the five concepts of new development and regarded that high-quality economic development had significant growth, and the unbalanced development in various regions was slowly decreasing.

Many scholars have also focused on the relationship between economic development and employment from different perspectives and have made great progress. Chen et al. [24] found that the traditional linear Austrian certainty law has failed in China, and the impact of economic expansion and economic contraction on unemployment has completely different asymmetric effects. Shahbaz et al. [25] stated a causal relationship between regional employment and economic development. Shi et al. [26] adopted the spatial autocorrelation method, concluding that economic development greatly impacted the employment of the primary industry and less so that of the tertiary sector. Li [27] applied the vector autoregression model to study the number of county employment in China and found that the economic growth of county areas has a significant inhibitory effect on county employment. Chen and Zhang [28] found that economic opening will upgrade the industrial employment structure in western China. Zhao et al. [29] applied the dual difference method to analyze the impact of the development strategy on employment and obtained that the development strategy in the Yangtze River Economic Belt can significantly promote employment. Wu and Li [30] regarded economic growth as promoting employment scale, and there was spatial heterogeneity. Xing and Yang [31] concluded that upgrading the industrial structure positively affects the employment labor force.

In conclusion, the existing literature has relatively mature evaluation systems for employment quality and economic development quality, which has laid a good foundation for this study. However, most of the existing literature focused on the impact of economic development on total employment or applied a single economic development indicator on employment quality. The effect of economic development quality on regional employment quality is still unclear. Therefore, this paper constructs an employment quality indicator system and an economic development quality indicator system and explores the influence of economic development quality on regional employment quality by fixed-effect panel regression models based on provincial panel data in China, during 2010–2019.

The rest is organized as the following. Section 2 gives the theoretical framework and research assumptions. Section 3 introduces the comprehensive evaluation model and panel regression model settings and presents the research variables for the relevant models. Section 4 analyses the research results, including multicollinearity, robustness, and heterogeneity tests. The last section provides the conclusions and some suggestions.

2. Theoretical Framework and Research Hypotheses

2.1. Economic Development and Employment Quality

Economic development is an essential factor in improving employment quality. China’s economy and society are moving towards a stage of high-quality development. Economic quality improvement in a region will attract a large number of employment talents, and economic growth and employment capacity will increase at the same time. In the new era, the employment structure will also change. Data and intelligence will promote economic development and create more jobs simultaneously. The comprehensive quality of employees puts forward higher requirements. In the high-quality development stage, labor supply and demand should be balanced and advance steadily. Enterprises can provide more labor demand, create more jobs, and generate more emerging industries. If there is a large supply of labor force in a region and the work technology is advanced, the investors will carry out corresponding locate activities. Then, employment quality in the region is high. Therefore, simultaneous labor supply and demand development will also improve employment quality. Wei and Gong [32] proved that technological progress and industrial structure upgrading were positively related to employment. Peng et al. [33] concluded that in the background of the new normal of economic development, excessive economic development would lead to fewer jobs and employment opportunities for college graduates. Zhang et al. [34] studied the relationship between employment and development in Sichuan and Chongqing and concluded that the coupling degree between the two regions was high and relatively balanced. Based on the above studies, this paper makes the following hypothesis:

Hypothesis 1 (H1).

Economic development quality has a significant influence on employment quality.

2.2. Regional Heterogeneity

There are apparent differences in the economic level, growth, and innovation in different regions of China. In addition, employment quality also has regional heterogeneity. The western region has a large area and relatively low economic development level, which is a vital and challenging area of shared prosperity; the eastern region is stable, leading the country and building the pilot demonstration area of socialism with Chinese characteristics. The Yangtze River Delta region is the highest level of economic development under the leader of Shanghai, the central region is the key objective of economic growth, and the central region is also constantly changing. Entering a new stage of development and high-quality integrated development will become a new trend. Many scholars have also conducted some studies in this regard. Chen and Zhang [28] showed that economic growth, opening up, and financial development had a significant role in promoting the employment structure in the western region. Hu and Zhou [35] concluded that the economic development in east China and south China had a positive effect on employment quality. Based on the above studies, this paper makes the following hypothesis:

Hypothesis 2 (H2).

Economic development quality has a different regional influence on employment quality.

2.3. Economic Growth

Economic growth is the rate of economic growth in a country or region. In 2019, China’s economy grew by 6.3%, the national economy improved, and the economic structure transformation advanced steadily. The development opportunities continued to upgrade, and significant breakthroughs were made in high-quality economic development. In China, Jiang and Liu [36] tested China’s data according to the Austrian law model and concluded that the law was unsuitable for China’s economic and employment growth model. Chen [37] concluded that employment did not grow with economic growth in China, and their correlation gradually weakened. Ma and Zhang [38] applied the Austrian affirmation law model to study the employment rate of college graduates and concluded that economic growth did not significantly impact graduates’ employment. Chen et al. [39] reported that the slow economic growth rate had a significant impact on employment, and when the economic growth rate declined, the employment rate would also decline. Based on the above studies, this paper makes the following hypothesis:

Hypothesis 3 (H3).

Economic growth positively influences employment quality.

2.4. Sharing Economy

The sharing economy is a new form derived from China’s high-quality development state. The sharing economy is designed to improve the utilization rate of idle goods and share the right to use idle resources with others through the Internet exchange platform. The sharing economy has become widespread with the rapid development of the digital economy. The sharing economy is characterized by openness, collaboration, and sharing. Everyone can share goods with others and maximize the value when the resources are idle, such as sharing transportation and sharing daily necessities. Yin [40] held that with the development of the sharing economy, the economic growth points are constantly increasing, providing more jobs for workers. Jiao [41] reported that the number of flexible employments in the sharing economy was too high, which may lead to employment instability. Wang and Zhang [42] showed that the sharing economy positively affected employment. Still, it also impacted the labor rights and interests sectors, requiring relevant government departments to introduce safeguard measures. Yang [43] reported that the sharing economy provided a more flexible employment platform for employees, and different workers could match their jobs according to their abilities. In this paper, the sharing economy is expressed by two indicators: disposable income per capita and consumer expenditure per capita. Based on the above studies, this paper makes the following hypothesis:

Hypothesis 4 (H4).

The sharing economy positively influences employment quality.

2.5. Economic Structure

Since the reform and opening-up, China’s economic structure has been continuously optimized, and the economic growth rate has continued to rise. The fifth Plenary Session of the 19th CPC Central Committee proposed to push a new “dual circulation”development pattern. Lin [44] held that the most critical thing in the economic structure is the industrial structure, and the growth of the industrial structure and the employment structure is unbalanced, which has some constraints on employment. Changes in the economic structure have an important impact on the employment structure [45]. Gan et al. [46] reported that it is necessary to rationalize the industrial structure and consider the factors of advanced industrial structure, so that industrial transformation can promote economic growth. Ma and Zhang [38] stated that the number of college-employed personnel in the secondary industry was higher than that in the primary industry, the tertiary industry attracted the most college graduates, and college graduates were the most employed in the social service industry. Yang [47] concluded that from 1994 to 2019, China’s economic structure was slowly unbalanced to equilibrium. Based on the above studies, this paper makes the following hypothesis:

Hypothesis 5 (H5).

Economic structure influences employment quality positively.

3. Research Design

3.1. Comprehensive Evaluation Model Settings

This paper applied a comprehensive evaluation model based on entropy weight to calculate the employment quality index and economic development quality index during the period 2010–2019. The raw data are standardized to facilitate the calculation and eliminate the dimension influence:

$$y_{ij}^{’}=\left\{\begin{array}{l}\frac{y_{ij}-\min (y_{ij})}{\max (y_{ij})-\min (y_{ij})},\hspace{1em}{y}_{ij} \mathrm{is} \mathrm{a} \mathrm{forward} \mathrm{indicator}\\ \frac{\max (y_{ij})-y_{ij}}{\max (y_{ij})-\min (y_{ij})},\hspace{1em}{y}_{ij} \mathrm{is} \mathrm{a} \mathrm{reverse} \mathrm{indicator} \end{array}\right.$$

(1)

where $y_{ij}$ represents the raw data of the ith indicator in the jth province ($i=1\dots n,j=1\dots m$).

Then, calculate the p-value by the proportion of the ith indicator in province j.

$$p_{ij}=\frac{{y^{\prime }}_{ij}}{{\displaystyle \sum _{i=1}^n{y^{\prime }}_{ij}}}$$

(2)

The following formula calculates the entropy value $e_i$ of each indicator:

$$e_i=-\frac{1}{\ln (n)}{\displaystyle \sum _{j=1}^m{p}_{ij}\ln (p_{ij})}$$

(3)

The weight of the corresponding indicator is given by the following.

$$w_i=\frac{1-e_i}{{\displaystyle \sum _{j=1}^n(1-e_j)}}$$

(4)

Finally, the comprehensive evaluation quality index of the jth province is calculated by the weighted sum:

$$inde{x}_j={\displaystyle \sum _{i=1}^n{w}_j{y}_{ij}^{’}}$$

(5)

3.2. Panel Regression Model Settings

To explore the influence of economic development quality on regional employment quality, we try to introduce control variables one by one. The specific model is as follows:

$$EQ{I}_{ik}={\beta }_1+{\beta }_{11}EDQ{I}_{ik}+{\epsilon }_{ik}$$

(6)

$$EQ{I}_{ik}={\beta }_2+{\beta }_{21}EDQ{I}_{ik}+{\beta }_{22}CP{I}_{ik}+{\epsilon }_{ik}$$

(7)

$$EQ{I}_{ik}={\beta }_3+{\beta }_{31}EDQ{I}_{ik}+{\beta }_{33}UR{G}_{ik}+{\epsilon }_{ik}$$

(8)

$$EQ{I}_{ik}={\beta }_4+{\beta }_{41}EDQ{I}_{ik}+{\beta }_{42}CP{I}_{ik}+{\beta }_{43}UR{G}_{ik}+{\epsilon }_{ik}$$

(9)

$$EQ{I}_{ik}={\beta }_5+{\beta }_{51}EDQ{I}_{ik}+{\beta }_{52}CP{I}_{ik}+{\beta }_{54}E{I}_{ik}+{\epsilon }_{ik}$$

(10)

$$EQ{I}_{ik}={\beta }_6+{\beta }_{61}EDQ{I}_{ik}+{\beta }_{62}CP{I}_{ik}+{\beta }_{63}UR{G}_{ik}+{\beta }_{64}E{I}_{ik}+{\epsilon }_{ik}$$

(11)

$$EQ{I}_{ik}={\beta }_7+{\beta }_{71}EDQ{I}_{ik}+{\beta }_{72}CP{I}_{ik}+{\beta }_{73}UR{G}_{ik}+{\beta }_{74}E{I}_{ik}+{\epsilon }_{ik}$$

(12)

Formulas (6)–(12) correspond to Models (1)–(7), respectively. $EQ{I}_{ik}$ means the employment quality index in the ith province in year k, and $EDQ{I}_{ik}$ represents the economic development quality index in year k. $CP{I}_{ik}$, $UR{G}_{ik}$, and $E{I}_{ik}$ are, respectively, the consumer price index, urban–rural gap and education expenditure in the ith province in year k. ${\beta }_i$ and ${\beta }_{ik}$ are constant terms, and ${\epsilon }_{ik}$ means the random error term.

3.3. Variable Selection

3.3.1. Explained Variable

The explained variable is the Employment Quality Index (EQI). Employment quality should include the social security, employment structure, and work availability of workers, in addition to the degree of wages obtained by combining workers with means of production. According to the available literature, there are no definite and unified indicators of employment quality. According to EQI indicators [10,12,48,49], an indicator system of employment quality is constructed from labor wage, social security, employment structure, and work availability, as shown in

Table 1. Indicator system of employment quality.

Target Layer	System Layer	Indicator Layer	Unit	Attribute
Employment Quality	Labor Wages	Average employee salary (WAGE)	Yuan	+
		Wage growth rate (DWAGE)	%	+
	Social Security	Social insurance ratio (INSURANCE)	%	+
		Ratio of the minimum wage to the average wage (MINWAGE)	%	+
	Employment Structure	The proportion of the urban employed population (TOWN)	%	+
		The proportion of the employed population in the tertiary industry (THIRD-INDUSTRY)	%	+
		The manufacturing employment rate (MANUFACTURE)	%	+
	Work Availability	Labor participation rate (LABOR)	%	+
		Urban registered unemployment rate (UNEMP)	%	−

. The comprehensive evaluation model based on the entropy method is applied to calculate the weight of the employment quality index in various provinces and cities.

Labor wages are the most important concern of employees, mainly measured by the average employee salary and wage growth rate. Social security includes social insurance ratio and minimum income guarantee, measured by the proportion of employees participating in the social insurance and the ratio of the minimum wage to the average wage, respectively. With China’s urbanization process, many rural surplus labor forces have turned to nonfarm industries under the support of the overall urban and rural employment policy. Additionally, the number of urban and rural employment has continuously increased. Additionally, to become a manufacturing power, it must rely on the real economy, revitalize the advanced manufacturing industry and improve the employment rate of the manufacturing industry. Therefore, the employment structure in this paper mainly includes urban and rural structure, industrial structure and manufacturing employment, which are measured by the proportion of the urban employed population, the ratio of the employed population in the tertiary industry and the manufacturing employment rate, respectively. Work availability includes labor participation rate and urban registered unemployment rate. Among them, the urban registered unemployment rate is a major indicator reflecting the Chinese employment situation. Although the urban survey unemployment rate is more accurate, considering the availability of provincial data, this paper uses the urban registered unemployment rate.

3.3.2. Explanatory Variables

The core explanatory variable is the Economic Development Quality Index (EDQI). Economic development quality is a multidimensional concept that refers to the comprehensive evaluation of the economic situation of a region. Reffing to the construction of EDQI indicators [50,51,52], a quality index system of China’s economic development is constructed from economic growth, sharing economy, and economic structure. The details are shown in

Table 2. Indicator system of economic development quality.

Target Layer	System Layer	Indicator Layer	Unit	Attribute
Economic Development Quality	Economic Growth	Gross regional product (GDP)	108 Yuan	+
		GDP per capita (RGDP)	Yuan/Person	+
	Sharing Economy	Per capita disposable income (PCDI)	Yuan	+
		Consumer expenditure per capita (CPP)	Yuan	+
	Economic Structure	Share of primary industry in GDP (FGDP)	%	−
		Share of the tertiary industry in GDP (SGDP)	%	+
		Share of fiscal expenditure in GDP (MGDP)	%	+

.

3.3.3. Control Variables

Referring to the research of Xu et al. [53], Xu and Yang [54], the consumer price index (CPI), urban–rural gap (GUR), and education expenditure (EE) are selected as control variables. The consumer price index is measured by the natural logarithm of the consumer price index. The urban–rural gap (URG) is measured by the natural logarithm of the proportion of the urban population to the rural population. Additionally, the education expenditure (EE) is calculated by the natural logarithm of the ratio of education expenditure to total fiscal expenditure. All the variables are shown in

Table 3. Variables in regress models.

Type	Variables	Specific Explanation
Explained Variable	Employment Quality Index (EQI)	It is measured by the comprehensive calculation of the above evaluation index system of employment quality.
Explanatory Variables	Economic Development Quality Index (EDQI)	It is measured according to the comprehensive calculation of the above economic development quality index system.
Control Variables	Consumer Price Index (CPI)	It is measured by the natural logarithm of the consumer price index.
	Urban–rural Gap (URG)	It is measured by the natural logarithm of the proportion of the urban population to the rural population.
	Education Expenditure (EE)	It is measured by the natural logarithm of the proportion of education expenditure in the total fiscal expenditure.

.

3.4. Data Sources

A panel data set involving 31 provinces and cities in China is selected to keep data continuity and consistency. The provincial data of indicators related to the economic quality index and control variables derive from the 2011–2020 China Statistical Yearbook. The data on related employment indicators draw from the China population and Employment Statistical Yearbook and the China Labor Statistical Yearbook.

4. Empirical Results and Discussions

4.1. Employment Quality Index

According to related employment data, the comprehensive evaluation model with entropy is applied to calculate the entropy value and weights of employment quality indicators from 2010 to 2019. The specific results are shown in

Table 4. Entropy value and weight of each indicator of employment quality.

System Layer	Weight	Indicator Layer	Entropy Value	Weight
Labor Wages	0.20	WAGE	0.96	0.16
		DWAGE	0.99	0.04
Social Security	0.24	INSURANCE	0.95	0.19
		MINWAGE	0.99	0.05
Employment Structure	0.42	TOWN	0.95	0.17
		THIRD-INDUSTRY	0.98	0.08
		MANUFACTURE	0.95	0.17
Work Availability	0.14	LABOR	0.98	0.07
		UNEMP	0.98	0.07

. According to the weights of indicators, it can be concluded that the employment structure has the greatest impact on the employment quality index, and its weight is 0.42. It indicates that the proportion of the urban employment population, the ratio of the employed people in the tertiary industry, and the manufacturing employment population greatly impacts the employment quality. Secondly, the social security and labor wages accounted for 0.24 and 0.20, respectively, indicating that occupations with higher average wage and social security, are more attractive employment. The weight of employment opportunities is 0.14 and is the lowest. For the weights of the three-level indicators, the average salary o the employees, the average social security participation ratio, the proportion of urban employment population, and the proportion of manufacturing employment are heavily weighted, with 0.16, 0.19, 0.17, and 0.17, respectively.

According to the comprehensive evaluation model based on the entropy method, the employment quality indexes of 31 provinces from 2010 to 2019 are calculated. The specific results are shown in

Table 5. Provincial employment quality index during 2010–2019.

Regions	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
Beijing	60.79	62.77	66.79	67.86	69.63	62.38	72.89	75.65	77.94	81.05
Tianjin	39.47	45.69	48.71	47.53	48.19	39.13	48.24	47.93	47.95	50.10
Hebei	20.06	21.70	23.13	23.67	25.06	15.35	25.23	25.08	25.65	26.76
Shanxi	26.06	27.35	29.41	29.62	30.07	20.99	29.33	30.66	31.33	32.46
Nei Mongol	23.32	24.28	25.18	26.77	28.42	19.30	29.29	29.68	30.11	31.27
Liaoning	31.53	33.86	34.71	36.75	37.44	27.64	35.64	35.85	36.34	37.29
Jilin	25.70	27.79	28.28	31.45	32.27	23.84	33.20	34.32	34.07	34.48
Heilongjiang	26.15	26.93	27.73	27.75	28.64	19.49	28.89	29.67	30.06	30.74
Shanghai	51.09	57.12	61.22	60.30	62.08	53.78	65.34	66.76	68.27	70.87
Jiangsu	29.89	31.73	33.10	42.65	45.28	35.25	45.22	46.17	46.71	48.14
Zhejiang	39.35	41.95	44.03	45.10	46.44	35.01	46.15	47.69	48.24	50.23
Anhui	19.78	22.47	23.40	24.50	25.50	13.43	26.50	28.47	30.74	31.33
Fujian	33.75	37.14	38.63	37.63	38.41	26.41	39.68	40.47	41.25	40.73
Jiangxi	20.28	21.92	23.55	24.80	27.44	17.85	28.94	28.91	28.60	29.84
Shandong	26.20	28.43	29.95	32.19	33.27	21.88	33.27	34.01	34.41	35.48
Henan	19.74	21.95	23.13	26.43	28.33	15.90	29.74	30.84	29.56	29.79
Hubei	21.87	24.00	24.94	26.97	29.38	18.95	30.68	31.52	31.53	32.48
Hunan	21.55	22.86	23.37	23.49	24.80	14.55	24.79	25.74	26.63	28.47
Guangdong	33.72	36.29	37.62	48.64	50.71	41.61	51.06	51.52	52.44	53.65
Guangxi	18.02	19.76	20.55	21.79	22.76	13.91	24.48	25.31	26.00	27.07
Hainan	26.32	27.72	28.69	31.19	31.56	22.18	31.94	33.04	34.88	37.06
Chongqing	23.99	28.00	29.83	31.90	34.16	24.11	35.01	35.97	37.31	38.34
Sichuan	21.37	23.35	24.22	27.38	27.57	17.02	27.67	28.94	31.21	32.71
Guizhou	17.38	19.79	21.08	22.10	23.18	13.88	24.84	25.89	27.14	28.42
Yunnan	19.42	20.42	21.35	22.85	23.71	13.59	25.61	26.55	27.14	28.15
Xizang	20.82	22.38	24.85	27.50	29.10	26.04	32.69	36.01	37.15	39.78
Shaanxi	22.64	24.43	24.24	27.04	28.34	23.19	30.88	32.38	32.34	34.03
Gansu	18.90	19.82	21.25	22.83	24.29	14.97	24.89	26.55	27.57	28.21
Qinghai	23.00	24.91	26.21	27.37	28.46	19.17	29.57	31.17	32.03	33.44
Ningxia	23.89	25.78	27.22	28.16	29.67	20.63	30.93	31.80	32.94	34.45
Xinjiang	30.07	32.24	32.84	33.44	34.45	25.40	34.80	35.63	36.52	38.34

and Figure 1.

From Figure 1, the provincial employment quality indexes from 2010 to 2019 were the lowest in 2012 and the highest in 2019. The eastern region has high employment quality indexes, such as Beijing, Shanghai, and Guangdong Province. In 2010, the employment quality index of Beijing was the highest, reaching 60.79. In 2019, Beijing still ranked first, reaching 81.1, up 4.1% from 2018. As the capital, Beijing is the political and cultural center in China. It has convenient transportation and many employment opportunities to attract plenty of high-level talents. The employment quality indexes of the central and western regions, such as Guangxi, Guizhou, and Yunnan, are relatively low. The employment quality index of Guizhou was the lowest, reaching 17.4 in 2010. Guizhou achieved transcendence in 2019, and the employment quality index reached 28.4, ranking 28th. In contrast, the employment quality index of Hebei was the lowest in 2019, only 17.4.

From a macro perspective, the employment quality index of some developed regions in China, such as Beijing and Shanghai, is higher than that of other regions. At the micro level, the number of employed people in Beijing increased from 10.316 million in 2010 to 12.73 million in 2019, and the overall employment rate of college graduates in Beijing reached 95.90% in 2019. The number of employed people in Shanghai increased from 10.908 million in 2010 to 13.762 million in 2019. Moreover, the GDP of Shanghai has sequencely exceeded CNY 3 trillion and CNY 4 trillion, increasing from CNY 1.79 trillion in 2010 to CNY 4.32 trillion in 2021. It has become the fourth largest city in the world.

Table 6. Statistics of employment quality index.

Year	Mean	Median	SD	Min	Max
2019	37.91	34	12.45	26.8	81.1
2018	36.57	32.3	12.01	25.6	77.9
2017	35.81	31.8	11.78	25.1	75.6
2016	34.75	30.9	11.6	24.5	72.9
2015	34.41	30.2	11.71	13.4	62.4
2014	33.83	29.4	11.34	22.8	69.6
2013	32.51	27.8	11.21	21.8	67.9
2012	30.62	27.2	11.18	20.6	66.8
2011	29.18	25.8	10.46	19.8	62.8
2010	26.97	23.9	9.751	17.4	60.8

shows that the employment quality indexes from 2010 to 2019 increased yearly. Additionally, the maximum value is 37.91, and the minimum value is 26.97. The standard deviation was 23.9 in 2010. The maximum was three times the minimum in 2019 and 3.5 times the minimum in 2010, indicating a significant difference in the employment quality index in different regions.

4.2. Economic Development Quality Index

According to the economic data of provinces and cities from 2010 to 2019, the entropy evaluation model based on the entropy weight calculates the entropy values and weights of the economic development quality indicators. The specific results are shown in

Table 7. Entropy value and weight of economic development quality.

System Layer	Weight	Indicator Layer	Weight	Entropy Value
Economic Growth	0.25	DGDP	0.06	0.94
		RGDP	0.20	0.96
Sharing Economy	0.40	PCDI	0.21	0.96
		CPP	0.19	0.96
Economic Structure	0.35	FGDP	0.06	0.99
		SGDP	0.14	0.97
		MGDP	0.15	0.97

.

In Table 7, among the weights of the secondary indicators, the sharing economy has the greatest impact on economic development quality, with a weight of 0.4. Secondly, the proportion of the economic structure is 0.35, indicating that the ratio of the added value of various industries in GDP also greatly affects economic development quality. The weight of economic growth is the lowest (0.25). The weight of per capita disposable income is the highest (0.21), and those of gross regional product and share of primary industry in GDP are the lowest (0.06), which has little impact on economic development quality.

The economic development quality index from 2010 to 2019 based on the above weights is shown in

Table 8. Economic development quality index.

Regions	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
Beijing	45.18	48.24	49.41	52.98	54.46	59.21	63.03	67.00	71.27	73.07
Tianjin	33.97	37.20	36.77	39.26	40.70	41.73	46.97	48.32	48.97	51.45
Hebei	19.25	21.75	18.68	18.88	19.70	22.95	25.87	27.72	29.72	32.27
Shanxi	24.52	24.95	20.88	20.87	21.48	23.97	26.26	35.36	32.64	32.64
Nei Mongol	27.48	30.80	30.81	31.49	32.07	34.17	35.62	37.18	38.74	39.45
Liaoning	21.45	27.94	26.95	28.93	28.51	27.27	29.72	33.41	35.86	36.23
Jilin	26.49	29.77	27.57	27.83	27.78	27.49	31.8	33.40	33.73	35.32
Heilongjiang	22.47	26.15	23.80	23.32	22.51	23.83	28.31	31.35	32.77	34.85
Shanghai	42.23	44.79	45.07	49.26	52.53	56.18	62.63	65.76	69.92	71.12
Jiangsu	26.58	29.01	27.86	30.44	32.02	34.84	36.45	39.89	41.48	42.98
Zhejiang	29.31	30.46	28.75	31.43	32.95	37.09	39.19	42.84	46.14	48.51
Anhui	22.52	24.81	22.32	23.33	23.48	24.30	27.99	31.01	33.52	33.24
Fujian	23.78	25.73	24.95	26.51	27.91	29.11	32.37	36.56	38.66	39.21
Jiangxi	22.25	24.69	21.10	23.25	24.09	25.55	27.92	29.96	33.33	34.02
Shandong	18.75	21.24	21.16	23.23	23.71	26.71	27.44	29.68	31.22	33.36
Henan	17.21	18.61	17.97	19.18	20.62	21.56	23.72	26.62	29.01	29.22
Hubei	22.44	24.41	21.98	23.89	25.40	26.76	29.33	31.82	34.78	35.10
Hunan	22.66	24.44	22.24	23.70	24.67	26.92	28.19	30.97	32.20	34.81
Guangdong	24.93	26.95	25.32	28.39	29.99	34.15	36.61	39.74	40.42	42.99
Guangxi	22.39	24.15	21.74	23.06	23.68	25.90	27.28	29.61	31.07	31.84
Hainan	28.83	30.37	28.82	30.11	30.86	31.78	33.94	35.15	38.22	40.13
Chongqing	26.29	32.29	28.91	28.35	29.49	30.65	33.20	34.68	34.65	37.40
Sichuan	24.03	25.00	23.00	23.59	24.38	24.66	28.20	32.16	33.73	33.15
Guizhou	26.64	33.00	32.42	31.91	31.34	31.65	31.09	33.48	33.54	34.72
Yunnan	24.29	28.79	27.71	28.5	26.93	27.07	29.42	32.87	33.81	34.54
Xizang	14.66	24.81	26.08	25.11	26.50	29.57	32.55	30.05	32.32	32.49
Shanxi	24.94	27.32	25.26	24.96	24.96	24.44	26.78	31.07	32.26	32.45
Gansu	28.01	30.24	27.46	28.73	28.86	29.87	33.55	34.88	38.95	38.42
Qinghai	44.86	47.75	47.65	47.25	47.29	50.48	49.83	46.92	48.37	49.24
Ningxia	32.26	33.78	31.44	31.90	32.19	34.27	37.78	41.20	39.46	38.92
Xinjiang	28.73	30.22	29.23	31.18	30.78	30.35	33.81	40.37	40.57	38.18

.

In Figure 2, the economic development quality from 2010 to 2019 shows an increasing trend year by year. Each province’s economic development quality index was the highest in 2019 and lowest in 2010. The economic development quality indexes are high in the eastern regions, such as Beijing, Shanghai, and Guangdong, ranking in the top three. Additionally, they are low in the central and western regions, such as Tibet, Guangxi, and Jiangxi. Among them, Beijing’s economic development quality index increased from 45.18 in 2010 to 73.03 in 2019. Additionally, the economic development quality index of Tibet is relatively low, rising from 14.66 in 2010 to 32.49 in 2019.

In

Table 9. Statistic of economic development quality index.

Year	Mean	Median	SD	Min	Max
2019	39.40	35.32	10.04	29.22	73.07
2018	38.43	34.65	9.81	29.01	71.27
2017	36.81	33.48	9.32	26.62	67.00
2016	34.09	31.80	9.88	23.72	63.03
2015	31.43	29.11	9.06	21.56	59.21
2014	29.74	27.91	8.35	19.70	54.46
2013	29.06	35.21	8.06	18.88	52.98
2012	27.85	26.95	7.62	17.97	49.41
2011	29.34	27.94	6.94	18.61	48.24
2010	26.43	24.94	7.03	14.66	45.18

, the average index of economic development quality reached the highest at 39.40, in 2019, and the lowest at 26.43, in 2010. The average economic development quality index declined in 2011 and grew gradually. The median value was the highest at 35.32, in 2019, and the lowest at 26.43, in 2010. The minimum standard deviation is 11.7, and the maximum standard deviation is 15.3. The maximum of the economic development quality index was 2.5 times the minimum in 2019 and was three times the minimum in 2010. It indicates great differences in economic development quality in different regions.

The economic development quality system is constructed from three dimensions and seven indicators and then studies the impact of economic development quality on employment quality. Intuitively, the higher the level of economic development in a region, the higher the quality of employment, the more high-quality talents it will attract, and the richer the salary will be. The predicted economic development positively affects regional employment, but whether the actual results are the same as the expected results should be further judged according to the regression results.

4.3. Descriptive Statistics

Table 10. Descriptive statistics of variables.

Variables	N	Mean	Median	SD	Min	Max
EQI	310	32.26	29.39	11.96	13.43	81.05
EDQI	310	32.26	30.41	9.700	14.66	73.06
CPI	310	4.631	4.629	0.0120	4.611	4.666
URG	310	0.313	0.224	0.643	−1.228	2.152
EE	310	−2.933	−3.002	0.349	−3.521	−1.693

shows the descriptive statistics of all variables in the model. The maximum employment quality index is 81.05, the minimum is 13.43, and the mean is 32.26, among which the maximum is six times the minimum. It indicates that there are differences in employment quality in different regions. Only one-third of the 31 provinces exceeded the average employment quality index, so the few provinces had a higher employment quality index. For the economic development quality system, the maximum economic development quality index is 73.06, the minimum is 14.66, the average of 32.26, and the maximum is five times the minimum. The economic development quality index varies significantly in different regions. Among 31 provinces, 11 have higher economic development quality than the average value, so few have a higher employment quality index. Regarding the control variables, the average consumer price index is 4.631, with a small gap between regions. The average price index exceeds the urban–rural gap and education expenditure. The average and standard deviation of urban-rural gap are 0.313 and 0.643. The average and standard deviation of education expenditure are −2.933 and 0.349.

4.4. Multicollinearity Test Results

A correlation test is performed between the variables. The results are shown in

Table 11. Correlation analysis for each variable.

Variables	EQI	EDQI	CPI	URG	EE
EQI	1
EDQI	0.772 ***	1
CPI	−0.0620	−0.096 *	1
URG	0.847 ***	0.715 ***	−0.138 **	1
EE	−0.347 ***	−0.000	−0.00900	−0.541 ***	1

Note: * p < 0.1, ** p < 0.05, and *** p < 0.01.

. The employment quality index (EQI) is strongly correlated with the economic development quality index (EDQI), and the coefficient is significant at the 1% level. Through the correlation analysis, economic development quality and employment quality are significantly correlated. For control variables, the consumer price index (CPI) and education expenditure (EE) have negative correlations with the employment quality index (EQI). The correlation coefficient of employment quality and an urban–rural gap is positive. The correlation coefficient is large, which means there may be multicollinearity.

Multicollinearity means there is no linear relationship between each variable. If there is multicollinearity between variables, the variance and covariance will be larger in parameter estimation; the parameter confidence interval will be larger; the hypothesis test will be misjudged; and the validity of F and t-tests will be failed. Therefore, multicollinearity tests for the variables are necessary. The main methods to test multicollinearity include the correlation coefficient test, variance expansion factor method, etc. In the correlation analysis, we found that the correlation coefficient of employment quality and education expenditure was greater than 0.8, and there may be multicollinearity. To verify whether they have multicollinearity, we tested it again with the variance expansion factor method. The test results are shown in

Table 12. Multicollinearity test.

Variables	VIF	1/VIF
GUR	5.210	0.192
Economy	3.620	0.276
Edu	2.530	0.396
CPI	1.040	0.963
VIF mean	3.100

.

If the VIF < 10, there is no multicollinearity between the variables. As shown in Table 12, all values of VIF are less than 10, so there is no multicollinearity between the variables. Regression analysis of the data is possible.

4.5. Analysis of the Regression Results

4.5.1. Unit-Root Test and Cointegration Test

To avoid the phenomenon of pseudo-regression in the model, we performed the LLC test as the unit root test on the data. The results of the LLC test with Stata16 are shown in

Table 13. LLC test results.

Statistic	EDQI	CPI	GUR	Edu
LLC	−10.3253 ***	−12.9083 ***	−15.4224 ***	−6.1905 ***
p-Value	0.0000	0.0000	0.0000	0.0000
Smoothness	Data smooth	Data smooth	Data smooth	Data smooth

Note: *** p < 0.01.

. The p-value of each variable is less than 0.05. It shows that all the data are stationary.

The cointegration test refers to judging whether there is a pseudo-regression phenomenon in the causal relationship between them, that is, to test whether the variable relationship is stable. The results of the Pedroni and Westerlund tests are listed in

Table 14. Cointegration test results.

Cointegration Test	Specific Test	Statistic	p-Value
Pedroni	Modified Phillips-Perron t	8.2808	0.0000
	Phillips-Perron t	−23.1290	0.0000
	Augmented Dickey–Fuller t	−14.7554	0.0000
Westerlund	Variance ratio	2.0940	0.0181

. According to the p-values, the null hypothesis is rejected, and there is a cointegration relationship.

4.5.2. Analysis of the Model Regression Results

The F test and Hausman test are performed for the panel data. The statistic value of the F test is 70.28, and the p-value is 0.0000. The null hypothesis is rejected, and a fixed-effect model should be chosen between mixed and fixed effects. Hausman test is performed on the sample data, and the null hypothesis is that a random effect model should be selected. The results show that the p-value is 0.6955, greater than 0.05. Then, the null hypothesis is accepted. The fixed and random effect models will converge to the actual parameter values. The fixed-effect results are generally robust. Therefore, a fixed-effect model is suitable for regression.

Control variables are introduced one by one to improve the regression results’ robustness and to analyze the impact of economic development on regional employment. The specific results are shown in

Table 15. Regression results for the fixed effects.

Variables	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)	Model (6)	Model (7)
EQI	0.616 *** (0.041)	0.610 *** (0.043)	0.435 *** (0.057)	0.635 *** (0.043)	0.416 *** (0.057)	0.444 *** (0.056)	0.427 *** (0.057)
CPI		−8.279 (18.289)			41.549 ** (20.356)		37.760 * (20.226)
URG			7.283 *** (1.651)		9.237 *** (1.901)	8.514 *** (1.701)	10.224 *** (1.925)
EE				−3.295 (2.672)		−6.985 *** (2.667)	−6.610 ** (2.663)
_cons	12.388 *** (1.331)	50.918 (85.126)	15.938 *** (1.519)	2.122 (8.431)	−176.472 * (94.282)	−5.224 (8.219)	−178.953 * (93.418)
N	310.000	310.000	310.000	310.000	310.000	310.000	310.000
R2	0.451	0.451	0.487	0.454	0.494	0.499	0.505
R2_a	0.389	0.388	0.427	0.391	0.434	0.439	0.444

Note: Standard errors in parentheses, and * p < 0.1, ** p < 0.05, *** p < 0.01.

.

Table 15 shows regression models performed by gradually adding control variables. In Model (1), no control variables are introduced, and only the quality of employment and economic development quality are in the model. The regression coefficient of EDQI is 0.616, significant at the 1% level, indicating that economic development plays a positive role in regional employment. Each additional unit of the economic development quality index will increase the employment quality index by 0.616. In Model (2), by introducing CPI as the control variable, the regression coefficient of EDQI changes to 0.610, and the regression coefficient of CPI is −8.279, but not significant. Model (3) replaces the control variable with the urban–rural gap URG. The coefficient of EDQI is 0.435, and the coefficient of URG is 7.283, both significant at 1%. It shows that the gap between urban and rural areas and economic development quality positively affects employment quality. The control variable in Model (4) is education expenditure (EE). The coefficient of EDQI is 0.635, significant at the 1% level. Additionally, the coefficient of URG is −3.295, but not significant. In Model (5), introducing two control variables with CPI and URG, their coefficients are positive and significant at the 1% level. It further shows that with URG and CPI, people’s wages are higher, and employment quality is better. The economic development quality, the price index, and the urban–rural gap can promote employment quality. In Model (6), EDQI, URG, and EE coefficients are all significant at the 1% level. The increase in the economic development quality and the urban–rural gap will increase employment quality. However, the employment quality will decrease with the increase in education expenditure. In Model (7), after introducing all control variables, the regression coefficients of EDQI, CPI, and URG are positive and significant, indicating these factors will promote employment quality. However, the regression coefficient of EE is negative and significant. The increased education expenditure will curb employment quality.

Overall, in the above regression models, the coefficients of the core explanatory variable EDQI are always positive and significant at the 1% level. It shows that EDQI plays a positive role in promoting the quality of employment. In Model (1)~Model (7), EDQI has different effects on employment quality. Additionally, it has the largest impact in Model (4). For each 1% increment of EDQI, the employment quality will increase by 0.635%. For the price index, the regression coefficient of Model (5) reaches 41.549, which is significant at the 5% level. Additionally, the regression coefficient of CPI in Model (5) and Model (7) is positive and significant, indicating that the improvement of CPI can increase employment quality. As for the urban–rural gap, it has the greatest impact on the employment quality in Model (7). For every 1% increment of URG, the employment quality will increase by 10.224%. In each model, the regression coefficient of URG was significant at the 1% level, indicating that the urban–rural gap positively affected employment quality. For education expenditure, the regression coefficient of EE fails the test in Model (4) but is significant at the 1% level in Model (6) and the 5% level in Model (7). The regression coefficients of EE in the three models are negative, indicating that education expenditure also has an inhibitory effect on the quality of employment.

To sum up, the quality of economic development, price index, and urban–rural gap all promote employment quality positively. Still, education expenditure has a restraining effect on the quality of employment.

4.5.3. Influence of Different Dimensions on Employment Quality

To further analyze the impact of each dimension of EDQI on regional employment, the economic growth, shared economy, and economic structure are taken as the explained variables. The regression results are shown in

Table 16. Regression results for each dimension.

Variables	Model (1)	Model (2)	Model (3)	Model (4)
Economic Growth	0.316 ***			0.243 ***
	(0.035)			(0.081)
Sharing Economy		0.214 ***		0.078
		(0.027)		(0.066)
Economic Structure			0.079	−0.138 *
			(0.074)	(0.079)
CPI	32.718 *	81.590 ***	67.240 ***	38.961 *
	(19.473)	(19.830)	(22.221)	(22.555)
URG	8.177 ***	7.290 ***	18.497 ***	8.564 ***
	(1.884)	(2.107)	(1.996)	(2.090)
EE	−0.594	−4.781 *	−6.419 **	0.670
	(2.604)	(2.623)	(3.137)	(2.884)
_cons	−133.753	−368.798 ***	−306.268 ***	−154.799
	(90.369)	(91.921)	(104.114)	(105.450)
N	310.000	310.000	310.000	310.000
R2	0.542	0.518	0.407	0.548
R2_a	0.486	0.458	0.334	0.488

Note: Standard errors in parentheses, and * p < 0.1, ** p < 0.05, and *** p < 0.01.

.

In Model (1), economic growth is positively correlated with employment quality with a regression coefficient of 0.316 and significant at the 1% level. The regression coefficients of CPI and URG are 32.718 and 8.177, respectively, and significant at the 10% and 1% levels. EE is negatively correlated with employment quality but not significant. In Model (2), the sharing economy positively correlates with employment quality, with a regression coefficient of 0.214, and is significant at the 1% level. The CPI and URG regression coefficients are positive and significant at the 1% level. EE passed the test at the 1% level. In Model (3), the regression coefficient of the economic structure is positive but not significant. In Model (4), three dimensions of EDQI are taken as the core explanatory variables to analyze the influence of different dimensions on employment quality. The results show that economic growth and sharing economy affect employment quality positively, but the economic structure influences it negatively. In three dimensions, economic growth is the most significant factor. From the control variables, the coefficients of CPI and URG are positive and significant. The coefficient of EE is positive but not significant.

4.5.4. Regional Heterogeneity

The samples are divided into the central, western, and eastern regions. The regression models are conducted by analyzing the influence of economic development quality on employment quality in different groups. The specific results are shown in

Table 17. Regression results for different regions.

Variables	Central Region	Western Region	Eastern Region
EDQI	0.244 (0.207)	0.237 * (0.136)	0.536 *** (0.082)
CPI	55.388 (43.297)	66.120 ** (30.840)	25.337 (38.547)
URG	13.369 ** (6.098)	16.601 *** (3.118)	1.900 (4.220)
EE	−10.491 * (5.365)	−8.389 ** (3.580)	1.431 (5.118)
_cons	−269.799 (196.42)	−307.909 ** (139.862)	−91.787 (181.268)
N	80.000	120.000	110.000
R2	0.453	0.599	0.501
R2_a	0.365	0.541	0.427

Note: Standard errors in parentheses, and * p < 0.1, ** p < 0.05, and *** p < 0.01.

.

In Table 17, economic development impacts employment quality positively in all three regions. In the eastern region, the regression coefficient of EDQI is 0.536 and significant at the 1% level. In the central region, the regression coefficient of EDQI is 0.244, which failed the significance test. Additionally, the regression coefficient of EDQI is 0.237 and significant at the 10% level in the western region. Among the control variables, the regression coefficients of CPI failed the test in the eastern and central regions; only that of the western region is positive ad significant at the 5% level. URG has the greatest impact on employment quality in the western region compared to other regions. The regression coefficients of URG are all positive, indicating that the urban–rural gap has a positive effect on regional employment quality. The coefficients of EE in the central and western regions are negative and significant at the 10% and 5% levels, respectively, indicating that education expenditure had a significant inhibitory effect on regional employment quality.

By region, each variable is significant, and the regression results are good in the western region. The economic development quality is significant in the eastern and western regions but not in the central region.

To sum up, the quality of economic development, price index, and the urban–rural gap can promote the quality of employment in the three regions, and education expenditure has an inhibitory effect on the quality of employment in the three regions. The quality of economic development has the greatest impact on the western region, the second in the central region, and the smallest in the west.

4.5.5. Robustness

(1)

Replace control variables

In regression, we add control variables one by one. Whether adding variables or reducing control variables, the regression coefficient of employment quality is positive and significant at the 1% level. It indicates that economic development quality has a positive effect on employment quality. The higher the quality of economic development, the higher the employment quality.

(2)

Change the regression model

The multiple linear regression model was used to regress the panel data of 31 provinces to test whether the regression results were consistent with those of the fixed-effect panel regression model. The results are shown in

Table 18. Results of OLS regressions.

Variables	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)	Model (6)	Model (7)
EDQI	0.606 *** (0.041)	0.605 *** (0.044)	0.467 *** (0.059)	0.629 *** (0.044)	0.453 *** (0.059)	0.475 *** (0.058)	0.463 *** (0.063)
CPI		−1.481 (19.277)			35.916 * (21.368)		52.743 * (28.092)
URG			5.811 *** (1.776)		7.415 *** (2.011)	7.060 *** (1.823)	10.452 *** (1.147)
EE				−4.391 (2.833)		−7.347 ** (2.860)	−1.482 (1.471)
_cons	13.179 *** (1.343)	20.068 (89.693)	15.397 *** (1.482)	−0.720 (9.067)	−151.085 (99.060)	−7.383 (8.990)	−234.543 * (129.50)
N	310.000	310.000	310.000	310.000	310.000	310.000	310.000
R2	0.457	0.457	0.480	0.463	0.486	0.493	0.778
R2_a	0.397	0.394	0.419	0.400	0.424	0.432	0.775

Note: Standard errors in parentheses, and * p < 0.1, ** p < 0.05, and *** p < 0.01.

.

In Table 18, the regression coefficients of EDQI are all positive and significant at the Q% level in Model (1)~Model (7), which is consistent with the panel regression models with fixed effects. For the control variables, the regression coefficient of the price index in Model (5) and Model (7) is positive and significant at the 10% level, indicating that the price index promotes employment quality. The regression coefficients of URG are all positive and significant at the 1% level, meaning that the urban–rural gap had a positive effect on employment quality. The regression coefficients of EE are all negative and significant at the 5% level in Model (6), indicating that education expenditure has an inhibitory effect on employment quality. Overall, the results of multiple linear regression models are consistent with those of the panel regression model with fixed effects.

(3)

Delete outlier data

In the panel data, there are some outlier data. For example, Qinghai’s economic development quality index is much higher than in other western regions. Additionally, the employment quality in Tibet is very high, but its economic development quality is the lowest in China. These outlier data will lead to the lack of certain reliability of the regression results. Therefore, we removed the data from Shanxi, Inner Mongolia Autonomous Region, Tibet, and Qinghai. Additionally, fixed-effect panel regressions were conducted for the remaining 27 provinces’ data during the period 2010–2019. The specific results are shown in

Table 19. Regression results after removing the abnormal data.

Variables	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)	Model (6)	Model (7)
EDQI	0.952 *** (0.045)	0.953 *** (0.045)	0.421 *** (0.048)	0.952 *** (0.037)	0.421 *** (0.048)	0.474 *** (0.063)	0.462 *** (0.059)
CPI		13.151 (37.196)			56.563 ** (27.836)		31.328 (21.244)
URG			11.205 *** (0.721)		11.351 *** (0.721)	10.093 *** (1.135)	8.396 *** (2.031)
EE				−11.907 *** (1.040)		−1.855 (1.463)	−6.972 ** (2.865)
_cons	1.551 (1.503)	−59.401 (172.398)	15.165 *** (1.427)	−33.375 *** (3.302)	−246.811 * (128.932)	8.373 (5.544)	−151.438 (98.096)
N	310.000	310.000	310.000	310.000	310.000	310.000	280.000
R2	0.596	0.597	0.774	0.717	0.777	0.775	0.498
R2_a	0.595	0.594	0.773	0.715	0.775	0.773	0.435

Note: Standard errors in parentheses, and * p < 0.1, ** p < 0.05, and *** p < 0.01.

.

In Table 19, among the regression results of the above seven models, the regression coefficient of EDQI was all significant at the significance level of 1%, and the regression coefficient was all positive. For control variables, the regression coefficient of CPI is positive and significant at the 5% level in Model (5), indicating that the price index promotes the quality of employment. The coefficients of URG are all positive and significant at the 1% level, meaning that the urban–rural gap positively affects employment quality. The coefficients of EE are, respectively, negative and significant at the 1% and 5% levels in Model (4) and Model (7), indicating that education expenditure has an inhibitory effect on employment quality.

In conclusion, by gradually adding control variables, changing the regression model, and deleting the outlier data for regression analysis, the regression results are consistent with the results of the fixed-effect panel regression, indicating that the regression results have robustness.

5. Conclusions

This paper uses provincial panel datain China from the period 2010–2019 to construct employment quality and economic development quality-indicator systems. It conducts a comprehensive evaluation model with entropy weight to measure provincial employment and economic development quality indexes. On this basis, fixed-effect panel regression models are applied to study the influence of economic development quality on regional employment quality, and the following conclusions are obtained.

First, employment structure and social security are the critical dimensions of employment quality. The employment structure has the largest weight, followed by social security. Regionally, the employment quality index of different regions has more apparent differences. Among them, Beijing, Shanghai, and Guangdong ranked in the top three, following central and western regions, with Guangxi Zhuang Autonomous Region and Gansu Province reaching the lowest.

Second, the sharing economy and economic growth are important dimensions. The sharing economy has the greatest influence on economic development quality, followed by economic growth. There are some differences in the economic development quality index among different regions. Only a few provinces with a higher economic development quality index, including Beijing, Shanghai, and Guangdong, are among the top three, while Tibet, Guangxi, and Jiangxi are the lowest.

Finally, economic development quality has a significant influence on promoting regional employment quality. In addition, the price index and the urban–rural gap significantly and positively impact employment quality, but education expenditure has a significant inhibitory effect. In different dimensions, all dimensions of economic development can promote employment quality. Economic growth has the greatest influence on employment quality, the sharing economy is the second, and the economic structure is the smallest. In different regions, economic development quality has the greatest influence on eastern employment quality, followed by the central region, with the smallest influence in the west.

Therefore, we should continue to increase per capita disposable income and per capita consumption spending, enhance the sharing economy, and vigorously promote the growth of employment quality. For the western region, it is necessary first to solve the contradiction of unbalanced economic development, implement more active employment policies, constantly expand the employment scale, carry out a series of special activities for employment services, reduce the unemployment rate, and raise GDP growth rate. For the central region, they should meet the market demand in the era of intelligence, digitization and networking to steadily improve employment quality. The eastern region, with high economic development, can launch a series of employment policies, such as social insurance subsidies and one-time house purchase subsidies. The policies can provide more jobs, attract high-tech talents to enter the local employment market, match jobs with technology, and rapidly improve employment quality. They should accelerate the realization of high-quality economic development, adhere to coordinated development within the region, and establish the leading demonstration area in the country.