The employment drift of rural migrants: increased employment pressure and lack of upward mobility channels

Zhang et al. (2019) argue that due to the reduction of employment opportunities for low-skilled workers, as well as the absence of upward mobility channels, China’s rural migrant workforce has been compelled to drift in low-end sectors.

Reviewing Zhang et al’s (2019) study on China’s employment system

The Evolution, Friction and Transformation of China’s Employment System: Micro Empirical and Institutional Analysis of the Labor Market

by Zhang Peng 张鹏, Zhang Ping 张平 and Yuan Fuhua 袁富华


Zhang Peng 张鹏, Zhang Ping 张平 and Yuan Fuhua 袁富华 are researchers from the Institute of Economics at the Chinese Academy of Social Sciences 中国社会科学院经济研究. They are further part of the Institute’s internal frontier research group on Chinese economic growth 中国经济增长前沿研究课题组.


Zhang et al. (2019) argue that due to the reduction of employment opportunities for low-skilled workers, as well as the absence of upward mobility channels, China’s rural migrant workforce has been compelled to drift in low-end sectors. This has a negative impact not only on the worker’s income growth, but also on their human capital accumulation. According to the authors, the employment drift is rooted in China’s maintenance employment system through which rural migrant’s wages have been kept at a low-level. Zhang et al. (2019), hence, call for a transition towards a sharing employment system in which migrant workers would in particular be able to benefit from productivity gains.

In-depth analysis

China’s labor force: an abundance of secondary education and shortage of tertiary education

China is faced with the contradiction of having both an abundance of workers with a low-level of human capital and a shortage of workers possessing a high-level of human capital. As the formal education system represent a major channel of acquiring human capital, the authors approximate the level of human capital through the level of education. More specifically, they equate primary and secondary education with a low-level of human capital level and tertiary education (college or higher) with high-level of human capital.

The share of China’s population having received at most a secondary education has since the founding of the PRC in 1949 increased substantially and equals now the levels of developed countries. Strikingly, the proportion of 20-to-39-year-olds with a secondary education as highest education level caught up with developed countries, such as the UK, the US and Japan, as early as the beginning of the 1990s and is nowadays at a higher level [1]. However, on the other hand, the share of China’s population with a tertiary-education (college and higher) lags behind. This is especially striking for the 40-64 years age group – of which only under 10% received a higher education. Reflecting China’s effort to develop its tertiary education capacity, this proportion has for the 20-39 year reached 21,4% in 2015. Nonetheless, the gap separating China from the level of developed countries remains considerable.

In the course of Japan’s and South Korea’s economic development, the labor force upgraded from primary education to tertiary education through two subsequent phases of education generalization. While through the generalization of secondary education, Japan and South Korea transitioned from a labor force with a low-level of human capital towards a medium level, the generalization of tertiary education enabled the upgrading to high-level human capital. Crucially, the development of the tertiary education system was undertaken during the later stages of the generalization of secondary education, when the share of the population with maximally a secondary education was still relatively low. This ensured the smooth upgrade towards a labor force dominated by tertiary education, as well as the provision of a highly educated labor force necessary for the county’s industrial upgrading. The importance of overlapping generalization efforts lies in the long-term nature of educational investments. (Yuan et al., 2016, p.11-17)

While China, on the other hand, successfully transitioned to a secondary education dominated-labor force in the 1990s, it subsequently failed to generalize tertiary education and lacks now in high-level human capital to pursue an industrial upgrade. [2] Yuan et al. (2016) argue that China’s current large proportion of secondary educated workforce represents an impediment to its industrial upgrade.

Rural migrant workers: high employment instability and the lack of upward mobility channels

China’s younger labor force – 20 to 39-year age group – is confronted with a high employment instability. This concerns especially those with primary and secondary education. Based on the data from the 5th wave of the China Household Income Project (CHIP 2013) [3], the authors show that slightly more than half of the 20-39 year-olds with at most a primary or secondary education, i.e a low-level of human capital, have experienced at least one change in their employment. Out of those with a secondary education 36.1% experienced an employment change once, 10.47% twice and 2.52% three times. In contrast, only 34.06% of the 20-to-39-year-olds with a tertiary education underwent a change in employment. This indicates that a higher-education level increases the stability of the employment relationship. (Zhang et al., 2019, p. 6-7)

In some instances, changes in employment have positive long-term effects, as workers might seek occupations with higher wages or additional training opportunities. However, Zhang et al. (2019) demonstrate that this is not the case for low-qualified workers in China. By computing a mobility index and applying it to the CHIP 2013 data set, the authors estimate the probability of young Chinese workers to change industry and measure to which industry they are likely to move. Their results show that overall Chinese workers in the age group of 20-39 years are unlikely to change industry. Moreover, young workers from low-end industries (agriculture, construction, manufacturing and low-end services) are extremely unlikely to change to high-end industries (high-end services as well as science, education, culture and health sector and public organizations 科教文卫与公共组织). Low-end workers are likely to change from agriculture to manufacturing and low-end services, from construction to manufacturing and from manufacturing to agriculture. The authors find similar results when looking at occupational mobility, i.e career changes. While young workers have an overall low probability of changing career, workers in low-end occupations – agricultural-, industrial- and service sector personnel – have almost no prospect of changing into high-end occupations. Consequently, workers with a low-education level, in particular rural migrant workers, are locked in low-end sectors and lack upward mobility channels. (Zhang et al., 2019, p.8-9)

Chinese young workers with a low-level of human capital – primary and secondary education level – are thus faced with both an increased employment instability and the impossibility of advancing into high-end sectors. Zhang et al. (2019) designate this as a ‘drifting state’ 漂移状态 [4]. With frequent changes in employment and no upward mobility channels, the process of human capital accumulation is disrupted.

Employment drift is leading to a negative feedback mechanism on workers income growth

The phenomenon of employment drifting implies that traditional job search models, which are premised on a ‘career-ladder’ whereby workers can accede to high-end sectors by searching for jobs with higher-wages, do not apply to China’s situation. On the contrary, due to the high frequency of employment changes, as well as the lack of upward mobility opportunities, a sizeable share of China’s labor force is drifting in low-end sectors without a chance of ascending to better paid employments in high-end sectors. To account for this Zhang et al. (2019) incorporate an employment drift shock 就业漂移冲击to their own job research model. The employment drift impact or shock is approximated through the number of employment changes: frequent changes in employment – i.e high employment instability – diminishes the company’s incentive to invest in training workers and thus deprive workers of opportunities to acquire skills, i.e accumulate human capital. After a further loss of employment these workers will subsequently be forced to accept work opportunities with low salaries so as to avoid unemployment. This process of employment drifting, hence, results in a negative feedback mechanism – where frequent employment changes lead to the stagnation or even reduction of wages and the dissipation of human capital, which further increases employment instability. (Zhang et al., 2019, p.10-12)

“The inadaptability of this employment system leads to a large number of people drifting in the low-end sector with no available upward mobility channels. The lag in the upgrading of human capital creates a negative feedback mechanism in the employment system.”

(Zhang et al., 2019, p.20)

Zhou et al.’s (2020) extended job search model reflects the existence of such a negative feedback mechanism on the wage of workers whereby an increase in employment changes has a negative impact on worker’s wages. First, the stronger the employment drift shock, the lower the workers’ average wage; second, the probability of a low wage increases with the employment drift shock.

Figure A (left): the y-axis represents the average wage-level (Waverage) and the x-axis the employment drift shock (θ). Figure B (right): the y-axis represents the probability density and the x-axis the wage-level (w).

To further quantify the impact of the employment drift shock on the wage level of young workers, the authors apply their model on the data of CHIP 2013 – more specifically on the 20-39 age group. Their results show that the income growth of young workers, who have experienced a change in employment, is 0.151 %-points lower, in comparison to those with stable employment. Moreover, for every additional employment change, their income growth is reduced by 0.066 %-points. Conversely, the duration of the employment relation has a positive impact on the worker’s income growth, as for every additional year of the employment duration the workers’ income growth increases by 0.318 %-points. This seems to validate Zhang et al.’s (2019) argument that in China employment instability, i.e employment drift, has a negative impact on the workers’ wage level, while employment stability has a positive effect. (Zhang et al., 2019, p.13-15)

“[…] Along with the increase of the employment drift shock, more and more workers are unable to move up the career-ladder through normal job-search channels and drift through low-end sectors by accepting jobs with salaries just above their acceptable minimum. In the end workers are unable to gain career advancements, their channel of human capital accumulation is interrupted, and correspondingly their average wage will continuously decline.”

(Zhang et al., 2019, p.13, transl. by author)

Employment drift is caused by China’s maintenance employment system

For Zhang et al. (2020) the phenomenon of employment drift is caused by China’s transformation from an economic model based on large-scale and low-end industry towards a high-end service sector economy. This economic upgrade, in particular, necessitates a fundamental change of China’s current employment system, which Zhang et al. (2019) have termed maintenance employment system 维持型就业系统.

China’s maintenance employment system is characterized by the large-scale transfer of surplus labor, i.e rural migrant workers 农民工, to urban industries [5]. In the first place, the wage level of the rural migrant workforce has been kept at the subsistence level – covering only the biological needs of workers and their families.  Past the Lewis turning point – where surplus rural labor has been fully absorbed into the manufacturing sector – wage increases were damped due to the lack of institutional bargaining mechanism. China’s maintenance employment system, through which wages of the rural migrant workforce are kept at a relatively low-level, has hence been perpetuated.

In contrast, throughout the development process of the US and several European countries workers have established formal organizations, which enabled them through a process of collective bargaining to take part in the sharing of productivity gains. This subsequently fostered the creation of high-income jobs and the continuous upgrading of the labor force’s human capital. In opposition, China’s relatively weak sharing mechanism hampered the upgrading of its labor force and entrenched the negative feedback cycle between employment instability and wage growth. The authors, hence, define China’s maintenance employment system in opposition to the West’s sharing employment system. (Zhang et al., 2019, p.16-17)

“The problem of employment drift in China’s labor market is not an accidental phenomenon but is part of the process of service sector transformation. The maintenance employment system that is part of the large-scale industrialization stage of economic development is obviously not suitable for the differentiated and innovative development requirements in the period of urbanization. A core component of China’s economic transformation is the transformation from a maintenance employment system to a sharing employment system”

(Zhang et al., 2019, p.20)

At the root of the perpetuation of China’s maintenance employment system lies the long-standing institutional division between rural and urban areas. While starting in the 1990s low-end urban industries attracted large numbers of rural migrant workers, high-end urban sectors absorbed China’s growing educated – high-level of human capital – labor force. Thus, the rural-urban division has been perpetuated in the formation of a dual employment system 二元就业体系 – that is two parallel labor systems: one the one hand, a rural migrant workforce with a low-level of human capital, which is drifting in low-end sectors, on the other hand, an urban workforce with a high-level of capital, which is able to benefit from a career-ladder in high-end sectors. Indeed, Rozelle et al. (2020), show that since the mid-2010s wage growth has been increasingly polarized: whereas wages continued to rise for professionals (high-level human capital) in formal skill-intensive industries, they declined for workers (low-level human capital) in informal labor-intensive industries.

With the slowing of large-scale industrialization and the reduction of employment opportunities in low-end sectors, China’s rural migrant workforce has faced increasing employment pressure. Furthermore, because of the absence of upward mobility channels within China’s dual employment system rural migrants have been unable to move into high-end sectors and thus been drifting in low-end sectors. As Zhang et al. (2019) show, this employment drift has a negative impact on the workers’ wage level and possibly also on their human capital level. This means that China is not only confronted with intensified employment pressure in low-end sectors but also more broadly with difficulties in the upgrading of its labor force. Subsequently, Zhang et al. (2019) argue that the transition away from China’s current maintenance employment towards a sharing employment system represents a central element in the upgrading of its labor force and ultimately in its economic transition towards a high-end economy.

[1] This is true when looking at the maximum level of education. The relatively large share of the labor force in developing countries with a tertiary education is not included in the proportion of the labor force with at most a secondary education. This means that while the enrollment rate for secondary education is higher in the US and UK than in China (see. Worldbank), the share of the labor force with at most a secondary education is higher in China as in most developing countries. Concretely, China’s labor force is composed of around 70% secondary educated workers (maximum education level). For the US and UK this share is only around 60% (Zhang et al., 2019, p.6).

[2] The authors rightly point out that China has undergone a generalization of secondary education. Their analysis, however, neglects to address the existing education inequalities within China, especially with respect to the rural-urban divide. Indeed, while more than 90% of urban hukou holders between the age of 15-17 years attend secondary education, this is only the case for around 70% of rural hukou holders. On the topic of rural-urban education divide see for example Scott Rozelle’s talk at the National Committee on U.S.-China Relations (available here).

[3] The China Household Income Project (CHIP) 中国家庭收入调查seeks to measure the distribution of personal income in China’s urban and rural areas. The data is collected on the basis of questionnaire interviews and includes in particular information on economic status, employment, level of education, sources of income, household composition, and household expenditures. The CHIP-surveys is notably organized by the China Institute for Income Distribution at the Beijing Normal University and conducted with the assistance of China’s National Bureau of Statistics. Up until this point, five waves have been conducted: in 1988, 1995, 2002, 2007 and 2013. For the 5th wave, i.e CHIP2013, see. CIIDBNU.

[4] The authors refer to employment drift, 就业漂移, as the deviation of a part of the labor force from normal job-search channels, due to market frictions, as well as the further deviation from normal paths of human capital accumulation, which subsequently leads to latter’s the partial or complete interruption.

Interestingly, the synonym漂泊 (verb) to drift or 漂泊者 (noun) drifter or floater is commonly used in regard to China’s rural migrant force. For instance, Liang Hong 梁鸿, Professor at the School of Liberal Arts at the Renmin University and Author of several excellent books on the lives of China’s rural migrants, explains: “[…] I think that the greatest characteristic of a ‘fractured society’ is that a whole social class is unable to be integrated into the overall social structure. They [China’s rural migrant population] have been forced to become drifters, forced to become society’s illness and problem.” (transl. by author, Liang Hong, 18 December 2017, speech at the third edition of the One Way Street Book Award 第三届单向街书店文学节, Chinese transcription available here). More generally, the term Beipiao 北漂 – a combination of北京 Beijing and 漂泊to drift – designates those working in Beijing without a local Hukou (household registration). The same applies to Shanghai with Hupiao沪漂, Shenzhen with Shenpiao深漂, or Guangdong with Guangpiao广漂.

[5] While Zhang et al. (2019) fail to provide a clear definition of maintenance employment system 维持型就业系统, it is to be assumed that the attribute maintenance维持型 refers to the maintenance of workers wage at a subsistence level.

The Paper

Zhang, Peng & Zhang, Ping & Yuan, Fuhua (2019). The Evolution, Friction and Transformation of China’s Employment System: Micro Empirical and Institutional Analysis of the Labor Market [English Title].

张鹏, 张平, 袁富华. 2019. 中国就业系统的演进、摩擦与转型——劳动力市场微观实证与体制分析[J]. 经济研究, 2019, 54(12): 4-20.Chinese version available here.


Rozelle, S., Xia, Y., Friesen, D., Vanderjack, B., & Cohen, N. (2020). Moving Beyond Lewis: Employment and Wage Trends in China’s High- and Low-Skilled Industries and the Emergence of an Era of Polarization: Presidential Address for the 2020 Association for Comparative Economic Studies Meetings. Comparative Economic Studies.

Zhang, Peng & Zhang, Ping & Yuan, Fuhua (2019). The Evolution, Friction and Transformation of China’s Employment System: Micro Empirical and Institutional Analysis of the Labor Market [English Title]. 张鹏, 张平, 袁富华. 2019. 中国就业系统的演进、摩擦与转型——劳动力市场微观实证与体制分析[J].经济研究, 2019 , 54 (12):4-20.

Yuan, Fuhua & Zhang, Ping & Lu, Mingtao (2015). The Human Capital Structure During the Process of Long-term Economic Growth [Transl. by author]. 袁富华, 张平, 陆明涛. 2015. 长期经济增长过程中的人力资本结构. 经济学动态, 第5期.