Rural Roots of Current Migrant Labor Shortage in China: Development and Labor Empowerment in a Situation of Incomplete Proletarianization

Abstract

This paper challenges and complements the view, widely held in sociological labor studies, that incomplete proletarianization weakens labor’s bargaining power in the city by allowing capital to externalize the costs of labor reproduction to the countryside. The authors argue instead that the preservation of migrant workers’ links to the rural economy plus rural development measures can, under certain circumstances, empower labor by increasing their marketplace bargaining power. Beginning with the puzzle of migrant labor shortages in China, and based on national data and a case study, the authors show that access to land and pro-rural state policies in the first decade of the twenty-first century together stimulated rural development in hinterland China and created more employment opportunities in agricultural and local nonfarm sectors. As a consequence, rural (migrant) laborers in China were able to rely on rural employment and choose not to participate in labor migration, thus contributing to the labor shortage and pressing employers in the city to increase wage rates and improve working and living conditions.

Introduction

Capitalist development worldwide works in concert with labor migration from rural to urban areas and from less-developed to developed regions. How labor migration interacts with labor’s bargaining power and development has attracted much academic attention. One established view argues that labor migration frequently constitutes an exploitative system that weakens migrant labor’s power generally.

Harold Wolpe was the first scholar to explicitly raise this view. In a 1972 article, he argued that native reserves in the context of apartheid South Africa functioned to subsidize the costs of labor reproduction (such as food, children’s education, and care for the elderly) for capital, thereby allowing capitalists to pay low wages to rural migrant labor in cities. This was a migrant labor system that externalized the costs of labor reproduction and prevented labor from full proletarianization. Michael Burawoy (1976) applied Wolpe’s thesis to the case of labor migration between Mexico and the state of California in the USA. He argued that the costs of the reproduction of Mexican farm labor in California were externalized to the other side of the national border, and thus wages in California could be kept low. Gay Seidman suggested that these studies had contributed to “a new approach to migration” that emphasized labor supply and labor control in capitalist development in various parts of the world (1999: 424–425). The research adopting this approach often suggests that an effective way to increase the bargaining power of migrant labor is to break the migrant labor system and cutoff migrants’ economic links to places from which people migrate. As far as rural-to-urban labor migration is concerned, it is necessary to change the status of rural migrant laborers from semiproletarian (migrant workers) to full proletarian (urban workers).

This approach has also been applied to labor migration and economic development in China. Anita Chan (2003) has argued that cheap migrant labor in China was derived from the hukou system that preserved rural migrant laborers’ links to rural areas and prevented them from settling down in cities.¹ In a co-authored article, she pressed this argument further by likening the hukou system in China to the pass system in apartheid South Africa (Alexander and Chan 2004). Ching Kwan Lee observed that some rural migrant workers returned home from the city when they became unemployed and thus suggested that economic links to the countryside might reduce rural migrants’ labor activism in places of destination (2007: 204–234). Pun and Lu (2010) argue that an unfinished process of proletarianization becomes a major source of frustration among young migrant workers in Chinese cities.

Viewing labor migration as an integral part of capitalist development, this approach has made a great contribution to our understanding of labor regimes in capital accumulation. However, its arguments rest upon two assumptions that are not necessarily true in all historical–local contexts. The first assumption is that economies in the places of origin, such as the countryside, are subsistence economies that have no potential to develop. In the works of Wolpe and Burawoy, the concept of subsistence economy has repeatedly been used in order to show the low level of development in places of origin. Secondly, any power labor might have comes only from labor institutions and labor movements, but these tend to be weak in a migrant labor system.

This paper will show that the two assumptions are not necessarily true. Based on our research on the recent migrant labor shortage and rural development in hinterland provinces of China, we find that the rural economy in hinterland China, from which most rural migrant laborers come, is not a subsistence economy but one that can, under certain circumstances, provide employment for a large number of rural laborers and contribute to rising incomes and welfare. Employment opportunities in the countryside have become so abundant and remunerative in recent years that many rural households can withdraw from the practice of labor migration, leading to migrant labor shortages in cities and coastal regions. Increasing employment opportunities in rural areas have also enhanced rural migrant laborers’ position in the labor market, a kind of labor empowerment that often evades academic attention. These findings thus suggest that incomplete proletarianization in the context of rural development² is also an important way to increase migrant labor’s power, in contrast to more common proposition that labor strength can only be based on the full proletarianization of rural migrant laborers.

In this paper, the term “hinterland provinces” refers to central and western regions in mainland China, and includes the 12 officially designated provincial units of western China and eight provincial units of central China.³ The term “coastal provinces” refers to eastern China and includes 11 provincial units (Fig. 1). In the process of interprovincial labor migration, generally hinterland provinces send their rural laborers to coastal provinces.

Fig. 1

Regional division of mainland China

In the following sections, we will take as a starting point the puzzling phenomenon of the migrant labor shortage in China and discuss the insufficiency of existing explanations. We then introduce the concept labor’s marketplace power based on Erik O. Wright and Beverly J. Silver’s works, on the basis of which we create our explanatory framework. Following this, we discuss, from the perspective of marketplace power, the relationship between rural development and labor migration in China over the past three decades. Finally, we use data from our fieldwork in a labor-sending county in eastern Inner Mongolia (H county), to show how rural development in a situation of incomplete proletarianization can contribute to labor empowerment.

The Puzzling Migrant Labor Shortage in China

The migrant labor shortage phenomenon first emerged as a serious issue in China in the spring of 2004 when almost all coastal regions reported such a shortage. An official study estimated that Guangdong province alone was short of at least 2,000,000 migrant laborers in 2004, and coastal regions taken together could employ 10 % more migrant laborers, that is 8,000,000 to 10,000,000 incoming workers (Jingji Cankao Bao 2004). The shortage increased in seriousness up until late 2008 when the worldwide economic crisis occurred. However, the crisis did not ease the problem for long, and the phenomenon returned as early as the summer of 2009. By the spring of 2010, Chinese media were flooded with reports of labor shortages from all over the country, including in hinterland provinces such as Anhui, Sichuan, and Gansu (Xinhua Net 2010a).

The Chinese migrant labor shortage came as a surprise to social scientists, who had long believed that rural China could provide an unlimited labor supply for urban sectors and industrial expansion for a very long period. The vast size of the labor force in rural China lent apparent credence to such a belief. In 2004, the rural population in China amounted to 757,000,000 of which 487,000,000 were in the active labor force (NBSC 2005).

Demographers attribute the migrant labor shortage to the lowering of the birth rate. For instance, Nielsen and Cai (2007) argue that China’s family planning policy in the last three decades has lowered the birth rate in rural areas significantly and led to a decrease in the number of young rural laborers, who make up the major source of migrant labor. Based on observation of these shortfalls, Fang Cai and his colleagues argued that China would soon reach the Lewisian turning point, that is, the period when labor supply from rural areas ceases to be unlimited (Cai et al. 2007). Demographic data do reveal significant changes in the Chinese rural labor force. Figure 2 shows that rural populations in their 20s and 30s had decreased to a great extent between 2000 and 2009, whereas those in their 40s and 50s increased, suggesting an aging trend among the rural labor force. In addition, the total number of potential rural laborers also decreased from 498,300,000 in 2000 to 469,700,000 in 2009 due both to the lowering of birth rates and to urbanization (Table 1).⁴ However, what demographers could not explain is why rural populations in their late 30s and 40s, whose numbers had changed little in the period, did not migrate to meet urban labor demand. This might be because they were less mobile than younger rural laborers, but they could nevertheless become active migrant populations if they possessed no other employment opportunities. Table 1 shows that rural populations from 35 to 49 years old totaled 181,200,000 in 2009; if this were extended to include rural people in their 50s, the number became 281,400,000. With such a large number of rural laborers, if most of them were to migrate, there would be no labor shortage in Chinese cities.

Fig. 2

The demographic change of rural labor in China between 2000 and 2009. Figure 2 and Table 1 are both constructed based on NBSC (2010d: 6, 45–47), and the 2000 National Census data, 1–7c, available at http://www.stats.gov.cn/tjsj/ndsj/renkoupucha/2000pucha/pucha.htm

Table 1 The demographic change of rural labor in China between 2000 and 2009

Some scholars attribute the labor shortage to the rapid economic growth of the 2000s and labor-intensive industries in coastal regions, which have increased the demand for labor supply from the countryside (Tangen 2010; Zhang and Tan 2005). However, it remains a mystery why the majority of rural laborers choose not to migrate, remaining instead in the countryside. According to the Chinese National Bureau of Statistics, the number of rural migrant workers was estimated at 145,300,000 in 2009,⁵ a large number indeed, but this is still less than one third (31.00 %) of the rural labor force, 468,800,000 in total (NBSC 2010b). That is to say, there were at least 323,420,000 rural laborers working in the countryside in 2009.

It might be argued that institutional barriers such as the hukou system have obstructed the flow of rural laborers to cities, thus causing the shortage. Scholars on the Chinese hukou system have pointed out that residents with urban hukou enjoy more welfare benefits than rural hukou holders. In addition, those who register their households in coastal provinces, usually the places of destination for labor migration, enjoy more benefits than those in hinterland provinces (Chan and Buckingham 2008; Wang 2005). However, closer scrutiny reveals that the hukou system is not the main cause behind the labor shortage. The differences in benefits between the countryside and the city, particularly for large and provincial cities, may prevent or discourage rural migrants’ permanent urban settlement, but they do not obstruct labor migration, which is determined by the accessibility of the labor market in places of destination. Moreover, the benefits associated with the hukou system have significantly declined in the 2000s as compared with previous decades, due either to the removal of certain urban benefits or to the offering of partial urban benefits, such as insurance programs and children’s education, to rural migrants (Huang and Zhan 2005; Zhan 2011). The differences in hukou benefits, including social insurance and government-provided subsidies, definitely exist, but they do not function to discourage migration. In addition, many employment and migration restrictions on rural migrants, often seen in the 1990s, were removed in the 2000s. Governments at both the central and local levels have been encouraging rural laborers to migrate since 2003 by offering skills training programs, employment assistance and wage guarantees (Zhao 2008). Thus, it is hard to explain in terms of institutional barriers why the labor shortage took place in a period when the hukou system was relaxed and the restrictions on migration were largely removed.

Another alternative explanation argues that the labor shortage has been caused by rural laborers’ lack of skills or qualifications (for example, age qualification) for urban jobs. This explanation was particularly popular when the problem of labor shortage first emerged in 2004. For instance, it was argued that the problem was not a shortage of migrant laborers but a shortage of skilled laborers (Mo 2004). However, the subsequent evolution of the labor shortage demonstrated otherwise. The labor shortage problem emerged in almost every sector that relies on migrant labor, including low-skill sectors such as restaurant and hotel services, sales and domestic services. According to data from public employment–service institutions (gonggong jiuye fuwu jigou) in more than 100 cities nationwide, labor demand in commerce and service sectors, which are the main sectors hiring rural migrant laborers and do not require much in the way of skill, exceeded supply as early as 2004 and continued thereafter. In terms of the severity of the labor shortage, these sectors even surpassed sectors demanding skilled laborers such as machine operators (All-China Information Center for the Market of Human Resources Zhongguo renli ziyuan shichang xinxi jiance zhongxin 2011). A lot of sectors that previously only hired young laborers started to relax age restrictions due to the difficulty in finding workers. The data show that people between 35 and 44 years old have also been in short supply after 2010. Shaohua Zhan’s dissertation fieldwork in Jiangsu in 2010 also found that the textile sector, which previously only recruited female laborers in their 20s and 30s as weavers, started to hire women in their 40s.

In short, none of the existing explanations can sufficiently account for the puzzling labor shortage phenomenon. In contrast to these explanations that focus on the demographics of the migrant population, urban labor institutions or the urban labor market, we switch our focus here to changes in rural areas and propose that the shortage is at least partly caused by rural development in hinterland provinces in the 2000s, which provides sources of income and job opportunities for rural laborers. A few scholars have raised similar propositions. For instance, Liu (2008) argues that the migrant labor shortage in China is partly caused by the increase in rural income and urban development in hinterland provinces. He Xuefeng (2011), an expert on the Chinese rural economy, recently commented that the migrant labor shortage was closely correlated with the issuance of pro-rural governmental policies in the 2000s. However, the contention that rural development has contributed to the migrant labor shortage has yet to be proven with solid empirical evidence. Moreover, the processes and mechanisms of the rural roots of the shortage remain largely unclear.

Another important and related issue is that the migrant labor shortage holds salient implications for our understanding of labor’s bargaining power. The labor shortage in China has directly contributed to the increase in wage rates of rural migrant labor. The monthly wages of rural migrant workers increased by 150 %, from 600 yuan in 2004 to 1,500 yuan in 2010 (Xinhua Net 2010b). As we will show below, confronting the difficulty of recruiting enough migrant labor, employers in places of destination have had to increase wages and promise better living and working conditions to attract migrant labor. How do we understand labor’s power in such a context? Why does the Chinese rural economy not function to depress the rates of wages, as other scholars suggest it should, but instead increases them? To answer these questions is our major task below.

Labor’s Marketplace Power and Employment Choices

Erik O. Wright (2000: 962) differentiates two types of workers’ power: associational power and structural power. Associational power results from “the formation of collective organizations of workers.” Examples of these organizations include trade unions, political parties or any other form of institutional representation of workers. Structural power is derived from workers’ location within the economic system. Wright further divides structural power into two subtypes: the first subtype results “directly from tight labor markets” and the second “from the strategic location of a particular group of workers within a key industrial sector.” Beverly J. Silver (2003: 13) calls the first subtype marketplace bargaining power and the second workplace bargaining power. According to Silver, marketplace bargaining power, which is simplified as marketplace power and a main focus of this paper, takes three forms, as follows: (1) the possession of scarce skills that are in demand by employers, (2) low levels of general unemployment, and (3) the ability of workers to pull out of the labor market entirely and survive on nonwage sources of income.

The classification of workers’ power constructed by Wright and Silver is actually a classification of sources of workers’ power, that is, their power comes from three sources, namely associations, the workplace and the marketplace, while these do not in and of themselves constitutes that power. This paper regards workers’ power to be their capacity to make and meet their own demands, including material demands such as higher income, more welfare benefits and better living conditions, and nonmaterial demands such as more political rights. In this paper, we discuss only material demands and leave nonmaterial demands for future research.

Wright and Silver’s classification draws attention in particular to marketplace power. However, this source of power for workers has still largely been neglected in labor studies. There might be two reasons why this is so. One is that a study of marketplace power requires a shift of attention to laborers in general, particularly those who rely on nonwage sources of income, such as peasants and small business owners, whereas labor studies usually focus on proletarianized workers, and even where labor scholars study semiproletarianized labor, they usually examine how this affects proletarianization. The other reason is that the way laborers exercise power in the marketplace in the attempt to get their demands met is fundamentally different from the ways they exercise power in the workplace or through collective organizations. The latter usually involve collective actions such as sit-ins, petitions, strikes, protests, and negotiations, whereas the former usually works through employment choices such as changing jobs, combining different sources of income and opting out of the labor market. Labor studies up to now usually focus on workers’ collective actions while paying little attention to their employment choices.

However, paying attention to nonproletarianized laborers and labor’s employment choices in the marketplace is important. A shift of attention to laborers who entirely or partly rely on nonwage sources of income (for instance, access to land) is important if trying to understand the emergence of various new kinds of labor struggles in recent decades and the possibility of uniting different sections of labor (peasants, informal workers, wage workers, the unemployed, and so forth; e.g., Moyo and Yeros 2005). Our paper will show that making employment choices is also an important means by which labor exercises power, that is, their capacity to make and get their material demands met.

How do employment choices lead to the meeting of labor’s demands, and how is this source of power affected by the conditions beyond one’s skills and qualifications? We propose that employment choices can limit the accessibility of labor to capital in the market, which in turn will press capital to meet labor’s demands in order to attract labor. That is, if a large number of laborers could opt out of one sector or withdraw from the labor market entirely, it would force employers in the sector to improve labor conditions, such as introducing higher pay and better working conditions, to attract labor. However, whether the laborers can opt out of one sector depends on (a) the availability of wage jobs in other sectors or (b) the existence of nonwage sources of income. Moreover, the conditions attached to jobs in other sectors and nonwage sources of income must be comparable to or even better than those in the sector concerned so that laborers are able either to quit or credibly threaten to quit their jobs.

As far as rural labor migration is concerned, if we see labor migration to the city as one sector of employment for rural laborers, rural laborers can exit the sector when they can find other wage jobs in rural areas or rely on nonwage sources of income such as agricultural or non-agricultural enterprise. However, rural wage jobs or nonwage sources of income must be comparable to or even better than labor migration to the city in terms of material rewards and conditions to allow rural laborers to exit that sector. An important way to meet this condition is to develop the rural economy to the extent that rural sectors can provide comparable and even better conditions than labor migration to the city. Figure 3 specifies a general path by which rural development leads to an increase in labor power.

Fig. 3

Rural development and labor empowerment

In the next two sections, we first discuss from the perspective of marketplace power the interrelation between rural development and labor migration in China since the 1980s, and then use a case study based on a labor-sending county in eastern Inner Mongolia to illuminate how rural development in the 2000s contributed to the shortage of migrant labor in the city and an increase in rural labor’s bargaining power.

Rural Development, Labor Migration, and Changing Marketplace Power: 1978–2010

Rural development and labor migration experienced three distinct periods after rural reforms were initiated in China in 1978, and each successive decade, that is, the 1980s, 1990s, and 2000s, generally represent one of the periods. We will use decades as short-hand periodizations even though the cutoff points do not match the beginning/end of decades exactly. For our purposes, the period of the 1980s includes 1978–1979 and 1990–1991. The period of the 1990s, on the other hand, stretches into the early 2000s. We will discuss the three periods in turn and show how rural development in each period affected labor migration and rural labor’s marketplace power.

The 1980s: Rapid Rural Development and Low-Level Labor Migration

The reforms in China, which first started in rural areas, led to rapid rural development in the 1980s. Grain output increased by one third from 300,000,000 to 407,000,000 metric tons, causing China to become a net grain exporter in 1985 for the first time since the Great Leap Forward in 1958 (Naughton 2007: 242). A more dramatic change was the explosive growth of rural industry. Between 1978 and 1991, the gross output of Chinese rural industry grew at an average rate of over 23.0 % a year.⁶

Rural development in the 1980s offered a great number of employment opportunities for rural laborers. The number of people employed in various kinds of rural enterprises, called Township and Village Enterprises (TVEs), increased from approximately 28,000,000 in 1978 to 100,000,000 in 1991 (The Ministry of Agriculture of China 2003: 5), representing about one fifth of the total rural labor force.⁷ More rural laborers found employment in agriculture or other non-agricultural activities. Katie Xiao Zhou (1996: 83–93) has shown that a large number of peasants became actively involved in both local and long-distance trade.

In contrast to rural sectors that provided expanding employment opportunities for rural laborers, labor migration in the 1980s had not absorbed much of the labor force. Before 1984, the volume of labor migration was very small and probably less than 10,000,000 people; after 1984, it had expanded to 26,000,000 in 1988, and may even have contracted between 1989 and 1991 due to the tightening of monetary policy to curb inflation (Zhan 2004). The small volume of labor migration in the 1980s was of course largely a result of the strict hukou system that prohibited peasants from going to the city in search of employment before 1984 (Cheng and Selden 1994). The system was relaxed in 1984 to the extent that peasants could go to the city for employment and business purposes provided they could find food and lodging, rather than relying on state provisions in the manner of urban residents. The policy change, in addition to the marketization of the urban economy after 1984, which made food and other necessities accessible to rural migrant workers, led to an increase in the number of rural migrant laborers. However, the level of labor migration remained low in the 1980s. We propose that this can be attributed in part to the growing employment opportunities in rural areas, in addition to the restrictions of the hukou system. Rural laborers are not forced to migrate to the city when they can find employment opportunities near their home villages. One study found that rural laborers preferred local nonfarm employment to labor migration if the two were comparable in terms of material rewards (Zhao 1999). Huang Yasheng (2008: 122) has shown that the rates of wages for peasants who worked in local industrial enterprises were actually higher than those of rural migrant workers in the 1980s.

In short, rural development in the period of the 1980s slowed the pace of rural labor migration because it provided an employment outlet for the rural labor force, which was near 500,000,000 in the 1980s. This point becomes clearer when compare the 1980s with the 1990s.

The 1990s: Sluggish Rural Development and High-Level Labor Migration

The period of the 1990s also saw rapid economic growth, but this growth was concentrated largely in urban and coastal areas. Huang Yasheng has termed the 1990s the decade of urban capitalism, in contrast to the previous decade of rural entrepreneurship that contributed to rapid rural economic growth (2008: 24–41). The rapid growth in urban and coastal areas generated a large number of employment opportunities, creating the “pull” effect for rural laborers to migrate. An equally important factor was the sluggish growth of the rural economy, which was unable to provide sufficient employment for rural laborers and thus created a “push” effect. TVEs, which were a major employment generator for rural laborers in the 1980s, experienced a short period of expansion and the number of TVE workers had increased to 135,000,000 by 1996. However, the growth was short lived. After 1996, a number of TVEs had either closed down or were in debt; and as a consequence, the number of TVE employees had decreased to 125,000,000 by 1998, subsequently slowly recovering to 132,000,000 by 2002 (The Ministry of Agriculture of China 2003: 73–74). Close scrutiny shows that the decrease took place primarily in hinterland provinces, that is, central and western provinces. The number of TVE workers in eastern provinces continued to grow after 1996, though not as rapidly as earlier. In central provinces the number of TVE workers decreased significantly, and that in western provinces increased little (Fig. 4). Between 1978 and 1991, TVEs in hinterland provinces absorbed 36,000,000 new rural laborers, whereas between 1991 and 2002, the increase was only 16,000,000 employees. That is to say, the capacity of hinterland TVEs to absorb rural laborers more than halved in the period of the 1990s.

Fig. 4

Number of TVE workers by region: 1978–2002 Source of data: China TVEs Statistical Data: 1978–2002, pp.71–74

The agriculture sector performed no better than rural non-agricultural sectors in providing rural employment in this period. Most rural laborers in China are able to cultivate a small amount of farmland,⁸ but the profit generated from agriculture was very little in this period. Many studies have shown that agriculture in the late 1990s was not able to generate much income for rural households (e.g., Chen 2001). As a consequence, the countryside experienced an agrarian crisis in the late 1990s and early 2000s, which Chinese scholars summarized as “Three Rural Problems” (sangnong wenti), literally, the problems of rural production, rural areas and rural people (Li 2002; Wen 2001). Because of this crisis, many rural laborers had no choice but to migrate to the city in search of employment and income.

The sluggish rural development of the 1990s spurred rural laborers to migrate, while at the same time the expansion of public and private investment in cities was generating a great number of jobs. As a result of their combined effects, the number of rural migrant laborers shot up from less than 20,000,000 in 1991 to 40,000,000 in 1992, 60,000,000 in 1994, and about 90,000,000 in 2000. By 2002, the Ministry of Agriculture estimated that the number of rural migrant workers had reached 94,000,000 (Zhan 2004).

The high level of labor migration in the period of the 1990s coexisted with the hukou system, which remained in place to control and even restrict migration. Both central and local governments, particularly those governments in receiving places, used the hukou system to issue and carry out a series of policies resulting in maltreatment of rural migrant laborers, including but not limited to deportation, detention, employment restriction, and imposition of fines. In this regard, urban employers behaved no better than government officials. They hired migrant workers at very low rates of wages and forced them to accept terrible living and working conditions (Alexander and Chan 2004; Solinger 1999: 100–148; 194–240). The situation was indeed little different from the migrant labor system described at the outset of the article, through which the state and capital worked together to control migrant labor. However, what has evaded scholarly attention is that such a system is premised on migrant labor’s low level of bargaining power within the marketplace, resulting in no other choice but to migrate in search of employment. In the case of China, sluggish rural development in the period of the 1990s forced rural laborers to migrate, irrespective of the misery associated with work conditions in urban and coastal areas. As long as employers in places normally attracting migration could find abundant labor, they had no incentive to raise wage rates or improve working and living conditions for migrant laborers. Between 1991 and 2002, the average rates of wages for migrant labor had stagnated at a level of approximately 500 yuan (about 60 US dollars at the time) a month (Tan and Liu 2003: 17), contrasting sharply with rapid GDP growth in the period (more than 9 % a year).

Therefore, the experience of the migrant labor system in the period of the 1990s does not necessarily suggest that proletarianization is the way to increase labor’s power. On the contrary, rural development, which provides employment options alternative to labor migration, can increase rural laborers’ power in the marketplace and allow them to negotiate with employers for higher rates of wages and better living and working conditions in city contexts. Such a path to labor’s empowerment actually strengthens migrant laborers’ status of incomplete proletarianization by offering them nonwage sources of income and rural employment. The changes in the period of the 2000s may, at least to some extent, show the feasibility of such a path, as we will discuss below.

The 2000s: Renewed Rural Growth and the Emergence of Migrant Labor Shortages

Widening regional disparity and the agrarian crisis together prompted the Chinese state to take action in an attempt to head off political instability and economic unsustainability. To do so, it launched two large-scale programs—Develop the West (xibu dakaifa) and Constructing the Socialist Countryside (Jianshe shehui zhuyi xinnongcun)—to offset the negative effects of earlier urban/coastal-centered policies. The program Develop the West, which was put into practice in 2000, aimed to address regional disparity between western and eastern regions. The program offered a wide range of economic support for 12 western provincial units, including but not limited to infrastructure funds, environmental protection funds, and financial assistance. The program Constructing the Socialist Countryside, which started in 2004 in an attempt to relieve the agrarian crisis in hinterland provinces, involves tax exemption/cancelation, agricultural subsidies, social programs in education, and health care and an increase in agricultural funds.

In addition, after 2002, the Chinese central government started to encourage labor migration, seeing it as both a way to relieve rural economic difficulties and a path to urbanization and modernization. As a consequence, previous barriers to labor migration were either removed or weakened and rural migrant laborers received more governmental support than in the period of the 1990s, as noted previously. Probably as a result of these promigrant labor policies, the volume of rural labor migration in official statistics continued to grow in the period of the 2000s. According to the National Bureau of Statistics, it increased from 105,000,000 in 2002 to 145,000,000 in 2009. This is a significant increase, but is still moderate compared with the large size of the rural labor force, 470,000,000 in 2009, and also not as dramatic as the expansion of migration in the 1990s. Moreover, the National Bureau of Statistics adopts a very broad definition of labor migration, seeing rural laborers as migrant laborers if they work for only 1 month outside their home townships. The Ministry of Agriculture uses a narrower definition, regarding rural laborers as migrant workers if they work for at least 3 months outside their townships. Adopting the Ministry of Agriculture’s definition, we can estimate that the number of rural migrant workers in 2009 was only 125,000,000. ⁹ Moreover, despite the belief that and statistics showing there was still a large number of redundant rural laborers in the countryside, the shortage of migrant labor started to emerge and became increasingly serious after 2004.

Our analytical framework, laid out in the previous section, suggests that there is connection between the migrant labor shortage and rural development in the period of the 2000s, which probably derives from the policy shift toward western provinces and rural areas. Between 2002 and 2010, annual per capita net income in rural areas grew 8.8 % a year, much higher than the period of the 1990s (1991–2002), which was 5.2 % a year, though still lower than the period of the 1980s (1978–1991), 10.1 % a year.¹⁰ Close scrutiny shows that the growth in the period of the 2000s came more from rural employment than labor migration. According to the national rural household survey, per capita rural household net income increased 2,577.6 yuan between 2002 and 2009, of which only 20.6 % came from labor migration while 25.0 % came from local wage employment, 38.9 % from local business (including both agricultural and non-agricultural activities) and 15.5 % from other nonwage sources of income (Table 2). In other words, it was rural development that contributed the major part of income increase in the countryside context during the period of the 2000s.

Table 2 Sources of increase in household net income between 2002 and 2009

As rural development was able to provide employment and income for rural households, it was reasonable for many rural laborers to choose not to migrate but to find employment opportunities in the countryside, and this was particularly so for people in their 40s and 50s. This can go a long way to explain why the migrant labor shortage emerged in the period of the 2000s when there still existed a large number of laborers in rural areas. People in the age range in which migration is most common, i.e., in their 20s and 30s, could also refuse to participate in or withdraw from migration if conditions in the cities were not above a certain threshold of acceptability. The existence of employment options was an important factor leading to the labor shortage phenomenon, forcing urban employers to increase wage rates and improve labor conditions in an effort to attract migrant labor. Figure 5 shows the magnitude of wage increases in six sectors that primarily hire rural migrant laborers in China. Between 2003 and 2009, the average wage rates in manufacturing, mining and commerce sectors more than doubled, and those in another three sectors also increased significantly. The data suggest that the rates of wages in high-risk sectors or those requiring skills increased much more than those involving low skill or risk. Nevertheless, the rates of wages in all sectors increased significantly: rates in manufacturing increased most, by 153.8 %, and those in hotel and catering services least, i.e., 57.5 %.

Fig. 5

Wage growth in sectors that hired migrant labor between 2003 and 2009. The figure is based on NBSC (2010d: 29) and is adjusted for inflation

With regards to laborers’ bargaining power within the job market, our analysis above shows that rural development in the period of the 2000s offered laborers employment alternatives to migration, thus increasing their bargaining leverage. As a result, rural laborers were able to force employers based in cities to satisfy their material demands (wages and living and working conditions) by choosing not to migrate or to exit the urban labor market. The question arises as to how rural development took place on the ground in the period of the 2000s, and how this changed rural laborers’ employment choices. In the following section, we use a case study to illuminate these processes and attempt to show concrete linkages between rural development and labor empowerment in a situation of incomplete proletarianization.

Case Study: Rural Development and Employment Choices in H County in the 1990s and 2000s

There is a wide range of local variation in hinterland provinces in terms of rural development, and therefore no one case can represent the entire region. The purpose of our case study is to show whether and how rural development has affected employment options available to rural laborers and their marketplace power in the local context, as a complement to the account of the national situation presented in the preceding section. While the case in question need not be representative, it must possess the following characteristics. First, it must exhibit a history of labor migration occurring after the reform, to show the relationship between rural development and labor migration. Second, the case should follow the general national trend of rural development, as delineated in the previous section. Third, the local economy must be diverse enough to provide rural laborers with both farming and nonfarm-related employment opportunities. As we will show below, H county is a case with such characteristics. Our fieldwork in H county took place between 2005 and 2011. The data collected are such as to best show the changes that occurred in the periods of the 1990s and 2000s, though we also collected data on the 1980s period. For this reason, our case study focuses on changes that took place in the periods of the 1990s and 2000s, to demonstrate the linkages between rural development and rural laborers’ employment choices and marketplace power.

Overview of the Case and Our Fieldwork

H county is located in the far east of Inner Mongolia (Fig. 6).¹¹ The county encompasses 15 townships and 311 villages and its population was approximately 600,000 at the end of 2010, of which 88 % was rural. The labor force consisted of about 350,000 people. Per capita rural household net income in H county was 4,746 yuan (about 750 US dollars) in 2010, lower than the average for central provinces, i.e., 5,510 yuan, but higher than that of western provinces, i.e., 4,418 yuan (NBSC 2011b: chapter 10–28). Per capita farmland in the county is about 5.5 mu,¹² larger than that in most hinterland provinces, which is usually 1–3 mu (NBSC 2011b: chapter 13–12). However, only 30 % of the farmland in the county is irrigated, and the land without access to irrigation can only produce one third of what the irrigated land produces. Apart from farmland, woodland and grassland also provide supplemental sources of income for rural households. The acreage of woodland is almost twice that of farmland, and grassland is also more expansive than farmland. The income from woodland is small because the trees are mainly used for environmental protection rather than for sale in the market. Nevertheless, peasants can at least use some of these trees to build houses. Grassland generates relatively more income than woodland since it provides fodder for animals such as sheep and goats.

Fig. 6

Location of H county

We made five field trips to H county between 2005 and 2011. All of these trips involved interviewing local officials and collecting local statistical documents and lasted 1–3 months. In addition, we investigated five villages—A, B, C, D, and E, located respectively in five townships—V, W, X, Y, and Z. The villages A, B, C, and E rely primarily on farming and animal husbandry, representing most villages in H county. The main undertaking of village D is rural industry (manufacture of fireworks). There are many villages like D in Y township, the only industrial township in H county. Outside Y township, there are only a few villages relying on rural industry. We conducted in-depth interviews with about 80 rural households and 30 township and county officials, on the issues of rural production, labor migration, and the implementation of rural policies. The households we interviewed are diverse in terms of economic activities and population characteristics. More interviews were conducted in villages B and C (about 30 interviews in each village) than in others (20 interviews in total) because we selected these as our main field sites in 2006 and 2010–2011.

In July 2008, we also drew a random sample from the data on all rural labor resources in the county. The County Bureau of Employment conducted a census in the second half of 2007 collecting every rural laborer’s employment information. Officials distributed standard statistical forms to every village team and asked team leaders to fill out the forms.¹³ These statistical forms provided valuable information on rural laborers’ employment choices. We randomly sampled 12 village teams and constructed a database comprised of 546 rural households and 1,387 rural laborers. We went back to three of the village teams to check the data. While local nonfarm employment was underreported, the information on labor migration was accurate. Local nonfarm employment was underreported because team leaders often recorded a person’s employment as being in agriculture when he or she was actually engaged in both agriculture and local nonfarm activities concurrently.

Rural Development and Employment Choices in the 1990s

The rural economy in H county provides local farm and nonfarm employment opportunities for rural laborers. In the period of the 1990s, however, these opportunities were so limited that labor migration became the most important employment choice for these laborers.

Agriculture provides employment for the majority of rural laborers. The land reforms of the 1950s and the contracting of land to individual households in the 1980s have permitted almost every rural resident in the county the chance to attain entitlement to a piece of farmland, and the land is distributed on a relatively equal basis. However, access to land does not necessarily make agriculture a good employment choice for local rural laborers. In the period of the 1990s, agriculture became so unprofitable that it could not even meet basic economic needs of the local rural population. According to national statistics, the net profit of 1 mu of grain or oil crops in 1999 was about a half of that in 1988, and did not increase until the period of the 2000s. By 2009, it had risen to twice as much as that in 1988 (Table 3).

Table 3 Net profits of principal crops in 1988, 1999 and 2009 (yuan per mu)

Our interviews in H county also showed that farming was unprofitable in the period of the 1990s, and that peasants did not want to farm their lands. In all five villages we visited, peasants and village officials told us that farming could not bring much income in the 1990s. The clearest indication that farming was not profitable was the fact that rents for farmlands were very low. The farmlands for rent came from households whose members migrated to the city, or from villages that had surplus collective land. However, the annual rent for 1 mu was only 25–50 yuan for irrigated land and land without access to irrigation was often rented out for less than 10 yuan, or even for free. Mr. Bai, who was the village party secretary of village C between 1991 and 1999, told us that the village rented out a certain amount of collective land at 5 yuan a mu. Even at such a low price, not many people wanted the land.¹⁴

In the period of the 1990s, farming was not alone in bringing low returns. Animal husbandry, which is another important source of income for local rural households, did not generate much income, either. Most rural households in the county raise sheep, goats, oxen, chickens, and pigs. Grass-fed animals such as sheep are particularly important because they can be grazed on grassland and thus do not add extra costs to households except labor, unlike pigs and chickens, which must be fed grain. Our fieldwork in village B in 2006 found that sheep fetched a mere 200 or 300 yuan per head in the 1990s, for a profit of little more than 100 yuan (including labor). The prices of oxen and pigs were also not high enough to attract much attention or investment from peasants.

As in most western provinces, TVEs in H county were not as developed as in coastal provinces. In the late 1980s and early 1990s, the number of TVEs increased in the county. However, this increase soon slowed to a stop as the Chinese state withdrew support for the rural economy in hinterland provinces. According to local statistics, the number of laborers employed in the secondary sector increased by only 2,000 between 1996 and 2000, and that in the tertiary sector by less than 5,000. The total number of workers in the local nonfarm sectors was 71,000 in 2000, approximately 20 % of the total labor force. Urban workers excluded, local nonfarm sectors absorbed 43,000 rural laborers, only 14.3 % of the rural labor force. In villages A, B, C, and E, our fieldwork uncovered that less than 10 % of rural laborers were working in local nonfarm sectors in the mid- to late 1990s. Only in village D did we find that a high proportion of rural laborers (at least half of its labor force) were engaged in nonfarm activities due to the fact that many households were making fireworks, a local tradition that can be traced back to the nineteenth century.

With agriculture being unprofitable and employment opportunities in local nonfarm sectors limited, rural laborers in H county were driven to migration in order to find work. As can be seen from Fig. 7, encircled by large and coastal cities, the county is an ideal place for supplying manpower for urban and coastal areas. In the 1980s and early 1990s, a certain number of male rural laborers migrated to Liaoning province, an industrial base in China, most of them ending up working at manual labor jobs. Based on our fieldwork in villages B and C, we estimated that about 10–20 % of the rural labor force migrated in the period.¹⁵

Fig. 7

Main destinations of H county rural migrant laborers

In the 1990s, outward labor migration grew rapidly, the majority of these migrant laborers going to large cities such as Beijing, Tianjin, Shenyang (the capital of Liaoning province) and coastal cities such as Dalian and Yingkou (Fig. 7). A small proportion went to cities within Inner Mongolia, mostly importantly, Chifeng, and cities in Hebei, Jilin, and Heilongjiang provinces.

In the mid-1990s, coastal cities such as Dalian and Yingkou started to employ rural laborers from H county for their export-oriented manufacturing sectors, particularly textiles and electronics; Beijing and Tianjin also required migrant labor for their expanding service sectors, including restaurants, hotels, and security and domestic services. These new outlets of labor migration changed the composition of the migrant labor force: an increasing number of female rural laborers in the county found jobs in the factories and service sectors in large and coastal cities.

Our fieldwork in villages A, B, C, and E found that about 30–40 % of rural laborers migrated in the 1990s, and that most households sent one or more laborers for labor migration. The interviews showed that the underdevelopment of the rural economy led labor migration to become the most important employment choice for many rural households. Mr. Liu (village C), born in 1968, went to Heilongjiang province with another six fellow villagers in 1991 and worked in a coal mine. He earned about 700 yuan a month but had to cover the costs of food and lodging on his own. He said, “Many fellow villagers migrated 10 years ago, and most of them went to Yingkou (a coastal city in Liaoning province), mainly because farming the land could not make money. Taxes were heavy while the yield was low. And the money from farming was less than that from labor migration.”¹⁶

Migration became so important in the 1990s that, without resorting to this pattern of employment, many rural households could not build houses, afford to see their sons marry, send their children to school, provide for medical care, or even purchase farm inputs such as fertilizers and pesticides. In other words, local rural life could not sustain itself without depending on labor migration. This was the case in all the villages that relied to any great degree on agriculture (A, B, C, and E).¹⁷ Mr. Chen (village B), born in 1982, migrated to both Shangdong province and Beijing to earn money, with which his family was able to build a house and provide for his own marriage in 2003 (the total cost of this was at the time about 40,000–50,000 yuan).¹⁸ Mr. Zhang and his wife (village B) went to Panjin city in Liaoning province in the 1990s and early 2000s to earn money first for the costs of their wedding and also for their two children’s education. They told us that they were not able to save any of the money that they earned and that all earnings were used to repay debt and support their children’s schooling. They had worked as migrant workers for more than 10 years when they returned to local agricultural activity in 2005.¹⁹

In the 1990s, rural laborers chose to migrate to urban centers because of a lack of employment opportunities inside the local economy rather than because they saw it as an attractive undertaking. A lot of interviewees remarked that work conditions for migrants in the period were terrible: wage rates were low and sometimes portions of due wages were withheld; they also faced the high risk of work injury or even death; they usually worked 12 h a day or even longer; living conditions, i.e., lodging and dining facilities in the city were often worse than in the countryside. We have not included their detailed accounts of migration experiences here since the issue has been covered by a number of studies (e.g., Solinger 1999: 100–148).

Between 1999 and 2003, a serious drought took place in H county, forcing more rural laborers to migrate in the period. Among the five villages we visited, A, B, C, and E saw further increases in outward migration in these years, and involved all sections of the rural labor force: young and old, female and male, skilled and unskilled.²⁰ However, this growing outward migration trend had stopped and even began to reverse by 2003 as the local rural economy started to provide more and better employment opportunities.

Rural Development and Employment Choices in the 2000s

In the period of the 2000s, the local rural economy, including both agricultural and non-agricultural sectors, grew rapidly and provided more local employment opportunities for rural laborers. Labor migration, though still important in the period, was no longer the single most important employment option. Farming, animal husbandry and local nonfarm sectors had all become important sources of income, in many cases comparable to labor migration if taking into account human capital such as education and skill.

Outward migration ceased to grow in H county after 2004. In fact, local statistics suggest a possible decrease. The number of rural laborers working in agriculture increased from 164,000 in 2004 to 215,000 in 2007, an increase of 51,000. For the same period, the total rural labor force had expanded by 24,000 persons. This indicates that at least 27,000 rural laborers returned to agricultural production between 2004 and 2007.

Local officials found it difficult to account for this shift because they believed there still existed a large number of surplus rural laborers willing to migrate. For this reason, as noted previously, the County Bureau of Employment conducted a census on the rural labor force in 2007. Our sample drawn from the census shows that among 1,387 rural laborers sampled, 312 worked as migrant laborers in 2007, accounting for 22.5 % of the total.²¹ Rural laborers who work in nonfarm sectors within the county for at least 3 months are seen as nonfarm workers. The data show that 263 people chose to work in local nonfarm jobs, accounting for 19.0 % of the total. The number was probably underreported because data collectors tended not to distinguish local nonfarm employment from agricultural work, as noted previously. In fact, more than half of the rural labor force (58.5 %) chose this as their main undertaking. Agriculture involves animal husbandry and cultivating cash crops, which are labor-intensive and more remunerative than regular farming (i.e., grain crops). Based on our fieldwork in the county, we estimate that at least 30 % of the agricultural labor force (about 17.6 % of the total) is devoted to animal husbandry and commercial crop growing (Fig. 8).

Fig. 8

Rural laborers’ employment choices in H county in 2007

The sample data also show that migrant laborers were mostly male, young, and more educated as compared with rural laborers who did not migrate. This suggests that young, educated, and male laborers were more likely to migrate than others. However, this could also be interpreted as the local economy providing employment options for women, the unskilled, and those who had reached later life stages, so that they were not forced to leave their hometowns to look for work. In the case of H county, this represented 77.5 % of the rural labor force. Moreover, not all laborers that did not migrate were women, the unskilled, or elderly. A significant proportion of them were also men, young, and skilled, and most of them became local nonfarm rural laborers or specialized in animal husbandry or growing cash crops.

The stagnation and even reversal of labor migration in H county had much to do with the rural development that increased and improved employment opportunities in local farm and nonfarm sectors. In agriculture, local peasants reported that farming and animal husbandry were much more profitable than before. As early as 2005, we found that a few households in village A were already expanding the scale of their animal husbandry enterprises and raising more pigs and sheep, due to the rising profitability. The village head told us that farming in 2004 and 2005 could bring in as much as migrating elsewhere to work, thanks to rising prices of agricultural products and good weather.²² Mr. Liu (village B), born in 1965, argued that, in 2006, farming was more lucrative than labor migration. He said,

The economy has become better in the last two years. We depended on labor migration (waichu dagong) in the past, but now this is not as good as farming. Going to look for work somewhere else you can earn eight to ten thousand yuan a year, just as much as if you farm ten mu of land. Moreover, you have to work every day when you work as a migrant worker while you do not need to work that much if farming at home. Of course people are different. Some people may prefer to work outside the home region.²³

Mr. Liu and his wife were farming about 30 mu of cash crops (two hectares), and earned more than 20,000 yuan a year. Including animal husbandry, their income exceeded 30,000 yuan a year, an amount possibly not attainable by migrating to work in a city.

The improvement in agriculture could also be seen from the changes in the rent levels for farmland, as well as the daily wage rate. Migrant laborers who leased out irrigated farmland for others to cultivate could charge at least 200 yuan a mu in 2006, as compared with only 50 yuan in the 1990s. By 2010, rents had increased to 300 yuan a mu. The changes demonstrate the increase in the value of arable land. The daily wage rate also increased remarkably. With the development of farming and animal husbandry, more and more households need to hire people to work on their agricultural land or to take care of animals. The daily wage rate to hire a laborer to do agricultural work increased from 10 to 20 yuan in the late 1990s to 40 yuan in 2006, and further rose to 60 yuan in the summer of 2008, plus in most cases, a decent lunch and a pack of cigarettes. In 2010, it rose to 80 yuan a day in the “busy” season.

The development of local nonfarm sectors also drew an increasing number of local rural laborers. Principal industrial sectors in H county include metallurgy, juice, fireworks and timber manufacture, food processing, and making cement. Industrial output in H county was growing rapidly, from 726,200,000 yuan in 2004 to 1.7 billion in 2007, and further to 3.5 billion in 2009, much more rapidly than in the 1990s (Fig. 9). We estimated conservatively that the number of people employed in industrial sectors was about 40,000 in 2008, increasing more than 15,000 people compared to 2000. Our fieldwork showed that Y township, the most industrialized township in H county and where village D is located, had alone employed more than 25,000 industrial laborers in 2008. In our interview, Mr. Sun, head of the township government office for local enterprises, told us that, after 2007, numbers of rural laborers within the township were not sufficient for the needs of local business, and the township started to recruit rural laborers from surrounding townships.²⁴

Fig. 9

Industrial output of H county: 1996–2007. Sources: H county Statistical Yearbooks 2007 and 2008

In addition to industry, other nonfarm sectors, including but not limited to transportation, construction, rural commerce, restaurants, preschool education, and health clinics, absorbed an even larger number of laborers. Based on our sample data and local statistics, we estimate, again conservatively, that the number of rural laborers employed in nonfarm sectors reached at least 60,000 in 2008, more than double as compared with the late 1990s.

The booming of local nonfarm sectors in the period of the 2000s started to draw migrant laborers back to H county, particularly those who had reached later stages in their lives, such as those in their late 30s and 40s. With savings from labor migration in hand, many of them were able to start small businesses in nonfarm sectors, either running factories resembling those in coastal regions, involving themselves in long-distance trade, particularly between H county and former destinations for migration, opening small stores to trade local products, purchasing trucks or busses to transport goods and people, or expanding the scale of animal husbandry and commercial farming operations. The County Labor Bureau showed that, up to 2007, about 200 returned migrant laborers had invested more than 40,000,000 yuan in the local economy, 200,000 yuan for each person on average. Moreover, this number excluded numerous small-scale investments made by workers moving back from the cities, all of which taken together would amount to a much larger figure.

Rural development in H county in the period of the 2000s induced more rural laborers to choose employment in local farm and nonfarm sectors than in the previous period, and their employment choices, together with those of rural laborers in many other hinterland rural areas, slowed the pace of outward labor migration or even reversed it, thus contributing to the emergence of the migrant labor shortage in large and coastal cities. There is no denying that some rural laborers, despite rural development, have remained and shall remain as migrant workers in coastal areas and large cities. However, a significant proportion of rural laborers—the majority of local nonfarm laborers, some agricultural laborers and some migrant laborers—are changing their employment strategy in response to new opportunities in the rural economy.

The case of H county shows the concrete linkage between rural laborers’ employment choices and the migrant labor shortage. The return of migrant laborers to local farm and nonfarm sectors in H county directly led to the reduction in labor supply to places typically chosen as a destination for migration: large cities such as Beijing, Tianjin, and Shenyang, and coastal cities such as Yingkou and Dalian. Mr. Wang (Z township), born in 1973 and a skilled migrant laborer who had worked in Dalian for more than 10 years, returned to Z township and opened a textile factory in 2006. One of the reasons for his return was that he considered it would now be easier to recruit workers back in his home region, as the factories in Dalian faced a labor shortage. At time of interview in July 2008, his factory was still short of more than 100 workers; he planned thus to hire more migrant workers when they returned home for the Chinese New Year.²⁵ His case suggests that there exists competition for rural labor between sectors such as agriculture and rural industry in H county and factories in large and coastal cities. Y township, the industrial township referred to above, also faced labor shortages in 2008 and was attempting to recruit rural laborers from surrounding townships. It seems logical that some of the laborers in these other townships might consider it more attractive to work there rather than migrate to large and coastal cities. In the meantime, we also found that rural households in villages B, C, and D also required extra workers to work on their farmlands or take care of their animals.

Increased employment options and competition among employers for the existing labor pool suggest increased bargaining power of local rural laborers within the marketplace, forcing employers to increase wage rates and improve living and working conditions. In H county, we have shown that there has been an increase in the daily wage rate. In addition, the average rate of wages in local rural industry sectors had also increased to 800–1,000 yuan a month in 2008, double the approximately 400–500 yuan typical of the early 2000s. That is to say, the rates of wages in large and coastal cities need to be higher than such levels because of higher living costs; otherwise, employers in these cities are unlikely to be able to recruit sufficient labor from H county.

The most direct evidence that employers in large and coastal cities have had to increase wage rates to attract rural labor from H county is the changes in their recruitment advertisements and attitudes toward local labor officials in H county. When we visited the County Bureau of Employment in September 2005, local officials were still worried about how to send out more rural laborers to cities such as Beijing, Tianjin, and Dalian. Mr. Yang, who was in charge of this sending out of labor (laowu shuchu) at the time, told us that he used to have to solicit employers in large and coastal cities to hire local rural laborers.²⁶ However, starting in 2006, the bureau began to be flooded with recruitment advertisements from large or coastal cities offering higher wage rates and better living and working conditions for migrant laborers. Ms Wu, who had worked for the Bureau since 2004, told us that she felt a reversal in the typical relationship with employers from those cities. She said, “Before, we used to beg them to hire our laborers, but now they beg us to send them our workers.”²⁷ The situation had remained the same when we interviewed her again in December 2010. However, while employers in large cities offered higher wage rates and better living and working conditions after 2006, it had become more difficult for local officials to convince rural laborers to migrate.

State Policy, Access to Land, and Rural Development

Why did the rural economy in H county experience significant growth in the period of the 2000s, whereas it had remained stagnant and even slumped into a crisis in the 1990s? The answer has much to do with changes made in state policy: Policies in the period of the 1990s constrained rural development, whereas those in the 2000s stimulated it.

The first set of policies to compare is those pertaining to agriculture and rural public services. In the 1990s, peasants in H county, as in other parts of rural China, had to pay taxes and fees for cultivating their farmland and raising animals. These regulatory burdens included a village administration fee, a township administration fee, farmland tax, tax on commercial crops, tax on selling animals, and tax on growing commercial trees. In addition, local authorities and particularly township governments, often collected fees from rural households ostensibly for building of public works or for other purposes. A conservative estimate suggests that per capita payment of these fees and taxes nationwide amounted to 156.6 yuan a year in 1999, 7.2 % of peasants’ net income (Zhao 2007). In some regions, the amount was as high as 300 yuan or more. According to our fieldwork, nearly all rural economic activities were charged taxes or fees in H county during the 1990s. For example, peasants had to pay fees for digging up wild sweet grass roots for sale, growing fruit trees in their own yard, and grazing sheep on hills (literally called fee for treading hills). Although peasants were each given a piece of farmland, they could not obtain much economic benefit for cultivating the land due to these constraining policies. As a consequence, an increasing number of peasants gave up their farmlands, choosing instead outmigration in search of work. The interviewees in villages A, B, C, and E reported that migrant laborers in the 1990s had to seek out fellow villagers to cultivate their lands so that the latter could pay the agricultural taxes and fees for them. Another force that pushed peasants to leave the countryside was policies on health care and children’s education. Due to the marketization of education and health care, peasants had to pay an increasing amount for these services. Unable to earn enough cash from the rural economy, more and more peasants had to migrate to large cities and coastal regions in order to afford these services.

In contrast, policies in the period of the 2000s stimulated the rural economy. Following the policy initiative of Constructing a Socialist Countryside, the central government used the No.1 Documents, which were issued at the beginning of every year and were seen as a symbol of policy priority, to address rural problems in seven consecutive years from 2004 through 2010. The most important of these policies included eliminating agricultural taxes and fees, providing subsidies for peasants who grow grains, increasing agriculture funds, and improving rural infrastructure such as roads and irrigation facilities. With no regulatory fees now required to be paid, at the same time that agricultural subsidies and funds were being applied, agriculture in the 2000s became much more profitable than in the 1990s. The increase in agricultural funds available also stimulated the rural economy in H county. For example, village B received an agricultural grant in 2003 and used it to drill 13 tube wells, which extended the acreage of irrigated farmland by about 1,000 mu. Peasants in the village were thus able to grow high-value commercial crops such as tobacco, corn seeds, and sugar beet on their farmlands. In addition, the central government also increased financial support for rural health care and education. Since 2003, the government has started to establish a new system of rural medical cooperation in the countryside, which allows rural households to be reimbursed for a portion of their medical expenses from government funds. The government also launched in 2004 subsidization of school children’s costs, including tuition, textbooks, and accommodation in rural schools. These policies relieved the pressure on rural laborers to earn cash income through labor migration to pay for medical expenses and children’s education. Nearly all interviewees in the four villages supported these pro-rural, propeasant policies. Mr. Li, born in 1956, was an example. He regarded the changes in state policy as a major reason for the economic improvement of his household. He remarked, “State policies are much better now than before. We had to pay all kinds of taxes and fees before 2002, and we were left with little after these levies. Now not only are these levies gone, but also we are given subsidies for farming our lands.”²⁸

Another set of state policies that stimulated the economy in H county came from the Project known as “Develop the West.” As a part of Inner Mongolia, one of the 12 western regions, H county was eligible to receive a number of funds from the project, including a grant for rural poverty reduction, one for environmental protection and another for infrastructure improvement. These funds all had the effect of promoting rural development. For example, the grant concerning environmental protection provided compensation for rural households that planted trees on their farmlands and for those that fed their animals on a ranch for a period of months instead of grazing them on the open grassland all year long. As a result, land was enabled to generate a secure and often higher income for rural households. Another important effect of these funds was that they provided more local nonfarm jobs for rural laborers. For example, with the funds from the central government, the county was able to start a number of large infrastructure projects in the 2000s. The amount of investment in fixed assets started to rise rapidly after 2002: local statistics show that this increased rapidly up to 3.4 billion yuan in 2007, ten times as much as that in 2002. The increase in investment in such assets provided rural laborers with many more local construction jobs than had been available before. In addition, with the inflow of financial resources, it became easier for rural laborers to find employment opportunities in trading, transportation, and other service sectors within the county.

It should be pointed out that the positive effects of state policies in the period of the 2000s on rural employment are also related to the current landholding system. As noted previously, the Chinese land system provides peasants with relatively equal access to land.²⁹ As a result, the supportive policies on agriculture and rural public services were able to benefit most rural households and stimulate labor-intensive small-scale agriculture. If access to land were unequal, the benefits might concentrate around a small number of large-scale farmers and could not create such a large number of employment opportunities. In the case of H county, small-scale agriculture based on this land system absorbed nearly 180,000 people in 2008, including many rural laborers in their 40s and 50s. Our ethnographic fieldwork in H county found that access to land allowed rural households to increase labor input into farming and animal husbandry when agriculture was profitable.

Access to land also helped rural laborers take advantage of growing local, nonfarm employment opportunities. Rural laborers who are engaged in local nonfarm employment can still receive most of the benefits of the lands and assets attached. Our fieldwork showed that most of them return home on a daily or weekly basis; thus, they can take meals at home and stay in their own houses at night. They are also able to assist their families with farming or family business activities after work. In addition, local nonfarm rural laborers are able to take care of the elderly and children themselves, thereby reducing the costs of these services. In other words, rural laborers working in local nonfarm sectors have preserved their status of incomplete proletarianization and have been able to maximize the benefits from both agricultural land use and wage-based jobs/non-agricultural enterprise. For this reason, many rural laborers choose to work at nonfarm-related jobs even though wage rates for this are lower than those gleaned from migrating to cities.

In summary, the changes H county has undergone demonstrate that, in the 2000s, rural development of hinterland provinces in general provided rural laborers with more and better employment opportunities, leading to a rise in their bargaining power within the marketplace. As a result, rural laborers have been able to choose not to participate in, or to exit, the labor migration path, causing the shortage of migrant labor in large and coastal cities. Employers in the normal places of destination for migrants have thus been obliged to increase wage rates and improve workers’ conditions to attract migrant labor. In this process, rising marketplace power has enabled rural laborers to get their material demands met. The rural development was made possible by pro-rural, pro-hinterland policies in the 2000s as well as being derived from the landholding system.

A further Discussion of Labor’s Power, Development, and Proletarianization

Our analysis above does not deny the existence of the migrant labor system as a mechanism that regulates the flow of migrant labor and serves to subsidize labor reproduction for capital as Wolpe, Burawoy, and Chan have described. We agree with Seidman that such a migrant labor system exists in various parts of the world, including China in the 1990s. However, it might not be true that labor migration in China could be called a migrant labor system before and after the 1990s. Moreover, it does not necessarily follow that migrant labor must be proletarianized in order to achieve either empowerment or development. Our findings suggest that the preservation of workers’ links to the rural economy, that is, the preservation of their status as incompletely proletarianized, plus efforts put into rural development, are also important avenues by which the power of migrant labor and of rural labor in general is increased.

Proletarianization as a road to labor empowerment comes with risks attached, and might not always produce the desired effects, particularly in developing countries. A major difference between semiproletarianized and fully proletarianized workers is that the former can withdraw from the labor market and live on nonwage sources of income while the latter cannot. For this reason, proletarianization can result in overcrowded urban labor markets and sharply decrease labor’s marketplace power relative to employers because they do not possess any realistic “exit” choice. This is particularly true in populous countries like China, India, and many African countries. Philip McMichael (2006) has argued that the dismantling of the peasantry and of small farms has the potential to nourish global capitalism with an endless supply of surplus labor that then depresses wages. On the other hand, the bargaining power of proletarianized labor depends much on labor’s organization and labor movements in the city, but these two factors are both historically and contextually contingent (Silver 2003).

This being said, nevertheless, merely incomplete proletarianization is insufficient to increase the position of rural migrant laborers in the marketplace. As this paper has shown, migrant laborers in 1990s China had low marketplace bargaining power even though they were not proletarianized. To increase rural migrant labor’s ability to negotiate, other conditions must be satisfied. First of all, land rights of peasants and migrant laborers should be secured.³⁰ As we have shown, access to land has made it possible for rural labor in hinterland China to live on nonwage income and helps them take up nonfarm employment opportunities. Second, pro-rural policies or projects should be drafted and implemented to foster rural development. Only in this way can a rural economy provide rural labor with employment opportunities that are sufficient and comparable to the benefits of labor migration. Finally, the development of labor-intensive rural nonfarm sectors is also necessary. Due to decreasing returns from agriculture, the rural economy must provide other more remunerative outlets for rural employment. As we have shown, labor-intensive rural nonfarm sectors in China have played a crucial role in generating employment opportunities for rural labor.

Another important issue is how labor’s power within the marketplace is related to labor activism, and this is in fact the focus of most labor studies. We suspect that migrant laborers with high marketplace power might be more likely to become involved in labor unrest and other movements because they are less afraid of being laid off due to the availability of jobs in the countryside and elsewhere. That is to say, labor’s marketplace power and other sources of power may be positively correlated. Increasing waves of migrant labor unrest in China in recent years might in part be attributable to increased marketplace power of migrant laborers. Although we have not systematically investigated this matter, Shaohua Zhan’s 2010 fieldwork in Jiangsu, a coastal province that receives migrant laborers, found that migrant workers were more willing to use strike action and other means of fomenting unrest to make demands on employers than they had been 5 years ago, because it was much easier for them to switch jobs in situations of labor shortage, or simply to return to their home towns. Understanding the relationships between marketplace power and other sources of power requires further research.

Conclusion

The current migrant labor shortage in China carries important implications for issues pertaining to development, the power of labor, and proletarianization. Through the investigation of Chinese rural labor’s employment choices, between local farm and nonfarm employment and labor migration, and how these are related to rural development and labor shortages; this paper draws the following tentative conclusions.

First and foremost, incomplete proletarianization plus rural development is a viable path to labor empowerment and economic development. This is particularly so for many developing countries where a large population lives in rural areas. Rather than workers cutting off their links to the rural economy and proletarianizing rural (migrant) labor, protecting their rural assets, especially land, and promoting rural development could provide them with more employment options and thus more bargaining power with employers.

Second, rural development in hinterland provinces from which emanates the bulk of migrant labor has been an important factor leading to the current migrant labor shortage in China. As the rural economy in the 2000s provided more remunerative employment opportunities in agricultural and local nonfarm sectors, rural laborers held back or withdrew from former labor migration patterns, contributing to the emergence of migrant labor shortages in large and coastal cities. The availability of employment in the countryside increases migrant laborers’ bargaining power in the city because they can press employers to meet their demands by applying the threat to return home and work in agricultural and local nonfarm sectors.

Finally, both state policies and the landholding system have been among the primary factors affecting rural development and contributing to the migrant labor shortage in China. In the period of the 1990s, the Chinese government constrained rural development by focusing its efforts on large cities and coastal provinces. In the 2000s, however, it issued a series of pro-rural policies, as well as much more support for western provinces than in previous periods. These policies stimulated rural development and created employment opportunities alternative to labor migration in agriculture and local nonfarm sectors in the hinterland, thereby strengthening rural (migrant) laborers’ marketplace power to the extent that they could choose not to migrate to cities, forcing urban employers to offer higher wages or other benefits to attract rural labor.