EN中文

Journal of Geographical Sciences >

2021 , Vol. 31 >Issue 12: 1737 - 1756

DOI: https://doi.org/10.1007/s11442-021-1920-2

City size and employment dynamics in China: Evidence from recruitment website data

  • HUANG Daquan , 1 ,
  • HE Han 1 ,
  • LIU Tao , 2, 3, *
Expand
  • 1. School of Geography, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China
  • 2. College of Urban and Environmental Sciences, Peking University, Beijing 100871, China
  • 3. Center for Urban Future Research, Peking University, Beijing 100871, China
* Liu Tao (1987-), Assistant Professor, specialized in urbanization and migration. E-mail:

Huang Daquan (1971-), Associate Professor, specialized in spatial planning and urban development. E-mail:

Received date: 2020-12-23

  Accepted date: 2021-08-10

  Online published: 2022-02-25

Supported by

Major Project of National Social Sciences Foundation of China(20&ZD173)

Copyright

Copyright reserved © 2021. Office of Journal of Geographical Sciences All articles published represent the opinions of the authors, and do not reflect the official policy of the Chinese Medical Association or the Editorial Board, unless this is clearly specified.
Fold

Abstract

This article explored China’s urban employment dynamics with particular focus on the city size effect. Big data derived from the largest recruitment website were used to examine the direct and indirect impacts of city size on employment demand by using mediating and moderating models. We also investigated the roles of the government and location factors which have seldom been considered in literature. Results showed that the concentration degree of new jobs is higher than that of stock employment and population across cities, implying a path dependency mechanism of job creation and employment expansion. Meanwhile, numerous job posts in inland central cities are probably a symptom of more even distribution of employment in future China. Econometric models further verified the significant correlation between city size and job creation. Moreover, industrial diversity, fixed asset investment, and spatial location have heterogeneous effects on employment demand in cities of different sizes and different levels of administration. These results can not only deepen our understanding of the crucial role of city size in urban employment growth but also demonstrate the future trend of labor and population geography of China. Policy implications are then proposed for job creation in cities of China and other developing countries.

Key words： employment dynamics; city size; job creation; industrial diversification; moderating effect; recruitment website

Cite this article

HUANG Daquan , HE Han , LIU Tao . City size and employment dynamics in China: Evidence from recruitment website data[J]. Journal of Geographical Sciences, 2021 , 31(12) : 1737 -1756 . DOI: 10.1007/s11442-021-1920-2

1 Introduction

Employment has always been a major issue of concern to individuals, cities, and countries (Huang et al., 2021). The global economic depression and rising unemployment caused by the 2008 financial crisis has made employment a more pressing issue for international society (Cai and Chan, 2009; Popov and Rocholl, 2018). An unavoidable concern facing every country and region in the world is how to promote the creation of jobs and decrease unemployment. In recent years, a growing number of empirical studies have investigated how to promote job creation and increase the vitality of urban employment. As the main carriers of economic development, innovation and technological progress are considered the key factors of sustained economic growth and urban employment expansion, which has been confirmed in developed and developing regions (Merikull, 2010; Lachenmaier and Rottmann, 2011; Horbach and Rennings, 2013; Kunapatarawong and Martinez-Ros, 2016). On this basis, other studies have discussed the differential impact of product innovation and process innovation on employment in different industries (Storey et al., 2002; Merikull, 2010). A common method used by the government for increasing employment is to increase investment, but how to ensure the sustainability of investment efficiency and the allocation of investment remains a matter of great concern. Excessive investment in large cities and insufficient investment in small and medium-sized ones are common in both transitional economies and mature market economy countries (Panayiotou and Medda, 2014; He, 2018). Additionally, given the role of cities as the key nodes of regional division of labor, the impact of urban traffic and location have also been a concern (Krumm and Strotmann, 2012).
Employment dynamics are an important indicator of urban employment vitality (Li, 2013). In the late 20th century, scholars gradually realized that it was not enough to study urban employment only from the dynamics of total employment and unemployment (Davis and Haltiwanger, 1992). Some scholars even argued that many theories would lose their power of explaining employment issues in the manufacturing sector when the creation and disappearance of jobs were separately considered (Davis and Haltiwanger, 1992). In reality, significant offsetting employment flows occur within and across industries and regions, which means that more intense job creation and disappearance lies behind the steady change in total employment (Dunne et al., 1989). The dynamics of this process are more important for understanding urban employment issues (Francis, 2009), and so, economists, sociologists, and urban researchers have begun to attend more to employment dynamics and job mobility (Li, 2013; Huang et al., 2015; Huang et al., 2017; Huang et al., 2021). This type of research usually employs longitudinal data and manufacturing statistics to study job creation and disappearance in an industry or enterprise. It yields a better understanding of the processes and mechanisms behind the dynamic changes in urban total employment, and provides policy makers with more effective employment promotion strategies (Essletzbichler, 2004; Åstebro and Tåg, 2017).
Urban employment and city size have been inseparable in previous urban employment research. The impact of city size on economic performance has long been an important theme of urban economics literature (Abdelrahman, 1990; Panayiotou and Medda, 2014; Zhang et al., 2016). Scholars try to explain the economic benefits of urban scale in depth through theoretical analysis, including, typically, the Marshall-Arrow-Romer model (MAR), Porter externalities, and Jacobs spillover (Glaeser et al., 1992). Economic prosperity also corresponds to increases in urban employment. Therefore, these studies have uncovered an implicit meaning of the relationship between urban size and employment. Economics explores the impact of various factors on urban employment and finds that city size is considered the core influencing factor on employment issues (Gabler, 1971). Some empirical studies show a positive correlation between urban size and employment demand and quantitatively estimate the impact of changes in urban size on urban employment demand (Gao et al., 2015; Hansen and Winther, 2018). Other scholars argued that considering only the size of the city is insufficient for understanding the performance of its economy. The impact of other factors, such as spatial location and level of diversification, needs to be discussed on the basis of city size, which is a more reasonable way than the relationship between single factors and employment (Coffey and Shearmur, 1998).
China is an ideal case for studying employment issues. China has experienced rapid urbanization in the 40 years of reform and opening up, and it will still be at a stage of high-speed urbanization in the future. With the expansion of the urban population, employment pressure has also increased. Against the backdrop of financial crisis, changes in the international trade environment, and domestic and regional competition, employment issues have become more important and prominent (Cai and Wang, 2010; Zhang and Yanxia, 2015). The report of the 19th National Congress of the Communist Party of China clearly stated that “employment is the greatest livelihood of the people.” The employment problem has risen to the level of national strategy. Using Chinese cities as a research area to study urban size and employment has unique advantages. First, China is a typical country in transition. Government and market forces have both played important roles in the process of employment promotion, giving researchers an opportunity to explore the role of these two forces. Second, Chinese cities vary greatly in characteristics such as size, degree of diversity, administrative hierarchy, and spatial location. Third, China’s rapid urbanization and transportation network have led to increased labor mobility and a complex urban system, which provide a systematic sample for urban employment research. It may also provide a good reference point for other developing countries.
The employment issue has received widespread attention in China. Many studies have analyzed the changing size and structure of employment, the current situation and future trends of the employment market, and the increasing challenge of urban unemployment (Cai and Wang, 2010; Lee and Hoshino, 2017; Dong, 2018; Kucera and Jiang, 2018). After the financial crisis, the impact of export decline on export-oriented and investment-driven employment in Chinese cities and the means for expanding domestic demand to promote employment have garnered attention (Cai and Chan, 2009). Scholars have also analyzed the impact of policies and regulations on employment (Chen and Funke, 2009). In addition to analysis at the national level, scholars have also empirically studied a number of factors influencing urban employment at the city level. These include the employment effect generated by urban size, transportation, investment, and new development zones (Gao et al., 2015; Zheng et al., 2017; Dong, 2018; He, 2018), as well as the impact of foreign trade on Chinese employment (Tang et al., 2016); the relationship between changes in scale, wages, and employment (Wheaton and Lewis, 2002; Fang and Lin, 2015); the heterogeneous effects of product and process innovation in employment growth (Hou et al., 2019). Existing empirical research has been based mainly on developed country employment models and theories, using statistical data to quantitatively study the impact of certain specific factors on regional employment.
There are deficiencies in the existing literature. First, employment dynamics provide a more reasonable approach to our comprehensive understanding of urban employment; however, more attention has been paid to total employment than to employment dynamics in existing research (Davis and Haltiwanger, 1992). The unavailability of data is an important restriction. This article uses employment demand information released by recruitment websites to measure employment dynamics and study urban employment issues. Second, differences in the influencing mechanism of urban employment dynamics in cities of different sizes have rarely been examined. Though some empirical studies have measured the impact of certain factors on the change in urban employment by using econometric models, such large sample studies often omit interactions between influencing factors and the way they combine in different regional contexts (Rodrik et al., 2004; Hansen and Winther, 2018). Third, urban size is a determinant of city employment vitality, but many influencing factors of employment vitality have an impact on urban size. Existing studies pay insufficient attention to this interaction. Fourth, existing research is mainly based on the practices and theories of developed countries. It is questionable to what extent these theories and conceptual frameworks are applicable in developing countries (Alpkokin et al., 2008)? China is a rapidly developing transition country. The focus on Chinese urban employment not only provides a point of reference for other developing countries but also enriches existing theories based on Western economies.
Therefore, this article attempts to study the relationship between city size and employment dynamics by using the urban employment demand information released by a recruitment website and using statistical data as an aid. The aim is to answer the following three questions: How do employment vitality and spatial distribution characteristics differ between cities? How do the factors underlying urban employment demand vary across cities with different sizes? What role does city size play in the formation of disparity in urban employment vitality?
The remainder of the paper is structured as follows. In the next section, a theoretical framework for understanding China’s employment dynamics is proposed. Section 3 introduces the data, models and variables. The results are described in section 4. The last section is the conclusion and discussion.

2 Understanding the employment dynamics of Chinese cities

City size is a crucial factor in the creation of jobs through a variety of urban agglomeration economies. The most direct impact of the difference in the sizes of cities is the scale of demand for products and services produced by urban residents in production and life, which has a linear relationship with city size. Further, scale increases the number of jobs through the production advantage of tradable products, and this relationship is non-linear (Coffey and Shearmur, 1998). Metropolises have higher labor productivity and more frequent technological updates because of input sharing, labor market pools, and knowledge spillovers (Glaeser et al., 1992), making the area attractive for the production of tradable products. The concentration of the tradable sector has increased urban employment and income, while rising incomes have contributed to the expansion of the non-tradable product sector. Relevant empirical studies have also confirmed the existence of the employment multiplier in the tradable sector, but the magnitude of the employment multiplier varies from industry to industry. The multiplier for traditional manufacturing is about 1.6, and for the core sector of the trade industry, it can reach as high as 5 (Gao et al., 2015; van Dijk, 2017). A metropolis is a cluster of highly dynamic industrial sectors that are separated from production. Sectors with high employment multipliers provide cognitive and cultural human capital for economic growth, such as advanced technology, business services, finance, and culture. Generally speaking, the spatial concentration of production activities and factors can bring additional benefits and/or cost savings. The attractiveness of small-scale administrative regions for production activities is usually limited by the lack of productivity, technological innovation, and innovation capacity. In China, efficiency is significantly increased when the size of the urban population exceeds one million (Zhang et al., 2016). Although some urban theories describe an optimal city size, studies of China generally claim that the benefits of small cities under the small city bias policy are insufficient (Au and Henderson, 2006). Therefore, we hypothesize a positive relationship between city size and employment vitality, that is, the employment vitality of big cities is higher than that of small cities.
The industrial structure and the employment structure are closely related concepts (Drucker, 2015), and the industrial structure reflects the proportion of the secondary and tertiary industries. The theory of regional development believes that the tertiary industry is ultimately the main accommodation place for jobs, which means that there will be more urban jobs in the tertiary industry. Therefore, it is generally believed that the higher proportion of tertiary industry in the employment demand will be higher, But the premise of this fact is that the city has a certain population size and economic level (Henderson, 2002), for underdeveloped regions or cities with relatively small scale in China, the tertiary industry, dominated by the service industry, also accounts for a high proportion, but obviously, the employment demand in these cities is far less than that in metropolises.
The debate between diversification and specialization is closely related to city size, and the economy of urban agglomeration is another factor with a significant effect on city prosperity and job creation. Some economists believe, based on the theory of industrial connections and comparative advantages between cities, that specialization is conducive to economic development and employment growth (Henderson, 1997). However, some scholars have indicated that a diversified industrial structure attracts labor forces from different industries to cities. Such agglomeration can promote innovation and increase labor productivity (Duranton and Puga, 2000). Compared with the benefits yielded by professional agglomeration in industry, some scholars believe that innovation and knowledge spillovers contribute more to urban growth (Glaeser et al., 1992). This debate on localization and urbanization economies as two types of urban agglomeration economy indicates two influence paths of size on city growth, both of which can be represented by economic and employment expansion. Empirical research finds that there is a positive correlation between urban diversification and city size (Begovic, 1992), which shows that the impact of the urbanization economy on urban employment exceeds that of the localization economy (Begovic, 1992; Desrochers and Leppala, 2011). However, the impact of the level of industrial diversification on the vitality of urban employment varies, and this affects the urban-scale system and the regional division of production in the system (AbdelRahman, 1996; Hansen and Winther, 2018). It is thus hypothesized that the greater the diversity of a city’s industrial structure, the higher the employment dynamics of the city.
Although the government plays an important role in employment promotion in all countries, in transitional socialist China, the state is extremely important and may exert influence through more diverse means of job creation than in Western countries. This dominance of the government could explain the differences in employment issues between China and developed countries (Zheng et al., 2017). In China, the government can influence the job creation capacity of cities in many direct and indirect ways. First, affected by a bias in the government’s allocation of resources to administrative centers, cities with different administrative levels differ significantly in terms of authority settings, resource allocation, and institutional arrangements (Zeng et al., 2016). This political bias means that cities with higher administrative levels, such as provincial capitals, have unique advantages that permit greater production and consumption, thereby driving employment growth. Additionally, cities with higher levels of administration are generally more appealing and, therefore, more populous.
Second, as a specialized production area affiliated with a main urban center, a development zone has a large employment multiplier because of the accumulation of high-end manufacturing and innovative industries, which directly promote urban employment and city size. However, development zones do not necessarily have a significant effect on economic growth and job creation. The extant literature shows that development zones play a significant role in promoting employment in developing countries but not necessarily in developed countries (Busso et al., 2013). The degree of the effect is also altered by the size of the central city (Zheng et al., 2017).
Third, polarization and diffusion effects can exist simultaneously in the development of central cities. The impact of this diffusion effect on surrounding cities increases their productivity through knowledge spillover but results in the spillover of some economic activities to central cities. For example, in order to reduce costs, standardized production activities migrate to smaller surrounding cities, driving increased employment there, but this spillover must maintain convenient connections to the central city (Ali et al., 2011). Therefore, the diffusion effect of the central city is affected by the distance attenuation effect, that is, its proximity to the central city. These regions have the opportunity to accommodate more employment spillover and proliferation (Krumm and Strotmann, 2012; Hansen and Winther, 2018). As a result, the role played by governments, especially those at high levels should not be overlooked in understanding employment dynamics of Chinese cities. Cities holding a high position in the administrative hierarchy and those located near to politically high-level cities are hypothesized to have high levels of employment dynamics.
Investment is the primary engine of industrial expansion, job creation, and city growth (Blomstrom et al., 1996; Madsen, 2002). Kahn proposed the employment multiplier of investment in 1963, claiming that the increase in net investment brings about a multiple of total employment increase and an increase in initial employment. Some subsequent empirical studies have confirmed the relationship between investment and employment growth. Investment plays an important role in some rapidly developing regions, including China (Willis, 1985; Ianchovichina et al., 2013; He, 2018). Enterprise production investment and infrastructure investment are the two main types of urban investment. An increase in the former can directly stimulate business expansion and the establishment of new enterprises that serve as the main source of new jobs (Adelino et al., 2017). Differences in urban infrastructure investment have an important impact on corporate site selection (Panayiotou and Medda, 2014; Zheltenkov et al., 2017). Moreover, infrastructure investment has a direct job creation and related multiplier effect on service industries. However, some studies have shown that the employment effect of investment has not always been significant because of differences in city size and short-term economic trends (Qin and Song, 2009). In fact, city size is also closely related to the employment creativity of investment. On the one hand, the economic benefits of unit investment and the ability to create jobs differ in cities of different sizes. On the other hand, investment through urban expansion can generate agglomeration economies, which in turn, can increase employment multipliers. Consequently, in general, investment in a city is hypothesized to have positive effect on employment expansion in the city.

3 Data and models

3.1 Internet recruitment data

The Internet has become the main channel for job seekers to obtain information about job opportunities. Because of the shortcomings of traditional recruitment methods, such as low coverage, poor efficiency, and high costs, recruitment websites have been increasingly favored by job seekers (Van Rooy et al., 2003; Suvankulov et al., 2012). In this study, 51Job (https://www.51job.com) was selected as the source for research data. As the earliest listed online human resources service company in China, 51Job has the largest number of business users and job applicant registrations, and the website includes nearly every industry in the urban economy. We collected employment demand data from the recruitment website through a cyclic program created in website crawler software. The data include the recruiting number, sector, salary, workplace, education and experience requirements, and detailed description of the position as well as basic information on the recruiting company.
Selecting 51Job as the only data source has its advantages and shortcomings. While duplication and mismatching of data from various sources are avoided, there is also the possibility of some inevitable data biases. The most significant one is caused by omission of data for some industries that have their own popular recruitment websites respectively. Moreover, the coverage rate of online platforms for recruitment information varies greatly across cities. The popularity of online recruitment is very high in big cities, whereas traditional recruitment methods remain dominant in less developed areas and small cities. Even with these biases, online recruitment data is an ideal data source for research on employment vitality. The number of new jobs created in each city is not available in any existing statistical source. This shortage makes it impossible to accurately validate the representativeness of the online data on the one hand; it also demonstrates the necessity and urgency to use non-statistical data for investigating this issue on the other hand.
The study area covers all of the 287 prefecture-level cities which are defined as all districts, excluding counties and county-level cities under their jurisdiction. A total of 2,897,645 items of position data were obtained for all cities during May and June 2018. After duplicate data were eliminated, 2,687,015 items remained. Each city had, on average, 9,558 new job requirements daily. The number of cities that reached the national average of daily job creation was 38, and these cities accounted for more than 80% of total daily job demand.
The distribution of China’s new job positions exhibited remarkable spatial agglomeration (Figure 1). At the national level, cities with high ability of job creation are mainly concentrated in coastal regions, while there is relatively low demand for new workers in cities in the northeast, southwest, and northwest. The Yangtze River Delta, Pearl River Delta, and Beijing-Tianjin-Tangshan Region as three major mega-regions create the largest numbers of new job positions. Significant spatial agglomeration also exists at the provincial level especially in central provinces and a couple of western ones where provincial capitals have extremely high employment vitality with very limited number of new jobs provided in other cities. Although the above spatial features of job creation are generally similar to those of the total employment in Chinese cities, the concentration of the former is much more notable than the latter. This contrast is evidenced by the difference between the shares of top ten cities in national total of the two quantities which are 31% for employment and 56% for new positions.
Figure 1 Urban employment and daily job creation in China

3.2 Model and variables

The regression model for urban employment demand is as follows:
employment=β1size+β27industrial structure+β3 second-tertiary ratio+β4 provincial capital+β5development zone+β6investment+β7location+constant+ε
where the dependent variable is city employment demand as measured by the volume of new job positions derived from the website 51Job.
The explanatory variables are city size, second-tertiary ratio, industrial diversity, provincial capital, the area of development zone, fixed asset investment and location (Table 1).
Table 1 Variable definition and data sources
Variable Definition Data source
Employment demand New job positions www.51job.com
City size City population China Urban Construction Statistical Yearbook
Second-tertiary ratio Ratio of the secondary to the tertiary industry China City Statistical Yearbook
Industrial diversity Herfindahl-Hirschman index China City Statistical Yearbook
Provincial capital Provincial level cities and provincial capitals=1, others=0 China City Statistical Yearbook
Development zone Area of development zone Ministry of Natural Resources
Investment Fixed asset investment China City Statistical Yearbook
Location Distance to central city China Urban Construction Statistical Yearbook
City size is measured by population of the city, which not only affects the employment of non-tradable product sectors in the city but also attracts the concentration of tradable product sectors through the benefits of agglomeration (Gabler, 1971; Zhang et al., 2016). Unlike other scholars who use the ratio of tertiary industry to secondary industry, we use the ratio of the secondary industry to the tertiary industry to measure the industrial structure. The reason is that the share of the secondary industry can be directly compared, and the tradable sector dominated by the secondary industry is an important basis for maintaining the scale and vitality of employment (van Dijk, 2017; Wang and Chanda, 2018). The ratio of the secondary industry to the tertiary industry reflects the advanced industrial structure. The tertiary industry accounts for a higher proportion of many small cities in China and areas with backward development levels, but it cannot explain its advanced industrial structure. Therefore, we use the ratio of the secondary industry to the tertiary industry to explain the employment problem.
Industrial diversity is measured by the Hirschman-Herfindahl index (HHI) (Henderson et al., 1995; Matsumoto et al., 2012; Tao et al., 2019). HHI can be calculated as follows:
$\text{Divi=1}\sum\nolimits_{n=1}^{Ni}{S_{i,n}^{2}}$
where Ni is industrial type in region i, and Si,n is the ratio of the number of persons employed in industrial category n in the region to all persons employed in the region. The more evenly distributed across industries, regional employment is, the greater the HHI and degree of diversity. When using the proportion of employed population in various industries to measure the degree of diversity of all urban industries, one must consider the group size of employed people. The ratio of fi is the number of people employed in the region to the total population of the region,
$\text{Divi=}{{f}_{i}}\cdot (1\sum\nolimits_{n=1}^{Ni}{S_{i,n}^{2}})$
Investment in fixed assets provides the necessary basic conditions for the sustainable development of cities. The direct means that local governments can use to promote employment expansion is to attract enterprises to settle in the city (Neumark et al., 2006). The scale of urban fixed asset investment is one of the factors considered in the location selection of enterprises. Spatial location has an impact on the employment of the city through the diffusion effect of the central city. It is generally believed that, because of the distance attenuation principle, cities closer to a metropolis can subsume technical knowledge spillover from the central city, which has a positive impact on employment (Krumm and Strotmann, 2012). Administrative centers often have advantages in terms of politics, capital, education, and transportation and are, therefore, generally preferred by companies. Provincial capital is included as a dummy variable to investigate the difference between provincial level cities and provincial capitals and other cities. Spatial location is measured by the distance of the city from the expressway of the region’s central city (selected as the provincial capital). To estimate the location of the regional central city, we assume the city as a circle and calculate its radius based on the city’s total built-up area. The radius is treated as the distance of the central city to itself. During the regression process, the logarithm of urban employment demand, urban population size, fixed asset investment, development zone size, distance from port, and space location were considered.
The data on the size of urban stock employment, the calculation of urban GDP, industrial diversity, and fixed asset investment employed in this study were derived from the China City Statistical Yearbook; the population data were derived from the sum of the registered population and the temporary population in the China Urban Construction Statistical Yearbook, which was also the data source for the built-up area. We used local statistical data to compensate for cities lacking data in the China City Statistical Yearbook. The descriptive statistics of variables are shown in Table 2.
Table 2 Descriptive statistics of variables
Variable N Min Max Mean SD
Employment demand 278 39 342,76 9,195.05 35,196.70
City size 278 74.54 2,479.00 204.02 259.10
Second-tertiary ratio 278 0.001 4.58 1.01 0.67
Industrial diversity 278 0.030 0.850 0.172 0.091
Provincial capital 278 0 1 0.31 0.05
Development zone 278 0 30,763 1,411.23 2,830.90
Investment 278 25.78 1,473.3 1,055.17 1,620.38
Location 278 12.00 1,936 227.54 192.10

3.3 Mediating and moderating effects

Such factors as industrial diversity, administrative hierarchy, the size of development zones, fixed asset investment, and spatial location, affect urban employment demand and urban size. Additionally, the impact of these factors on urban employment varies for differences in city size. In this sense, the size of the city may have mediating and moderating effects on the impact of multiple factors on urban employment which can be estimated simultaneously in statistical models (Xie, 2010). For example, scholars have different views on the employment effect of industrial diversity, which is largely a result of differences in the size of a city (Duranton and Puga, 2000; Fu and Hong, 2011). Regarding the moderating effect, we divided the cities into three levels: large cities with a population of more than three million, medium-sized cities with a population of from one to three million, and small cities with a population below one million. The three levels were valued 2, 1, and 0 respectively. The formula for the moderating effect is as follows:
employment = aX + bS·X+constant+ε
where X includes industrial structure, industrial diversity, administrative level, the size of development zone, fix asset investment and spatial location; S is city size level; and a and b are coefficients.
For mediation effects, the most commonly used approach is the stepwise method (Baron and Kenny, 1986; Hayes and Preacher, 2014; Judd and Kenny, 1981). The basic operation is to centralize all variables and test whether the regression coefficients of the independent and dependent variables are significant and whether the regression coefficients of the intermediate and dependent variables are significant. If both of the above coefficients are significant, this means that the effect of the independent variable on the dependent variable is at least partially achieved via the intermediate variable. A Sobel test is required if at least one is not significant, that is, the mediating effect is determined by testing the significance of the statistic z = aˆbˆ/Sab, where aˆ and bˆ are the estimates of a and b, Sab is the standard error of aˆbˆ, and Sa and Sb are the standard errors of aˆ and bˆ.
${{S}_{ab}}\text{=}\sqrt{a_{{}}^{\wedge 2}S_{b}^{2}+b_{{}}^{\wedge 2}S_{a}^{2}}$
According to the introduction of the estimation and testing of the multiple mediation model, the additional regression equations of the multiple mediation effect corresponding to the theoretical model of this study is as follows:
employment=γ1 second-tertiary ratio+γ2 industrial diversity+γ4 provincial capital+ γ5 development zone+γ6 investment+γ7 location +constant1+ε1
size=α1 second-tertiary ratio+α2industrial diversity+α4 provincial capital+α5 development zone+α6 investment+α7 location+constant2+ε2

4 Results

4.1 Regression results and spatial variations

Table 3 shows the regression results for the factor influencing urban employment demand. The significant effect of city size is observed in the national and three regional models. More importantly, the employment demand increases by 1.5% when city size increases by 1% which indicates that new positions are concentrated more significantly in large cities than urban population. The effect of the expansion of city size on the increase in employment demand is non-linear, similar to the situation in developed countries (Coffey and Shearmur, 1998). Large cities can create more job opportunities for their residents and thus have greater potential to attract job seekers from outside. This result agrees with classic urban growth theories that are rooted in agglomeration economies. The reasons are multifold. First, according to previous theoretical analyses, large cities have higher consumer demand for non-tradable products. Second, big cities have significant advantages in terms of production in the tradable sector. At the same time, metropolitan areas have higher employment vitality because of the employment multiplier generated by the tradable product sector. Additionally, different types of manufacturing have different employment multipliers. The core sectors of the trade industry have higher employment multipliers for traditional manufacturing. And this kind of high multiplier effect is generally concentrated in large cities, which is the reason for the non-linear relationship between city size and employment demand. For small cities, their productivity levels are relatively backward (Zhang et al., 2016). The main source of employment in small cities is in the non-tradable sector, which provides services for urban residents. This proportion of employment is more than 60% in small and medium-sized cities in developed countries, and it is even higher in developing countries (Miller and Bluestone, 1988; Lombardo and Ravenna, 2012). Migration in China, the primary goal of which is the search for high-quality employment, has received widespread attention, and our findings provide evidence that large cities can indeed create more new employment. However, this agglomeration effect is found mainly in the eastern region where 1% increase in the size of cities can generate 1.8% growth of job opportunities. In contrast, the elasticity coefficient of urban growth on employment in central and western regions is less than one which means large cities do have more job opportunities but their advantage in employment is lower than that in city size measured by urban population. The result in central China is similar to that of studies of some developed countries such as France and Japan. The growth of cities in these countries does not converge on an optimal size or divergent growth of large cities, but maintains parallel growth (Eaton and Eckstein, 1997). In western China, the job creation ability of large cities seems weaker compared with their coastal counterparts and also small and medium-sized cities, indicating a convergence trend of city size.
Table 3 The regression results of factors affecting urban employment vitality
Model 1
Nationwide		 Model 1a
Eastern region		 Model 1b
Central region		 Model 1c
Western region
City size 1.538***(0.365) 1.841**(0.657) 0.915**(0.307) 0.571**(0.208)
Second-tertiary ratio ‒0.007***(0.002) 0.364**(0.169) 0.054**(0.123) 0.125**(0.041)
Industrial diversity 0.169**(0.026) 0.195**(0.067) 0.721**(0.232) 0.031*(0.017)
Provincial capital 0.107*(0.059) ‒0.387*(0.211) ‒0.093*(0.051) 0.709(1.86)
Development zone 0.03*(0.016) ‒0.075*(0.044) 0.029*(0.018) ‒0.001(0.057)
Investment 0.382*(0.198) 0.131*(0.051) 0.024*(0.013) ‒0.67**(0.231)
Location ‒0.079***(0.015) ‒0.543***(0.115) ‒0.428**(0.141) ‒0.153**(0.054)
N 278 97 83 75
F 127.394*** 31.95*** 27.25*** 9.929***
Adjustment R2 0.757 0.693 0.703 0.455

Note: *Significance at a.05 level, **significance at a.01 level, ***significance at a.001 level. Standard error in parentheses.

According to classical labor transfer theory, with improvements in the level of urbanization, labor shifts from the primary industry to the secondary and tertiary industries, and eventually the tertiary industry becomes the prime locations for accommodating labor. The results indicate a more complex story in China. Nationwide, the ratio of the secondary industry to the tertiary industry is positively related to employment demand, indicating that China remains in the stage of large-scale accumulation of labor of secondary industry. In contrast, the results of three regional models have all demonstrated a positive relationship between the dominance of tertiary industry in urban economy and a city’s ability of job creation. Second, many small cities in China receive low urban benefits (Zhang et al., 2016). The production and consumption of products in these cities are mainly dominated by the non-tradable product sectors that belong mostly to the service industry. Therefore, tertiary industry accounts for a high proportion, but employment demand is sluggish. The proportion of manufacturing gradually increases when the urban scale gradually increases because of the emergence of urban agglomeration benefits. As city size continues to expand, tertiary industry’s proportion again rises. Thus, the relationship between the proportion of secondary and tertiary industries explains differences in employment demand between different size cities. This empirical judgment can be tested by a hierarchical regression of city size.
On the whole, industrial diversification has a positive effect on employment. The motivation of agglomeration production comes from complementary producer networks, diverse local labor networks, and learning processes. Hence, diversification can increase labor productivity and knowledge spillovers by promoting innovation (Desrochers and Leppala, 2011). Moreover, industrial diversification can attract labor forces from different industrial backgrounds. It has been found that a city’s knowledge spillover is mainly inter-industry rather than within-industry (Glaeser et al., 1992). Therefore, while the diversified industrial environment increases the size of the local consumer market, it is also conducive to strengthening the agglomeration economy. This is in accordance with most empirical results in previous literature. Studies in developed countries also found that specialized production is more conducive to innovation and urban economic growth (Henderson, 1997). The significant positive effects exist in all three main regions of China and the largest marginal effect is observed in the central region. It implies that industrial diversity is most important in regional development at the middle stage of industrialization.
The role played by the state in job creation shows a significant regional variation. In general sense, cities at a higher level of administration and those hosting more and larger development zones have greater ability of job creation. An urban development zone reflects the government’s preferences regarding resource organization, solves the problem of land coordination and the spatial coordination of an enterprise, and enables some specific enterprises to gather in the development zone. However, different economic benefits are generated by development zones in some developed countries (Busso et al., 2013). In some developed countries, local policies generally target backward areas. In Nordic countries, governments often focus on capital or first-tier cities, and do not sufficiently invest in or develop second-tier cities (Parkinson et al., 2015). The construction of development zones in Chinese cities is undertaken with a consideration of both efficiency and equity (Alder et al., 2016). As a result, the nationwide pattern is not the case for within the three regions at different stages of development. Specifically, the intra-regional variation in the west cannot be explained by the two variables, indicating that state-led development promotion strategies may stimulate regional economic growth, but have very little achievement in job creation. Controlling for other variables, cities with advantages in administrative resources and development zones are less capable to help enterprises develop well or less favored by enterprises therein. The latter are not able to or not willing to provide more positions in these cities with strong state power. Results of the central region model are a combination of the above two. In this region, more developable area is helpful for a city and its enterprises to create more jobs, while a higher level of administration is not.
Spatial variation also exists in the effect of investment on job creation. The results of models for full sample, eastern, and central regions are consistent to our hypotheses. More investment will create more jobs in Chinese cities. Nevertheless, the relationship between investment and job creation is negative in western China. This might imply a crowding-out effect of state investment. The majority of investment in the vast west comes from the central state or state-owned banks and concentrates on huge projects with limited relation to local development. This result does not signify the meaningless or unnecessity of such investment. But it does remind the central and local governments to pay more attention to the effectiveness and efficiency of future investment in remote areas.
There is a significant negative correlation between employment demand of a city and its distance from the provincial capital. The spillover effect of central cities makes more jobs available in neighboring cities. This conclusion is similar to the results of research in developed and other developing countries (Hansen and Winther, 2018). Regressions by region show that the spillover effect is the greatest in the eastern region, moderate in the central region, and the weakest in the underdeveloped west. This provides a realistic basis for the development of China’s megaregions in the developed coastal region, while the national strategy to cultivate megaregions in inland provinces seems relatively difficult to success.

4.2 Moderating effects of city size

The adjustment of city size on the impact of various factors of employment vitality is verified by using city size as a moderator. Tables 3 and 4 show the regression results. The moderating effects of city size exist in the influence of industrial diversity, development zone, investment, and location.
Table 4 The regression results of the moderating effect
Model 1 Model 2
City size 1.538***(0.365) 1.529***(0.316)
Second-tertiary ratio ‒0.007***(0.002) ‒0.009**(0.063)
Industrial diversity 0.169**(0.026) 0.101**(0.945)
Provincial capital 0.107*(0.059) 0.124*(0.082)
Development zone 0.03*(0.016) 0.025*(0.013)
Investment 0.382*(0.198) 0.215**(0.819)
Location ‒0.079***(0.015) ‒0.152**(0.054)
Second-tertiary ratio*City size 0.021(0.015)
Industrial diversity*City size 0.075**(0.026)
Provincial capitals*City size ‒0.727(0.514)
Development zone*City size 0.335**(0.129)
Investment*City size 0.112**(0.042)
Location *City size 0.181*(0.127)
N 278 278
F 127.394*** 39.328***
Adjustment R2 0.694 0.771

Note: *Significance at 0.05. **Significance at 0.01. ***Significance at 0.001. Standard error in parentheses.

The effects of industrial diversification on urban employment appear to be different for cities of different sizes. The positive effect of industrial diversification on the increase in employment demand of a city is greater in large cites than that in small ones. This difference is not large in quantity but significant in statistical sense. This common but different positive correlation differs from that was found in developed countries like Japan and United States (Henderson, 1997). In China, cities in nature are regional centers of administration, economy, education and all other public and commercial services. This comprehensive function leads to a high level of industrial diversity in most Chinese cities. Against this background, diversification became an inevitable path of development and only those with a higher level of industrial diversity are able to provide more job opportunities for local labors and migrant workers. For smaller cities, a specialized industrial structure is more conducive to promoting innovation and reducing production costs, giving its products an advantage over a larger area. Companies in the same industry will also move in, In addition, the high productivity industrial structure has a high employment multiplier, which can drive more employment in the service industry. Therefore, in small cities, specialization is a more reasonable choice.
A common urban development strategy in China is to stimulate the employment effects of development zones. The results show that the impact of the size of the development zone on urban employment is significantly affected by the size of the city. A development zone is generally a gathering place for economic activities such as innovative economy and standardized manufacturing sectors in the city. Only when the scale of the entire city reaches a sufficient level can the development zone achieve higher economic benefits and employment promotion.
The employment spillover effect of regional central cities will affect the employment demand of surrounding cities to a certain extent. From the regression results, it can be seen that the regions close to the central cities across the country have higher employment vitality, but this proximity effect is also limited by the size of the city. Generally, the urban employment vitality increases with the increase of the scale at the same distance from the regional central city. Because the economic activity of the central city is spreading out, the carrying capacity of the surrounding cities is different, and it is easier for large-scale cities to establish associations with the central city. Urban employment needs are fewer, so the diffusion effect of regional central cities only appears in cities of sufficient size (Henderson, 1997).

4.3 Mediating effects of city size

As we hypothesized, many factors have effects not only on the ability of a city to create new positions for job seekers but also on the size of the city. The mediating effects of city size on the impact of these factors on job creation are thus tested. The results in Table 5 show the existence of the mediating effects on industrial structure, diversity and location.
Table 5 The regression results of the mediation effect
Model 3
Employment		 Model 4
City size		 Model 1
Employment		 Mediating effect Proportion
City size 1.538***(0.365)
Second-tertiary ratio ‒0.013**(0.005) ‒0.004*(0.002) ‒0.007***(0.002) 0.006 46%
Industrial diversity 0.511**(0.182) 0.223**(0.089) 0.169**(0.026) 0.342 66%
Provincial capital 0.285(0.209) 0.116***(0.023) 0.107*(0.059)
Development zone 0.049*(0.031) 0.013(0.011) 0.03*(0.016)
Investment 0.788**(0.301) 0.264(0.211) 0.382*(0.198)
Location ‒0.115***(0.025) ‒0.024**(0.009) ‒0.079***(0.015) ‒0.039 23%
N 278 278 278
F 78.312*** 47.738** 127.394***
Adjustment R2 0.641 0.442 0.757

Note: *Significance at.05. **Significance at.01. ***Significance at.001. Standard error in parentheses.

The size of the industrial structure reflects the relative proportion of the tertiary and secondary industries in a city. The theory of the stage of regional development holds that during the process of economic development, the labor force is undergoing a shift from the primary industry to the secondary industry and then to the tertiary industry. Eventually, the tertiary industry becomes the main place to accommodate urban employment. When the proportion of industries at the tertiary stage is higher within an economically developed area, higher employment vitality results. However, after adding the variable of city size, this variable exerts a mediating effect. The reason for this is that, in China’s urban system, it is not only economically developed regions that have a high demand for tertiary industry. In cities with populations less than one million, the advantages of tradable product production are not obvious, which leads to the proportion of employment in the non-tradable product sectors that serve the local area is higher, and these sectors are all in the tertiary industry.
The mediating effect of city size on industrial diversity is relatively high. As mentioned above, there is no consensus on the impact of industrial diversity on urban employment, and a variety of perspectives exist in empirical research. The main reason for this difference is the difference in the size of the study area and its macroeconomic background, which has also been confirmed by previous studies. City size matters when regulating the impact of industrial diversity on urban employment. Industrial diversity has different or even opposing effects for cities of different sizes.
City size also plays a mediating role on the impact of spatial location on urban employment. The vitality of urban employment increases the closer the distance to the central city, if the central city is fixed; but the intermediary effect of the city size appears when the city size variable is controlled, which means that the central city’s employment spillover to surrounding cities is also affected by the size of surrounding cities. The spillover effect of the central city requires a certain degree of carrying capacity for surrounding cities, which is reflected in the labor market and the infrastructure required for economic activities; only large-scale cities have a carrying advantage.

5 Conclusion

It is questionable how applicable the urban employment theory is to developing countries whose urban systems and economies have not reached the level of developed countries, which are based on a mature market economy and high levels of urbanization. This article used data from 278 cities in China to study the impact of city size on urban employment. We found, first, that employment vitality is more concentrated than existing employment and resident population, and eastern coastal regions and central provincial capitals are regions with high employment demand. Second, the size of cities and urban employment demand are significantly positively related, but urban employment demand grows non-linearly as scale expands. Third, city size has important mediating and regulating effects, which is of great significance for understanding urban employment and for formulating policy.
Our research has revealed both the differences and similarities between China and other developed countries in terms of the mechanism of urban employment vitality. The reasons behind these differences, which have been created by China’s development stage, urban-rural relations, and institutional environment, may be shared by many developing countries. China’s continuing rapid urbanization means that a large gap remains between levels of urbanization in China and that in developed countries, resulting in a job market in some less-developed regions largely different from cities in developed countries. Roles differ as well. Smaller prefecture-level cities mainly provide services for urban residents and production services for surrounding agricultural areas, so the highest proportion of tertiary industry occurs in these cities. However, some small and medium-sized cities in developed countries participate in the regional division of production activities, showing high levels of specialization and a high proportion of secondary industries. Additionally, China’s development system is another distinguishing feature, including the control of hukou policies, administrative level bias, and regional bias of resources. The differences in the spatial distribution of China’s employment vitality largely derive from the impact of national development policies. The country’s pioneering development in the east has particularly enabled the eastern coastal regions to obtain significant development advantages after reform and opening up, while central and western regions have provided much of the labor for eastern development, and the northeast has begun to decline under the influence of market economic reform.
In China, the scale of large cities will continue to grow at a faster pace than that of small and medium-sized cities, as opposed to cities of various sizes in some developed countries that maintain the same growth rate. In rapidly urbanizing China, many agricultural populations will migrate to cities, and these immigrants will still prefer large cities because of the influence of market forces. Therefore, the growth of large cities will be faster than that of small and medium-sized cities. The level of diversification of medium-sized cities in the urban system is also different. Diversification of medium-sized cities in China promotes employment, but medium-sized cities are often highly specialized in developed countries. Small and medium-sized cities in China serve as service and production centers for nearby areas and are generally relatively independent, while small and medium-sized cities in developed countries often have production and organizational links with metropolises. Given that these factors are also common in many developing countries, this study is also of significance for urban research and urban policy making in other developing countries.
City size matters when it comes to the issues of employment in developing countries. City size affects the vitality of urban employment in two respects: the advantages of tradable products and the demand for non-tradable products. Due to the agglomeration benefits resulting from the proximity of factor space, metropolises have more efficient production and more frequent technological updates (Glaeser et al., 1992), thereby increasing urban employment and income, whereas the tradable sector has a multiplier effect on the untradeable sector. The productivity and agglomeration benefits of large cities are not comparable to those for small-scale administrative regions, and the ability of these regions to update and innovate is limited. The relationship between urban size and employment is non-linear (Coffey and Shearmur, 1998). The government has always regarded scale as the key to urban regulation. The bias that controlled large cities and encouraged small cities and towns onto the main path of urban system development has been proved wrong because the development of large cities is limited and their benefits are few. However, the mechanisms of the market have played a more important role in economic development after the reform. The various economic advantages of big cities have led to the rapid accumulation of labor, capital, and other means of production in large urban areas. This is also a key to the sustained and rapid growth of employment.
The results of this paper offer practical implications for policy makers. First, large cities should no longer be controlled strictly from rational expansion. As highly dynamic employment centers, the enlargement of the scale of big cities is not only beneficial to the development of the cites per se, but also of great significance to solve the employment problems of the whole country. This is the case especially for the western region where the importance of cultivating core cities is more prominent than well-developed coastal regions. Second, the conclusion shows that the distance to the central city has a significant impact on urban employment, which on the one hand confirms the diffusion effect of the central city and on the other hand confirms the importance of large developing cities. In this sense, the recently proposed national strategy of promoting the development of large metropolitan areas not only deserves more efforts to guaranteeing its effective implementation but also needs more attention be paid to the interlinkage between the central city and surrounding area. Third, given the positive and significant effect of industrial diversity on urban employment expansion, the government should encourage the diversification of cities. For instance, the dual function of cities, production and consumption, can be better balanced rather than emphasizing anyone of them. Fourth, despite the general correlation between investment growth and employment expansion, the employment effect of investment is weak in western China, which reminds local governments to divert the attention from investment attraction to enlarging the job creation ability of investment.
Big data provide new directions and ideas for the study of employment issues; but there are also some limitations that lie mainly in data bias. On the one hand, we used recruitment data derived from only one website to avoid data duplication. Although the 51Job is the largest platform with most comprehensive data, some industries have their special recruitment websites. This is a balance between data missing and duplication which cannot be resolved simultaneously. On the other hand, the share of traditional face-to-face recruitment methods remains high in China especially in less-developed cities, leading to more serious data bias of recruitment websites in these regions.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Abdel-Rahman H M, 1990. Agglomeration economies, types, and sizes of cities. Journal of Urban Economics, 27(1): 25-45.

DOI

[2]
Abdel-Rahman H M, 1996. When do cities specialize in production? Regional Science and Urban Economics, 26(1): 1-22.

DOI

[3]
Adelino M, Ma S, Robinson D, 2017. Firm age, investment opportunities, and job creation. Journal of Finance, 72(3): 999-1038.

[4]
Alder S, Shao L, Zilibotti F, 2016. Economic reforms and industrial policy in a panel of Chinese cities. Journal of Economic Growth, 21(4): 305-349.

DOI

[5]
Ali K, Olfert M R, Partridge M D, 2011. Urban footprints in rural Canada: Employment spillovers by city size. Regional Studies, 45(2): 239-260.

DOI

[6]
Alpkokin P, Cheung C, Black J et al., 2008. Dynamics of clustered employment growth and its impacts on commuting patterns in rapidly developing cities. Transportation Research Part A: Policy and Practice, 42(3): 427-444.

[7]
Åstebro T, Tåg J, 2017. Gross, net, and new job creation by entrepreneurs. Journal of Business Venturing Insights, 8: 64-70.

DOI

[8]
Au C, Henderson J V, 2006. Are Chinese cities too small? The Review of Economic Studies, 73(3): 549-576.

DOI

[9]
Baron R M, Kenny D A, 1986. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6): 1173-1182.

PMID

[10]
Begovic B, 1992. Industrial diversification and city size: The case of Yugoslavia. Urban Studies, 29(1): 77-88.

DOI

[11]
Blomstrom M, Lipsey R E, Zejan M, 1996. Is fixed investment the key to economic growth? Quarterly Journal of Economics, 111(1): 269-276.

DOI

[12]
Busso M, Gregory J, Kline P, 2013. Assessing the incidence and efficiency of a prominent place based policy. American Economic Review, 103(2): 897-947.

DOI

[13]
Cai F, Chan K W, 2009. The global economic crisis and unemployment in China. Eurasian Geography and Economics, 50(5): 513-531.

DOI

[14]
Cai F, Wang M, 2010. Growth and structural changes in employment in transition China. Journal of Comparative Economics, 38(1): 71-81.

DOI

[15]
Chen Y, Funke M, 2009. China’s new labour contract law: No harm to employment? China Economic Review, 20(3): 558-572.

DOI

[16]
Coffey W J, Shearmur R G, 1998. Factors and correlates of employment growth in the Canadian urban system, 1971‒1991. Growth and Change, 29(1): 44-66.

DOI

[17]
Davis S J, Haltiwanger J, 1992. Gross job creation, gross job destruction, and employment reallocation. Quarterly Journal of Economics, 107(3): 819-863.

DOI

[18]
Desrochers P, Leppala S, 2011. Opening up the ‘Jacobs spillovers’ black box: Local diversity, creativity and the processes underlying new combinations. Journal of Economic Geography, 11(5): 843-863.

DOI

[19]
Dong X, 2018. High-speed railway and urban sectoral employment in China. Transportation Research Part A‒Policy and Practice, 116: 603-621.

DOI

[20]
Drucker J, 2015. An evaluation of competitive industrial structure and regional manufacturing employment change. Regional Studies, 49(9): 1481-1496.

DOI

[21]
Dunne T, Roberts M J, Samuelson L., 1989. Plant turnover and gross employment flows in the U.S. manufacturing sector. Journal of Labor Economics, 7(1): 48-71.

DOI

[22]
Duranton G, Puga D, 2000. Diversity and specialization in cities: Why, where and when does it matter? Urban Studies, 37(3): 533-555.

DOI

[23]
Eaton J, Eckstein Z, 1997. Cities and growth: Theory and evidence from France and Japan. Regional Science and Urban Economics, 27(4/5): 443-474.

DOI

[24]
Essletzbichler J, 2004. The geography of job creation and destruction in the US manufacturing sector, 1967‒1997. Annals of the Association of American Geographers, 94(3): 602-619.

DOI

[25]
Fang T, Lin C, 2015. Minimum wages and employment in China. Iza Journal of Labor Policy, 4: 30.

[26]
Francis J, 2009. Agglomeration, job flows and unemployment. Annals of Regional Science, 43(1): 181-198.

DOI

[27]
Fu S, Hong J, 2011. Testing urbanization economies in manufacturing industries: Urban diversity or urban size? Journal of Regional Science, 51(3): 585-603.

DOI

[28]
Gabler L R, 1971. Population size as a determinant of city expenditures and employment: Some further evidence. Land Economics, 47(2): 130-138.

DOI

[29]
Gao H, Lu M, Sato H, 2015. Inclusive urban employment: How does city scale affect job opportunities for different people? Asian Economic Papers, 14(2): 98-128.

[30]
Glaeser E L, Kallal H D, Scheinkman J A et al., 1992. Growth in cities. Journal of Political Economy, 100(6): 1126-1152.

DOI

[31]
Hansen K E H, Winther L, 2018. Employment growth in Danish towns and regions since the crisis: Industrial structure, city size and location, 2008‒2013. Geografiska Annaler Series B‒Human Geography, 100(3): 244-262.

DOI

[32]
Hayes A F, Preacher K J, 2014. Statistical mediation analysis with a multicategorical independent variable. British Journal of Mathematical & Statistical Psychology, 67(3): 451-470.

[33]
He Y, 2018. Foreign direct investment, economic growth and employment: Evidence from China. International Research in Economics and Finance, 2(1): 15-25.

[34]
Henderson V, 1997. Externalities and industrial development. Journal of Urban Economics, 42(3): 449-470.

DOI

[35]
Henderson V, 2002. Urbanization in developing countries. World Bank Research Observer, 17(1): 89-112.

DOI

[36]
Henderson V, Kuncoro A, Turner M, 1995. Industrial development in cities. Journal of Political Economy, 103(5): 1067-1090.

DOI

[37]
Horbach J, Rennings K, 2013. Environmental innovation and employment dynamics in different technology fields: An analysis based on the German community innovation survey 2009. Journal of Cleaner Production, 57: 158-165.

DOI

[38]
Hou J, Huang C, Licht G et al., 2019. Does innovation stimulate employment? Evidence from China, France, Germany, and the Netherlands. Industrial and Corporate Change, 28(1): 109-121.

DOI

[39]
Huang D, He H, Liu T, 2021. The spatial distribution and influencing factors of employment multipliers in China’s expanding cities. Applied Sciences, 11(3): 1016.

DOI

[40]
Huang D, Liu T, Kong F et al., 2021. Employment centers change faster than expected: An integrated identification method and application to Beijing. Cities, 115: 103224.

DOI

[41]
Huang D, Liu Z, Zhao X, 2015. Monocentric or polycentric? The urban spatial structure of employment in Beijing. Sustainability, 7: 11632-11656.

DOI

[42]
Huang D, Liu Z, Zhao X et al., 2017. Emerging polycentric megacity in China: An examination of employment subcenters and their influence on population distribution in Beijing. Cities, 69: 36-45.

[43]
Ianchovichina E, Estache A, Foucart R et al., 2013. Job creation through infrastructure investment in the Middle East and North Africa. World Development, 45: 209-222.

DOI

[44]
Judd C M, Kenny D A, 1981. Process analysis: Estimating mediation in treatment evaluations. Evaluation Review, 5(5): 602-619.

DOI

[45]
Krumm R, Strotmann H, 2012. The impact of regional location factors on job creation, job destruction and employment growth in manufacturing. Jahrbuch für Regionalwissenschaft, 33: 23-48.

[46]
Kucera D, Jiang X, 2018. China and the great trade collapse: Employment effects of falling exports to the EU and US. International Economics and Economic Policy, 15(3): 629-659.

[47]
Kunapatarawong R, Martinez-Ros E, 2016. Towards green growth: How does green innovation affect employment? Research Policy, 45(6): 1218-1232.

DOI

[48]
Lachenmaier S, Rottmann H, 2011. Effects of innovation on employment: A dynamic panel analysis. International Journal of Industrial Organization, 29(2): 210-220.

DOI

[49]
Lee J, Hoshino S, 2017. GIS-based evaluation of spatial interactions by geographic disproportionality of industrial diversity. ISPRS International Journal of Geo-Information, 6(11): 352.

DOI

[50]
Li J, 2013. Job mobility in postreform urban China. Chinese Sociological Review, 45(4): 81-109.

DOI

[51]
Lombardo G, Ravenna F, 2012. The size of the tradable and non-tradable sectors: Evidence from input-output tables for 25 countries. Economics Letters, 116(3): 558-561.

DOI

[52]
Madsen J B, 2002. The causality between investment and economic growth. Economics Letters, 74(2): 157-163.

DOI

[53]
Matsumoto A, Merlone U, Szidarovszky F, 2012. Some notes on applying the Herfindahl-Hirschman Index. Applied Economics Letters, 19(2): 181-184.

DOI

[54]
Merikull J, 2010. The impact of innovation on employment firm- and industry-level evidence from a catching-up economy. Eastern European Economics, 48(2): 25-38.

DOI

[55]
Miller J P, Bluestone H, 1988. Prospects for service sector employment growth in non-metropolitan America. The Review of Regional Studies, 18(1): 28-41.

[56]
Neumark D, Zhang J, Wall B, 2006. Where the jobs are: Business dynamics and employment growth. Academy of Management Perspectives, 20(4): 79-94.

DOI

[57]
Panayiotou A, Medda F, 2014. Attracting private sector participation in infrastructure investment: The UK case. Public Money & Management, 34(6): 425-431.

[58]
Parkinson M, Meegan R, Karecha J, 2015. City size and economic performance: Is bigger better, small more beautiful or middling marvelous? European Planning Studies, 23(6): 1054-1068.

DOI

[59]
Popov A, Rocholl J, 2018. Do credit shocks affect labor demand? Evidence for employment and wages during the financial crisis. Journal of Financial Intermediation, 36: 16-27.

DOI

[60]
Qin D, Song H, 2009. Sources of investment inefficiency: The case of fixed-asset investment in China. Journal of Development Economics, 90(1): 94-105.

DOI

[61]
Rodrik D, Subramanian A, Trebbi F, 2004. Institutions rule: The primacy of institutions over geography and integration in economic development. Journal of Economic Growth, 9(2): 131-165.

DOI

[62]
Storey J, Quintas P, Taylor P et al., 2002. Flexible employment contracts and their implications for product and process innovation. International Journal of Human Resource Management, 13(1): 1-18.

DOI

[63]
Suvankulov F, Lau M C K, Chau F H C, 2012. Job search on the internet and its outcome. Internet Research, 22(3): 298-317.

DOI

[64]
Tang X, McLellan B C, Zhang B et al., 2016. Trade-off analysis between embodied energy exports and employment creation in China. Journal of Cleaner Production, 134: 310-319.

DOI

[65]
Tao J, Ho C, Luo S et al., 2019. Agglomeration economies in creative industries. Regional Science and Urban Economics, 77: 141-154.

DOI

[66]
van Dijk J J, 2017. Local employment multipliers in US cities. Journal of Economic Geography, 17(2): 465-487.

[67]
van Rooy D L, Fairchild Z, Alonso A, 2003. In with the new, out with the old: Has the technological revolution eliminated the traditional job search process? International Journal of Selection and Assessment, 11(2/3): 170-174.

DOI

[68]
Wang T, Chanda A, 2018. Manufacturing growth and local employment multipliers in China. Journal of Comparative Economics, 46(2): 515-543.

DOI

[69]
Wheaton W C, Lewis M J, 2002. Urban wages and labor market agglomeration. Journal of Urban Economics, 51(3): 542-562.

DOI

[70]
Willis K G, 1985. Estimating the benefits of job creation from local investment subsidies. Urban Studies, 22(2): 163-177.

DOI

[71]
Xie Y, 2010. Regression Analysis. Beijing: Social Sciences Academic Press. (in Chinese)

[72]
Zeng C, Zhang A, Xu S, 2016. Urbanization and administrative restructuring: A case study on the Wuhan urban agglomeration. Habitat International, 55: 46-57.

DOI

[73]
Zhang W, Yang D, Huo J, 2016. Studies of the relationship between city size and urban benefits in China based on a panel data model. Sustainability, 8(6): 554.

DOI

[74]
Zhang Y, 2015. Social impacts of the global financial crisis on China and its policy initiatives. China Journal of Social Work, 8(1): 65-77.

DOI

[75]
Zheltenkov A, Syuzeva O, Vasilyeva E et al., 2017. Development of investment infrastructure as the factor of the increase in investment attractiveness of the region. IOP Conference Series‒Earth and Environmental Science, 90: 012122.

DOI

[76]
Zheng S, Sun W, Wu J et al., 2017. The birth of edge cities in China: Measuring the effects of industrial parks policy. Journal of Urban Economics, 100: 80-103.

DOI

Options
Outlines

/

〈 〉