WPS3890 Migration in Towns in China, a Tale of Three Provinces: Evidence from Preliminary Tabulations of the 2000 Census Anqing Shi* Development Research Group The World Bank Abstract There is a concern that the growth of towns in China has been stalled recently and with it, the creation of non-farm jobs in rural industries. This study uses the 2000 census tabulations to look at this issue by examining in-migration in towns in three provinces in China, Zhejiang, Henan, and Sichuan, their educational attainment, original place, and occupational composition. In addition to the diversified patterns of town in-migrants revealed in the three provinces, the paper finds that town in-migrants generally possess a higher level of educational attainment than the local population in towns, especially in the less-developed western and central regions. This inflow of human capital could foster development in towns. There is also evidence that as economic opportunity increases in towns, such as in the richer coastal province of Zhejiang, better educated people in rural areas are likely to shift their jobs from the farm to the non-farm sector in towns nearby, instead of leaving the countryside to migrate to other provinces. This could reduce migration pressure on big cities. Finally, the labor market in towns in the less-developed west and central regions is more flexible in accommodating in-migrants, whereas in the developed province of Zhejiang the labor market is segregated between migrants and the local population. Key words: migration, town development, Chinese census, labor segregation World Bank Policy Research Working Paper 3890, April 2006 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. *I thank Bert Hofman, David R. Dollar, Luis Serven, and Songsu Choi for very helpful comments on the earlier version of this paper. 1. Introduction There has been a remarkable growth in small and medium towns in China over the past two decades. Traditionally, scholars have long characterized the pre-reform rural communities as essentially self-sufficient, inward-looking, and isolated. Their economic structure was described as "cellular" and "fragmented" (Skinner 1964, 1985) based almost exclusively on farming, with towns playing only a supporting role for agricultural production.1 In the modern times, these more dispersed centers are still critical in helping develop and modernize agriculture by providing both consumer and producer services from banking to marketing, and basic services such as health services and educational facilities (Johnson 1970). The market reforms in the late 1970s and early 1980s provided changes in institutional environment conductive to the growth of towns. Some changes such as increasing foreign and domestic investment (Chen et al., 1995; Fei 1986; Wei et al., 1999), regional development strategy in favor of more developed coastal region, and massive rural-urban migration (Ma 1996; Wei 1997) have great impact on the growth of town and village enterprises (TVEs), which propelled the town development. Town development in China represents a large economic and social agenda. By the year of 2000, there were about 20,000 small towns, with a population of 166 million (NBSC 2002). It plays an important role in absorbing excessive rural labor force, expanding rural markets, raising rural incomes, and facilitating economic growth in urban centers. There has been a concern that the growth of towns has been stalled recently and with it the creation of non-farm jobs in rural industries, although there is little agreement why and how. Likely explanations include financial and institutional barriers, product cycles, lower stock of human capital, labor market barriers, and the lack of agglomeration economies. These issues have further been compounded with recent problems in environment and land, as well as regional income disparity. This paper explores the issue of town development in China by examining town in-migration in three provinces, Zhejiang, Sichuan, and Henan. Specifically, we center around three issues: current status of town in-migration, human capital stock of in- migrants, and labor market barriers. Three provinces represent different macroeconomic conditions in China: Zhejiang is located in the east coast, Henan in the central region, Sichuan in the western region, with Sichuan and Henan legging economically far beyond Zhejiang. We mainly rely on the three volumes of aggregated statistics from tabulation of 2000 population census data released, supplemented by electronic version of detained tabulations for towns released by the National Bureau of Statistics recently (NBSC 2002). Both short form and long form (9.5% of total population) asked the questions about migration. In the published tables, in-migration into towns is cross-tabulated by some 1For a historical treatment of this issue, see Huang 1990. 2 social and economic characteristics such as educational attainment, occupation, and region, which provides us with an excellent opportunity to look into the issue of migration into towns and its determinant and consequence.2 Specifically, we focus on three issues: (1) Does migration into towns stall? (2) Has development in town absorbed more educated people? Do these in-migrants in towns come locally or from other provinces? (3) What is the consequence of in-migration on labor market in towns? Does labor market in towns become more flexible to accommodate highly skilled labor or labor market remains segregated between migrants and non-migrants? The paper is organized as follows. In the next section, we briefly discuss policy on town migration and provide an overall view of town in-migration in China. Then, we discuss population composition, employment opportunities and other economic conditions in towns in three provinces and patterns of town in migration. It is followed by examining the in-migrants' differentials in education and the impact of in-migration on town labor market. The paper is concluded with a summary and some policy recommendations. 2. Brief view on town migration in China Migration used to be an area of heavy state control, and conventional hukou system classifies all the people as either "agricultural population" or "non-agricultural population". The conversion from the one status to another was subject to strict quota controls. With market reforms in 1978, tremendous non-agriculture employment opportunities have been created in town and village enterprises (TVEs), attracting large migrants form rural areas. Grains and almost all kinds of commodities have been available from markets at reasonable prices, which essentially enabled people to choose where to work and reside despite their hukou status. The migration policy became increasingly flexible as reforms proceeded (Chan 2001). A "self-supplied food grain" hukou, a new type of urban hukou in rural market towns, was issued to farmers who ran business or undertook industrial or commercial activities in rural market towns. Later, a more formal urban hukou was issued to those people who had stable non-agricultural jobs and accommodations in selected cities and designated towns. Thus, a more flexible migration policy could have facilitated migration into towns. Data on national migration flows (Table 1) tend to speak for the possible impacts of migration policy and development of TVE. It shows that the migration into towns was at its highest during the period of 1982-87; of total migration population, about 40% of them went to the towns. This was probably due to booming of TVEs in the rural towns in the late 1970s and earlier 1980s. The in-migration into towns started to diminish since 2In the time of completing this paper, author does not have access to the 1 per 1,000 micro-data set of long form questionnaire, and will do more empirical testing whenever data are available. 3 the middle of 1980s, and reached its lowest level in 1990-1995; during this period migrations into cities increased significantly thanks to urban market reforms initiated in the middle 1980s and flexibility of urban hukou system (Yang and Zhou 1999). During the period of 1995-1996, the migration into towns started to pick up, reaching 19 percent. This recent increase in the flow of migrants into towns probably has been due to the more favorable migration policy toward small cities and designated towns residents by granting them urban hukou, and the increase in FDI in many small towns, especially in the coastal areas. Table 1 Percentage of flow of in-migration to cities, towns and rural townships. place of destination 1982-87 1985-90 1990-95 1995-2000 ________________________________________________________________________ (%) city 36.6 61.7 61.4 59.4 town 39.8 20.1 10.0 19.16 rural 23.6 18.2 28.6 21.43 Source: Data for the period of 1982-87, 1985-90, and 1990-95 are from Yan (1998), data for the period of 1995-2000 are derived from the tabulation of 2000 population census data (the long form of 9.5% of census sample data). 3. Economic opportunities in towns in three provinces Before we elaborate the town in-migration in these three provinces, we present the population composition and employment opportunities in towns in these three provinces. Towns are well developed in Zhejiang, and less so in Henan and Sichuan. According to census data of 2000 (Table 2), there were 20.15 percent of population living in towns in Zhejiang province; in contrast, there was only 8.31 percent of population living in towns in Henan, and 12.26 percent in Sichuan. Zhejiang is also more urbanized than other two provinces. There was 28.5% of population living in cities whereas there was only 15% of population living in cities in Henan and Sichuan respectively, even lower than national average; majority of population, three quarters, lived in rural areas in Henan and Sichuan. Table 2 City, town, and county population in three provinces in China (in million): 2000 Province total city town county (percentage in parenthesis) Zhejiang 45.93 13.10 9.25 23.57 (28.53) (20.15) (51.33) Henan 91.24 13.80 7.58 69.851 (15.13) (8.31) (76.56) Sichuan 82.35 12.21 10.10 60.04 (14.83) (12.26) (72.91) National 1,242.61 292.63 166.14 783.84 (23.55) (13.37) (63.08) ________________________________________________________________________ Source: NBS "Tabulation on the 2000 population census of the People's Republic of China. Table 1-1, 1- 1a, 1-1b, 1-1c. 4 The data (Table 3) further suggest that manufacturing becomes a leading sector in towns in Zhejiang, whereas in other two provinces agriculture remains the leading sector. About 40 percent people employed in manufacturing sector in Zhejiang, whereas in Henan and Sichuan, there were only 12 percent and 14 percent employed in the manufacturing sector respectively. But, retail and hotel sector seem well developed in these two provinces. This could also have created more job opportunities. It is also interesting to note that the percentage of people employed in the local government institutions were higher in the less developed regions such as Henan and Sichuan than in the developed region of Zhejiang. 3 Table 3 Employment by industrial sector in towns in three provinces: 2000 Province 1 2 3 4 5 6 7 8 9 (%) Zhejiang 16.54 6.76 39.50 4.74 17.38 1.27 4.59 5.60 3.54 Henan 43.21 5.37 11.92 4.36 12.68 1.73 2.64 8.44 9.46 Sichuan 32.50 6.55 14.07 5.33 18.12 1.66 4.32 10.00 7.22 Sources: calculated from Tabulation on the 2000 Population Census of the P. R. China, Table 4-1b, page 989-1042. Note: 1, agriculture; 2, construction; 3, manufacturing; 4, transportation, and telecommunication; 5, retail and hotel; 6,finance; 7, services; 8, health, education, and research and development; 9. government institution. The living conditions are also much better in towns in Zhejiang than in other two provinces. Table 4-1 presents the availability of tapped water and composition of fuel resources for cooking in households in towns in three provinces. Data show that in Zhejiang and Sichuan, majority of households in towns were equipped with tapped water, whereas in Henan only about half of town residents had access to the tapped water, which was lower than the national average of 70 percent. Furthermore, most of households in towns in Henan (two-thirds) used coal for cooking, about twice as much as likely the national average (one-third). The percentage of using coal for cooking in Sichuan was slightly higher than the national average. In Zhejiang, gas became the major fuel source for cooking, which was about twice as much as likely the national average (41%). Nationwide, a little over 20% of households were using wood/grass as their cooking fuels in towns in China, and the level of using wood/grass as fuels for cooking were found to be lower in all three provinces examined in this paper. 3Scholars have emphasized that at the local town level there was a "local state corporation" in parts of rural China, an entity which related local government closely with enterprises and people (Oi 1992, 1999). This might explain why some of rural towns had higher share of population working in the local government institutions. 5 Table 4-1 Percentage distribution of tapped water and fuels for cooking in towns in three provinces in China: 2000 _______________________________________________________________________ province tapped water fuel resources for household cooking yes no gas elect coal wood/grass others ______________________________________________________________________ (%) Zhejiang 84.17 15.83 80.70 0.37 2.68 14.59 1.66 Henan 56.53 43.47 11.56 0.21 76.51 11.25 0.46 Sichuan 72.15 27.85 36.06 10.02 38.04 14.42 1.46 National 69.32 30.68 40.97 2.66 34.82 20.46 1.10 _______________________________________________________________________ Sources: calculated from Tabulation on the 2000 Population Census of the P. R. China, Table 8-3b, page 1859-1860. Similarly, there were wide gaps in bath facilities among three provinces, although the difference in kitchen availability in households appeared small. In Zhejiang, about half of households (48%) in towns did not have bath facilities, whereas in Henan, three- fourths of households did not have bath facilities (Table 4-2). Table 4-2 The percentage distribution of kitchen and bath facilities in towns in three provinces in China: 2000 ________________________________________________________________________ province kitchen bath facility yes no yes no own share with central own other others heatingunit ________________________________________________________________________ (%) Zhejiang 87.90 2.45 9.65 0.44 38.70 12.48 48.39 Henan 84.12 0.84 15.04 0.60 17.24 6.31 75.84 Sichuan 86.11 2.12 11.78 0.63 31.96 11.41 56.00 Sources: calculated from Tabulation on the 2000 Population Census of the P. R. China, Table 8-3b, page 1859-1860. 4. Migration into towns Migration is quite responsive to the economic opportunities and conditions in towns (Knight et al. 1999). Table 5 shows the trend of in-migration in towns in three provinces. Town in-migration has kept increasing in relatively rich province of Zhejiang over the last five years; 1.64 percent of population migrated into towns in 1996, and 6.49 percent migrated in 2000. During the period of 1996-2000, as much as 18.21 percent of population migrated into towns. Whereas in Henan, a relatively poor province, the 6 growth of town in-migration has been staggering over the last five years, hovering around 1.5 percent in average each year during the 1996-2000 period. During whole period of 1995-2000, in-migration only accounted for 7.3 percent of town population. In Sichuan, which was economically in the middle range among three provinces, in-migration into towns has shown an upward trend during this period although growth was not been as pronounced as that in Zhejiang province. Table 5 Migration in towns as percentage of total population:2000 live before 11/1- 1996 1997 1998 1999 2000 total of in townOct31 12/31 last five years since 1995 1995 born (%) Zhejiang 61.70 19.71 0.38 1.64 1.96 3.21 4.91 6.49 18.21 Henan 70.51 21.88 0.33 1.15 1.28 1.73 1.79 1.34 7.29 Sichuan 57.65 26.99 0.38 1.69 2.16 3.22 4.01 3.84 14.92 ________________________________________________________________________ Sources: calculated from Tabulation on 2000 Population Census of the P. R. China, Table L7-2. A main expectation underlying the advocacy for town development is to have the towns to absorb excessive laborers who would otherwise head for large cities. In the following two tables, we present the geographic composition of in-migrants to see if towns absorb excessive labor force from the rural areas. In Table 6, we categorize town in-migrants by original place of residence. Data suggest that towns in three provinces have indeed absorbed rural population. In Zhejiang, 38 percent of in-migrants were originally rural residents, living in villages. In Henan and Sichuan provinces, 53 percent and 37 percent came from rural areas respectively. If we include those who were originally rural town residence, then the majority of town in-migrants came from rural areas. Whereas the percentage of in-migrants from city has been small, accounting for 4.6%, 15.6% and 12.4% respectively for Zhejiang, Henan and Sichuan. Table 6 Town in-migration by original residence place: 2000 Province village town rural town urban urban residence residence residence residence (%) Zhejiang 38.49 45.19 11.69 4.64 Henan 52.99 20.38 11.00 15.63 Sichuan 37.34 27.89 22.38 12.40 Sources: calculated from Tabulation on 2000 Population Census of the P. R. China, Table L7-6. 7 Another geographic aspect of migration is to look at the issue of whether in- migrants are from same province or other provinces. Table 7 suggests that in Zhejiang, about 41 percent of towns in-migrants were from other provinces. In contrast, in Henan and Sichuan, majority of migrants were from same province, and there was only about 8 percent from other provinces. Table 7 Town in-migration by province and region: 2000 Zhejiang Henan Sichuan (%) From same province 59.08 91.93 92.68 From other provinces 40.92 8.06 7.32 100.00 100.00 100.00 Of which: coast 11.03 38.33 22.79 central 60.74 34.00 15.47 west 28.23 27.67 61.74 total 100.00 100.00 100.00 ________________________________________________________________________ Sources: calculated from Tabulation on the 2000 Population Census of the P. R. China, Table L703b, and Table L704b. Note: regional classification: Coast includes following provinces: Heilongjian, Jinin, Liaoning, Beijing, Tianjian, Hebei, Shandong, Jiangsu, Shanghai, Zhejiang, Fujiang, Guangdong, Guanxi and Hainan. Middle includes following provinces: Nei Mongol, Shanxi, Henan, Anhui, Hubei, Hunan and Jiangxi. West includes following provinces: Xingjiang, Qinghai, Gansu, Ningxia, Shaanzi, Sichuan, Guizhou, Yunnan and Tibet. Table 7 further shows that among those of inter-provincial in-migrants in Zhejiang, majority (89%) came from central and western regions. Thus, many in- migrants in fact crossed provincial lines and traveled a quite long distance to seek economic opportunities in towns in Zhejiang. Whereas in Henan and Sichuan, even though the inter-provincial in-migrants only accounts for less than 10 percent, many of them (61.7% in Henan and 77.2% in Sichuan respectively) were from the western and central regions. Migration was largely motivated by economic opportunities. Indeed, when census long-form asks about the reasons for migration into towns (Table 8), more than half (60 percent) of in-migrants in Zhejiang responded that purpose of migration into towns was the engagement in production and business, job transfer or assignment. In contrast in Henan and Sichuan, there were only 37 percent and 34 percent of in-migrants who came for the purpose of engaging in production and business, job transfer or assignment. 8 Table 8 The reasons for migration into towns in three provinces in China:2000 1 2 3 4 5 6 7 8 9 (%) Zhejiang 54.92 3.92 3.03 6.55 5.55 5.96 14.29 2.99 2.79 Henan 17.79 12.31 7.17 7.33 7.68 13.70 21.96 4.53 7.54 Sichuan 18.71 9.90 5.06 21.30 6.20 9.47 14.44 8.47 6.43 Sources: calculated from Tabulation on 2000 Population Census of the P. R. China. L705B. Note: 1, engage in production and business; 2. job transfer; 3, job assignment; 4,study and training; 5, resettlement; 6, marriage; 7, move with family member; 8, seeking help from relatives; 9, others. The age composition of in-migrants also seems to support for the view that migration is economically rewarding. When we tabulated in-migrants by age (figure 1), we found that most town in-migrants were in the categories of economic active groups of age between ages of 20 to 34. This inverted U-shaped relation of town in-migrants confirms the findings of other studies (Hare 1999; Zhao 1999). People in these age groups are most likely to migrate, and probability of migration decreases with age. In rural area, a piece of land that could be hold as a social security against old ages, whereas in towns, old people have to rely on some kind of social security, which has not been well established so far, preventing elderly moving into towns (Cai 2003). Figure 1 Age differentials in migrants into town: 2000 The age distribution of in-migrants in town by provinces 25 20 15 % 10 5 0 0-- 5-- 10- 15- 20- 25- 30- 35- 40- 45- 50- 55- 60- 65+ 4 9 -14 -19 -24 -29 -34 -39 -44 -49 -54 -59 -64 age group Zhejiang Henan Sichuan National Source: calculated from Tabulation on the 2000 population census of the P. R. China. Table 7-5, page.1818-1827, and L705B. 9 5. Educational differentials in migration The literature also emphasizes the importance of human capital as a precondition of migration (Schultz 1982; Todaro 1969). For example, one percent of national sampling survey conducted in 1995 shows that among all the migrants, 30.8 percent had received senior high and higher formal education, comparing to 11.3 percent for the same category of national average (Yang 1997). Zhao (1999) further found that better educated people in rural areas were not necessary to leave countryside to migrate to other provinces and likely to shift their jobs from agriculture to the non-agricultural sector in towns within own region. One of reasons lies in the fact that educated rural people regard migration as a loss of their social status. Thus, it is unclear whether the development of towns attracts a large part of educated labor force, and whether development of towns attracts educated laborers from its own region who otherwise heads to other regions as suggested by Zhao. In Table 9, we represent the findings on educational attainment of those who migrated into towns, and of general population in towns in three provinces. Table 9 Educational attainment of town in-migrants (6 years old or above) by migration status: 2000 Province 1 2 3 4 5 6 7 (%) Zhejiang in-migrant 4.15 28.89 43.91 12.76 5.00 3.94 1.33 town pop 8.03 35.32 38.23 11.67 3.29 2.66 0.79 Henan in-migrant 3.58 17.59 37.37 15.19 13.65 8.95 3.67 town pop 5.94 27.12 41.31 14.17 5.39 4.92 1.16 Sichuan in-migrant 3.80 20.09 29.79 24.26 12.79 7.18 2.09 town pop 6.19 32.20 36.21 12.94 6.96 4.41 1.09 Sources: calculated from Tabulation on 2000 Population Census of the P. R. China, Table 1-8b, and Table L7-6-1b and Table L7-6-2b. note: 1, literate; 2, primary school; 3, secondary school; 4, high school; 5, vocational school; 6, college; 7, university or above. In representing the findings on educational attainment of those who migrated into towns, and of general population in towns in three provinces (Table 9), we have found that in-migrants generally possess higher level of educational attainment than those of general population. In Zhejiang, for example, there were 17.8 percent of all in-migrants who have completed high school or professional technical school, whereas only 14 percent of town general population who have completed high school or professional technical schools. 10 Furthermore, this association is found to be much more appealing in less- developed western province of Sichuan and central region of Henan province than that in more developed coastal region of Zhejiang. In Henan, there were 28.8 percent of in- migrants who completed high school or professional technical school whereas the corresponding figure for town local population was 19.6 percent. In Sichuan, the contrast is more revealing, with corresponding figure of 37.1 percent for in-migrants and 19.9 percent for local population. Table 10 Educational attainment of town in-migrants (6 years old or above) by province: 2000 By province 1 2 3 4 5 6 7 (%) (Zhejiang) intra-province migration 4.89 26.42 36.64 17.55 7.33 5.64 1.53 inter-province migration 3.09 32.45 54.41 5.84 1.62 1.48 1.11 (Henan) intra-province migration 3.55 17.33 37.09 15.16 14.06 9.22 3.59 inter-province migration 3.94 20.49 40.61 15.47 9.05 5.90 4.54 (Sichuan) intra-province migration 3.82 20.02 29.51 24.88 12.79 7.05 1.92 inter-province migration 3.58 20.94 33.35 16.37 12.75 8.78 4.23 Sources: calculated from Tabulation on 2000 Population Census of the P. R. China, Table L7-6-1b and Table L7-6-2b. note: 1, literate; 2, primary school; 3, secondary school; 4, high school; 5, vocational school; 6, college; 7, university or above. We also further breakdown the level of educational attainment of town in- migrants by original place as we are interested in a question of whether inter-provincial in-migrants possess higher level of educational attainment than those in-migrants who are from the same province. Data (Table 10) show that in Zhejiang, among inter-provincial in-migrants, only 10 percent of them had high school education attainment or higher, whereas among intra-provincial in-migrants, as high as 32 percent of them had high school education attainment or higher. Thus the fact that a higher percentage of educated migrants coming from same province than other provinces seems to give some supports to what Zhao (1999) has found in her study that better educated people in rural areas probably were not necessary to leave countryside to migrate to other provinces and were likely to shift their jobs from agriculture to the non-agricultural sector in towns nearby instead. However, when we further look at the data on educational attainment of in- migrants by original place in Henan and Sichuan, we do not find this kind of sharp contrast between those who came from same province and other provinces. For example, in Sichuan, about 44 percent of intra-province in-migrants had high school education or 11 above, whereas about 42 percent of inter-province in-migrants had high school education or above. The contrast is really small. The same is hold true for Henan. In other words, Zhao's hypothesis can only be applied to the rich region such as Zhejiang, not in relatively poor western and central regions such as Henan and Sichuan. 6. The migration and labor market segregation Although studies on the labor force segregation in towns are few, there are some studies examining the consequence of migration on labor market segregation (Meng 2001; Yao 2001). While Yao found that migrant was 17.6 percent less likely to have a white-collar job than a local resident, controlling for personal characteristics, Meng found a sharp occupational differentials between local population in the cities and in-migrants. To explore this issue of whether labor market became segregated in towns, we present the occupational composition of in-migrants and of local population in towns in three provinces (Table 11). We found that in Zhejiang two-thirds of in-migrants (63 percent) worked as blue-collar production workers, which was higher than town local population. Only about 43 percent of town local population worked as blue-collar production workers. Data further reveal that in-migrants were generally less likely to take white-collar jobs such as managers and technicians as compared to the town local population. Thus, labor market in towns in Zhejiang seems to be quite segregated, with in-migrants being more likely to take blue color production work, which might require hardship in their working conditions. In Table 12, we further explore this issue by breaking down occupational composition of towns in-migrations according to whether they come from the same province or from other provinces. Clearly, data reveal that in Zhejiang in-migrants from other provinces were even more likely to be blue-collar production workers than in- migrants from same province. For example, 83 percent of in-migrants from other provinces were engaged in blue-collar production jobs, whereas only 43 percent of in- migrant from same province did so. This probably could be explained by the lower educational attainment of those cross-province migrants. Thus the probability of being a blue production worker is twice higher for in-migrants coming from other provinces than the in-migrants from the same province. Likewise, for the white-collar jobs, the chances are much smaller for the in-migrants from other provinces as shown in the Table. Thus labor market segregation not only exists between town in-migrants and local population, but also exists between those who came from same province or came from other provinces. However, we also find quite different story in other two relatively poor provinces. Data in Table 11 do not show a sharp contrast in percentage distribution of blue-collar production workers between those in-migrants and local population as we have seen in the Zhejiang data. Only slightly higher percentage of in-migrants worked as blue-collar production worker than that for the town local population, 24% vs. 16% in Henan and 23% vs. 19% in Sichuan. Thus, data suggest labor market has been much more flexible in Henan and Sichuan than in Zhejiang. 12 In contrast, in Henan and Sichuan, a good proportion of in-migrants worked as technician. For example, in Henan, about 20 percent of in-migrants worked as technician, higher than the percentage of town local population, which had only 12.55 percent of them working as technicians. The pattern also holds true in Sichuan. In addition, there were 28 percent and 34 percent of in-migrants working in service sector in towns in Henan and Sichuan provinces respectively, which were also higher than that for town general population. Furthermore, among the town in-migrants in Henan and Sichuan, data of Table 12 do not suggest a severe labor market segregation, which exists between those in- migrants from same province and those from outside provinces as we observed in Zhejiang data. For example, generally there are more in-migrants from other provinces working in the blue-collar production sector than in-migrants of same province in Henan (37.89% vs. 22.60%) and Sichuan (23.13% vs. 26.53%); however the difference was much smaller than that we observed in Zhejiang (83.31% vs. 43.24%) For the same token, although the percentage of workers engaging in the white- collar occupations such as manager, technician, and sale-related jobs was lower among inter-province in-migrants than among the intra-province in-migrants, the gap is much smaller. All of this suggests that labor market is more flexible in Henan and Sichuan, whereas it is more segregated in Zhejiang. Table 11 Occupation composition of towns in-migrants (15 years old or above) by migration status: 2000 ________________________________________________________________________ province type of population 1 2 3 4 5 6 7 ________________________________________________________________________ (%) Zhejiang town in-migrant 1.38 7.71 3.80 19.59 4.02 63.46 0.04 local population 2.57 8.84 4.93 23.38 17.22 43.00 0.05 Henan town in-migrant 3.90 20.48 11.25 28.38 11.85 23.95 0.18 local population 3.19 12.55 7.65 17.45 43.01 16.00 0.14 Sichuan town in-migrant 3.62 19.41 8.59 34.22 10.47 23.43 0.26 local population 3.14 14.50 6.57 24.41 31.81 19.38 0.20 ________________________________________________________________________ Source: calculated from Tabulation on 2000 Population Census of the People's Republic of China, Table 4- 2b page 1167-1201, and Table L7-1b and L7-2b. note: occupation title: 1, manager and administrative personals, 2. professionals and technician, 3. staff and other auxiliary workers, 4. salesperson, 5. agriculture, animal husbandry, and fishery, 6. production workers, 7. unclassified. 13 Table 12 Occupation composition of town in-migrants (15 year old or above) by province: 2000 _______________________________________________________________________ province by province 1 2 3 4 5 6 7 ________________________________________________________________________ (%) Zhejiang within province 2.56 13.28 6.54 28.62 5.70 43.24 0.05 outside province 0.22 2.24 1.11 10.72 2.37 83.31 0.04 Henan within province 4.05 21.30 11.49 28.36 12.02 22.60 0.17 outside province 2.29 12.06 8.79 28.62 10.03 37.89 0.33 Sichuan within province 3.70 19.84 8.47 34.02 10.57 23.13 0.25 outside province 2.77 15.06 9.88 36.02 9.44 26.53 0.30 ________________________________________________________________________ Source: calculated from Tabulation on 2000 Population Census of the People's Republic of China, Table L7-1b and L7-2b. note: occupation title: 1, manager and administrative personals, 2. professionals and technician, 3. staff and other auxiliary workers, 4. salesperson, 5. agriculture, animal husbandry, and fishery, 6. production workers, 7. unclassified. 7. Conclusion Our analysis reveals a diversified pattern of town migration in the three provinces. Zhejiang exhibits steady and significant growth in town in-migration over 1995-2000, whereas the other two provinces, Sichuan and Henan, remain quite staggering. Nevertheless, there is one thing in common that majority of town in-migrants were from rural areas. The diversified pattern of town in-migration in the three provinces probably suggests that migration has been quite responsive to the economic opportunities and living conditions in towns. Many migrants went to towns in Zhejiang province for the sake of seeking business and job opportunities. Nearly half of migrants came from other provinces, especially from the western and central regions. In contrast, a lack of a strong manufacturing sector and good living conditions in Henan and to some extent in Sichuan did not attract outsiders. Most in-migrants came from the same province for non-business reasons. Our study also suggests that town in-migrants generally possess a higher level of educational attainment than the general population in towns; this is particularly true in less-developed provinces such as Sichuan and Henan. Thus, migration could benefit the development of towns, especially in the western and central regions. Data further suggest that as a town develops, as in the case of the richer province of Zhejiang, better-educated people probably did not leave the countryside to migrate to the cities in other provinces and instead were more likely to shift their jobs from agriculture to the non-agricultural sectors in towns nearby within the province. Thus, the development of towns provides a buffer that could hold more migrants, reducing the migration pressure on large cities. 14 The findings further reveal that the labor market in less-developed regions such as Henan and Sichuan became more flexible to accommodate skilled in-migrants; a good proportion of in-migrants worked in white-collar jobs. In Zhejiang the labor market seems to be segregated, with most of the in-migrants engaging in blue-collar jobs. Thus as towns develop, the consequences of migration on labor market segregation will be an increasingly important policy issue for better town development strategy. 15 References Cai, Fang 2003. Migration and Social-economic Insecurity: Patterns, Processes and Policies. Manuscript. International Labor Office, Geneva. Chan, Kam Wing. 2001. "Recent Migration in China: Patterns, Trends and Politics." Asian Perspective, vol. 25, no.4 (127-153). Chen, C., L. Chang, and Y. M. Zhang. 1995. "The Role of Foreign Direct Investment in China: Post-1978 Economic Development," World Development. 23(4):691-703. Fei, Xiaotong. 1986. On Small Towns and Other Issues. China: Tianjin People's Press. Hare, Denise. 1999. "Push versus Pull Factors in Migration Outflows and Returns: Determinants of Migration Status and Spell Duration among China's Rural Population." Journal of Development Studies 35(3):45-72. Huang, P. C.C. 1990. The Peasant Family and Rural Development in the Yangzi Delta, 1350-1988. Stanford, CA: Stanford University Press. Johnson, E.A.J. 1970. The Organization of Space in Developing Countries. Cambridge: Harvard University Press. Knight, John, Lina Song, and Huaibin Jia. 1999. "Chinese Rural Migrants in Urban Enterprises: three Perspectives." Journal of Development Studies 35(3): 73-104. Ma, L.J.C. 1996. "The Spatial Patterns of Inter-Provincial Rural-to-Urban Migration in China, 1982-1987." Chinese Environment and Development. 7 (1):73-102. Meng, Xin. 2001. "The Informal Sector and Rural-Urban Migration---A Chinese Case Study." Asian Economic Journal 15(1):71-89. NBSC (National Bureau of Statistics of China). 2002. Tabulation on the 2000 Population Census of the People's Republic of China. Beijing: China Statistical Press. Oi, J.C. 1992. "Fiscal Reform and the Economic Foundation of Local State Corporation in China." World Politics. 45 (1):99-126. Oi, J.C. 1999. Rural China Takes Off. Berkeley, CA: University of California Press. Schultz, T. P. 1982. "Lifetime Migration within Educational Strata in Venezuela: Estimates of a Logistic Model." Economic Development and Cultural Change 30(3):559-594. Skinner, William. 1964. "Marketing and Social Structure in Rural China." Journal of Asian Studies, 24, pp.3-43. 16 Skinner, G. W. 1985. "Rural Marketing in China: Repression and Revival." China Quarterly, 103:393-413. Todaro, Michael. 1969. "A Model of Labor Migration and Urban Unemployment in Less Developed Countries." The American Economic Review 59(1):138-148. Wei, Y.H. 1997. "Interregional Migration in Socialist Countries: The Case of China." GeoJournal. 41 (3):205-214. Wei, Y.H. 1999. "Regional Inequality in China." Progress in Human Geography. 23(1):49-59. Yan, Sangpi. 1998. "Chongguo Jiushiniandai dichujia rungou qianyide sheteijiqi jizhi." (The Empirical Evidence of Population Migration between Regions in 1990's in China) Journal of Sociological Study. Vol. 2(67-74). Yang, Dennis Tao and Hao Zhou. 1999. "Rural-Urban Disparity and Sector Labor Allocation in China." Journal of Development Studies 35(3): 105-133. Yang, Jianchun. 1997. "Scale, Direction and Structure of China's Migration-Analysis on 1 Percent National Sampling Survey." Reference Bulletin of Statistical Research, No2. Yao, Yang. 2001. "Social Exclusion and Economic Discrimination: The Status of Migrations in China's Coastal Rural Area." Working Paper E2001005, China Center for Economic Research, Beijing University. Zhao, Yaohui. 1999. "Migration and Earnings Difference: the Case of Rural China." Economic Development and Cultural Change, 47(4), pp.767-82. 17