Information about Asia and the Pacific Asia y el Pacífico
Chapter

11. Labor Mobility in China and India: The Role of Hukou, Caste, and Community

Author(s):
Eswar Prasad, Steven Dunaway, and Jahangir Aziz
Published Date:
September 2006
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

Arvinder Singh

Major debates on the industrialization of agrarian economies view labor mobility—sectoral and spatial, from agriculture to industry and from villages and small towns to cities—as central to understanding the pace and nature of industrialization. Many see labor mobility as the result of the commercialization and capitalization of agriculture, which leads to an increasing number of peasants who are alienated from the land and must be absorbed by the expanding urban and industrial sector. Indeed, industrialization is seen as hinging on the successful, rapid, and smooth transfer of the so-called surplus labor from agriculture to nonagriculture. Land and labor productivity rises in agriculture as a result of the labor transfer, which enables it to contribute more capital resources to the developing industrial sector through terms of trade (TOT) favorable to industry. The labor transfer mobilizes the savings of the rural households directly or through taxes and provides a ready “home market” for industrial sector goods. Speedy development of the industrial sector itself, along with rural industrialization and increasing urbanization, takes care of the transferred labor from agriculture.

Not only do the two sectoral, artificially neat distinctions present problems, but these models of the movement of labor and capital from agriculture to the nonagricultural sectors are extremely state-centered. It is invariably the state that is seen as restricting, facilitating, or regulating the transfer of labor according to the needs of industrialization. In fact, neither the commercialization and capitalization of agriculture alone, which may indeed alienate peasants from the land and compel them to seek a livelihood elsewhere, nor the might or the policies of the state may explain why peasants are known to “irrationally”cling to the land (or, for that matter, to be too eager to move out even when the state intends to “keep” them in agriculture). Peasants often cling to the land despite the pressures of destitution. They may refuse to join the ranks of the factory proletariat even when they are made to give up the land.

Because states and state-centered theories are so preoccupied with extracting capital and food grains from agriculture for the supposed nascent industrial sector, labor is virtually treated as a residual issue. For one thing, by the time the state intervenes to “develop” agriculture for the sake of industry (a typical postcolonial developing-economy endeavor, modeled on Soviet-type planning), the industrial sector may already be fairly developed, generating its own surpluses and working on its own specific needs for skilled and unskilled labor. By this time, the industrial sector may not be short of capital in a blanket sense nor in need of help by cheap labor from agriculture. For this reason, planners should take into account simultaneous developments in agriculture and nonagriculture.

In addition, because of an aversion to depending on food imports, states may attempt to force peasantries to produce and deliver food grains for the national economies. These efforts can affect the retention or displacement of labor from agriculture. It would be better to recognize the ways that specific state policies affect labor in agriculture than to ascribe results vaguely to the “state.”

The TOT, or the price scissors, so central in the literature on the ruralurban interface, fail to see beyond financial surplus from the agricultural sector. The fact is that adverse TOT could depress agriculture and make more labor 1 redundant. Policies on urbanization, on rural industrialization, and on raising incomes of agricultural households through investments in agriculture, which are often interrelated, would slow or hasten the movement of labor from agriculture. Equally important, labor’s exodus from agriculture, whether on account of state policies or otherwise, might influence rural incomes, rural industry, and urbanization. It would help if the literature on rural nonfarm activities and employment and rural industry, and the demographic work on urbanization and migration, were linked more meaningfully with the economic literature on industrialization and labor mobility. The city studies as such, though interesting and growing in number, would be more useful if they would come out of the city limits and start looking at wider local economies of which the city is a part or with which it is linked.

Notwithstanding the occasional admission that the process of land alienation (or that of agriculture’s dwindling relative importance in contributing to income or in engaging or supporting labor) is a long one, spread over generations, the models remain static by and large. Besides, a persisting or thriving agriculture is possible, is compatible with industrialization,2 and has implications for rural-urban labor mobility. Even the role of differing forms of agrarian structures—land organization, ownership, and rights—and the size of holdings in rural—urban labor transfers has hardly been adequately captured in the literature.

Also, a wider class framework and a better appreciation of social differentiation in the economy than that suggested by the two-class polarization/pauperization framework—labor versus capital—would help in better understanding the dynamics of labor movement, both forced and voluntary, from agriculture to nonagriculture or from the rural to the urban sector. The polarization thesis is questionable in both agriculture and industry.3 The thesis is simplistic. It is not the best way to analyze agrarian transformation and the way it links to the wider process of industrialization in the economy through sectoral movement of labor, capital, and enterprise. Creating a labor market, or a working class, in an industrializing agrarian economy is not all about conflict between labor and capital. The creation of a labor market would be better understood in the context of the wider social structure and institutions in which the developing economy is embedded. Also, not all peasants who leave the land will become part of the manual labor workforce. Some will become white-collar workers and entrepreneurs. Besides, there are also nonpeasant sections of the agricultural population to consider, such as artisans and traders. The social and economic transition of artisans and traders is more difficult to capture in the polarization/pauperization perspective.4 Then there is also the question of social and class relationships among land-owning peasants, landless labor, artisans, merchants, and industrialists; the relative social mobility of these classes should be considered.

Industrialization and Social Structure

Developing economies may be structured strongly around social and communal identities—for example, around caste and religion in India. Different social groups or communities provide different aspects of production in the economy. Land may belong to certain caste groups or religious communities; landless agricultural labor is provided by another caste group (usually lower menial castes); industrial capital may be in the hands of still others—traditionally the elite urban castes; and trade usually involves the traditional merchant communities (like Baniyas).

The following tables show how this is indeed the case in Punjab, a northern province in India, which also happens to be a classic Green Revolution site and is known for its rapid economic development over the past three to four decades. In fact, Punjab is a site of very large-scale labor migration—both into and out of the province. The data presented here make it clear that production and exchange in Punjab are almost neatly compartmentalized along religious and caste lines. The peasants are Sikhs (Jats), whereas merchants are Hindus (Khatris/Aroras/Baniyas). Sikhs cultivate and organize agricultural production, Hindus trade. Peasants live in villages, merchants live in towns. The occupational barriers between agriculture and nonagriculture are clear and strong.

Table 11.1 shows that the rural population in Punjab is predominantly Sikh and the urban population is predominantly Hindu. Of the rural households, 70 percent are Sikh whereas 65 percent of the urban households are Hindu.5 The dominance of Sikh households in rural areas is greatest in the districts of Amritsar, Faridkot, Ludhiana, Bathinda, and Sangrur; while the dominance of Hindu households in urban areas is most pronounced in the districts of Jalandhar, Hoshiarpur, Gurdaspur, and Firozpur.

Table 11.1.Rural and Urban Population in Different Districts in the Indian Punjab by Religion of the Head of the Household(In percent)
RuralUrban
DistrictSikhsHindusSikhsHindus
Punjab70.0827.7432.2565.24
Gurdaspur50.8040.7019.7377.21
Amritsar90.827.0141.6056.96
Firozpur64.0634.7920.8576.72
Ludhiana87.2011.9435.0463.21
Jalandhar52.7346.2120.6277.71
Kapurthala70.2728.7133.0565.92
Hoshiarpur41.6657.0622.9375.58
Rupnagar58.1940.4135.5562.03
Patiala62.6735.3334.0664.72
Sangrur79.5816.4934.7950.48
Bathinda86.7612.5436.5962.40
Faridkot89.799.6042.7556.29
Source: Singh (1999, p. 193).
Source: Singh (1999, p. 193).

Table 11.2 makes it even clearer that the Sikhs in Punjab are concentrated heavily in rural areas. The table shows that 84 percent of the Sikhs live in rural areas and only 16 percent in urban areas. The occupational structure and religious composition of towns in Punjab make it clear that most merchants are Hindu and live in towns.

Table 11.2.Percentage of Sikh Households in Different Districts in the Indian Punjab According to Residence (Rural/Urban)
Sikhs
DistrictRuralUrban
Punjab83.8016.20
Gurdaspur89.4310.57
Amritsar79.9820.02
Firozpur90.759.25
Ludhiana75.4224.58
Jalandhar82.2017.80
Kapurthala81.9818.02
Hoshiarpur90.749.26
Rupnagar82.4117.59
Patiala78.8021.20
Sangrur87.8312.17
Bathinda87.5012.50
Faridkot86.4413.56
Source: Singh (1999, p. 193).
Source: Singh (1999, p. 193).

The towns in Punjab in which trading is predominant are precisely the ones where Hindus compose the majority of the population. Table 11.3 presents the percentage of the Hindu population that is engaged in trade in 42 towns the census has classified as primarily trading towns.6 The names of the towns are arranged in descending order by population size.

Table 11.3.Percentage of Hindu Population and of Main Workers Engaged in Trade in Primarily Trading Towns in the Indian Punjab(In percent)
TownHindusMain

Workers

in Trade
Abohar83.7030
Ahmedgarh56.2835
Amloh52.7328
Banga70.1929
Bassi51.4229
Batala66.6731
Bhogpur65.8534
Bhucho Mandi67.1737
Budhlada71.8341
Dasua68.9232
Dhuri63.5136
Dinanagar86.5034
Doraha47.6330
Fatehgarh Churian45.8734
Garhdiwala72.3637
Giddarbaha61.7132
Goniana63.2546
Guru Har Sahai74.8238
Jagraon43.3233
Jaitu57.5228
Khanna61.5833
Kot Kapura50.5330
Lehragaga71.5240
Malout71.8433
Mansa55.5231
Morinda51.8728
Mukerian77.7843
Nabha64.7731
Nawanshahr75.0531
Patran72.4248
Qadian56.8938
Raman64.6533
Rampura Phul55.7431
Samana68.9341
Samrala48.0929
Shahkot59.4538
Sirhind50.1727
Sunam60.0133
Talwandi Bhai45.1135
Tapa59.0936
Tarn Taran25.2830
Urmar Tanda55.3929
Source: Singh (1999, p. 194).
Source: Singh (1999, p. 194).

The mandi towns are more striking. They owe their very existence to trade in agricultural goods. These towns are more prominent in the Malwa region, which comprises the southern districts of Bathinda, Sangrur, Faridkot, and Firozpur. Trade in these towns (including Abohar, Kot Kapura, Malout, Mansa, Sunam, Dhuri, Samana, Rampura Phul, Ahmedgarh, Budhlada, Patran, and Bhucho Mandi) has been conventionally handled by Baniyas.7

The existing industrial capital in Punjab is primarily owned by the urban trading castes, the high-caste Hindu elite—Khatris, Aroras, and Baniyas—all of whom have trading backgrounds; however, artisans (Ramgarhias), who set up the small engineering goods workshops and factories that are connected with Green-Revolution technology, also own industrial capital. The religion and caste of the owners of the top 10 corporations in Punjab are telling (see Table 11.4 for religion and caste/subcaste of the chairperson and see Table 11.5 for the religion and caste of the members of the boards of directors of these companies). As these tables show, these companies are owned predominantly by Hindus. Of the 10 chairpersons, 9 are Hindu—3 are Khatri, 5 are Baniya, and 1 is Arora. The chairperson of the remaining company, Ranbaxy, is an Arora Sikh. Out of the 80 members of the boards of directors of these companies, as many as 70 are Hindus, 45 of them being Khatri, Baniya, or Arora.

Table 11.4.Religion and Caste of the Owners of the Top Ten Corporate Companies in the Indian Punjab
Name of CompanyReligious Affiliation

of Chairman
Sub-Caste of the

Chairman
Caste

Group
Ranbaxy, MohaliSikh1Arora
JCT, HoshiarpurHinduThaparKhatri
Hero Cycles, LudhianaHinduMunjalArora
Mahavir Spinning, LudhianaHinduOswalBaniya
Jagatjit, KapurthalaHinduJaiswalBaniya
Oswal Agro, LudhianaHinduOswalBaniya
Vardhman, LudhianaHinduOswalBaniya
ICT, MohaliHinduThaparKhatri
Malwa Cotton, LudhianaHinduOswalBaniya
JCT Fibres, HoshiarpurHinduThaparKhatri
Source: Constructed by the author from a list of the names of the chairpersons and the members of the boards of directors of the top corporate companies in Punjab. The list was provided in a report by the Centre for Monitoring Indian Economy (January 1995). The table also appeared in Singh (1999, p. 153).Note: The companies have been arranged in descending order according to sales in 1994.

Could not be ascertained.

Source: Constructed by the author from a list of the names of the chairpersons and the members of the boards of directors of the top corporate companies in Punjab. The list was provided in a report by the Centre for Monitoring Indian Economy (January 1995). The table also appeared in Singh (1999, p. 153).Note: The companies have been arranged in descending order according to sales in 1994.

Could not be ascertained.

Table 11.5.Religious and Caste Composition of the Board of Directors of the Top Ten Corporate Companies in the Indian Punjab
Total Size of theReligion of the MembersMembers of the Boards Belonging to the same Sub-Caste as Chairman
Name of CompanyBoard of DirectorsSikhsHindusOthers(Number includes chairman)Khatris (Hindu)Aroras (Hindu)Baniyas 1Others
Ranbaxy. Mohali927136
JCT, Hoshiarpur9183513
Hero Cycles, Ludhiana981772
Mahavir Spinning, Ludhiana917112115
Jagatjit. Kapurthala82631133
Oswal Agro, Ludhiana661123
Vardhman, Ludhiana981245
JCT, Mohali9182315
Malwa Cotton, Ludhiana994162
JCT Fibres, Hoshiarpur33221
80770318101735
Source: Constructed by the author from a list of the names of the chairpersons and the members of the boards of directors of the top corporate companies in Punjab. The list was provided in a report by the Centre for Monitoring Indian Economy (January 1995). The table also appeared in Singh (1999, p. 154).

All Baniyas are Hindus.

Source: Constructed by the author from a list of the names of the chairpersons and the members of the boards of directors of the top corporate companies in Punjab. The list was provided in a report by the Centre for Monitoring Indian Economy (January 1995). The table also appeared in Singh (1999, p. 154).

All Baniyas are Hindus.

At the top of the corporate ladder in Punjab, Khatris are represented by Thapars, Aroras by Munjals, and Baniyas by Oswals. Artisans are mainly Ramgarhias. Thapars represent the old Khatri industrial elite, whereas Munjals are newcomers, having emerged from the refugee middle class. The line that divides Baniya and Khatri in commerce seems to extend to industry too, where there is also a caste distinction. While the Baniya are rice shellers, the Khatris are more likely to enter the kind of modern industry not really related to agriculture.8

Notice how the lower castes are almost completely missing from industry and trade. As Tables 11.6 through 11.9 show, the Dalits, or the scheduled castes, play an extremely marginal role in trade in Punjab; commodity trade there is predominantly in the hands of the higher castes. Scheduled caste traders seem to be catering to the lower end of the markets. Fixed costs, working capital, and turnover show that scheduled caste traders operate at a much lower level than higher caste traders (see Table 11.6). Interestingly, scheduled caste traders operate self-reliantly, with a low frequency of loans.

Table 11.6.Estimated Value per Enterprise of Fixed Assets, Working Capital, Outstanding Loan, and Turnover of Smaller Trading Enterprises in the Indian Punjab by Caste Group, 1990–91
Caste GroupEnterprisesFixed

assets

(Rs.)
Working

Capital

(Rs.)
Loans

(Rs.)
Turnover

(Rs.)
(Number)(Percent)
Higher castes217,46884.4324,951.8338.502.414.224.29306.766.79
Dalits39,04413.168,844.975,065.231,064.7674,486.64
Scheduled tribes1,0280.409,363.8228,014.8511.49121,330.83
Not reported320.01130,215.9437,501.880.00274,495.31
Total257,572100.0022,461.2733,391.813,727.92270,812.79
Source: Singh (1999, p. 147).Note: Includes proprietary and partnership enterprises within the same household.
Source: Singh (1999, p. 147).Note: Includes proprietary and partnership enterprises within the same household.

Scheduled castes are more evident in trade in rural areas. One-fourth of the trading enterprises located in rural areas in Punjab are owned by scheduled castes, whereas in towns, they account for only 8 percent of firms (see Table 11.7 and Table 11.8).9 However, trade based in rural areas is itself of small quantitative importance in Punjab (see Table 11.7 and Table 11.9).10

Table 11.7.Estimated Value per Enterprise of Working Capital and Turnover of Smaller Trading Enterprises in Punjab by Caste Group and Location (Rural/Urban), 1990–91
Enterprises (percent)Working Capital (Rs.)Turnover (Rs.)
Social GroupRuralUrbanRuralUrbanRuralUrban
Higher castes74.6391.4711,15354,33480,217439,563
Dalits25.048.063,4448,68445,583139,003
Scheduled tribes0.330.455,73839,91868,836149,411
Not reported0.000.0216,20038,92250,625289,420
Total100.00100.009,20550,77071,508414,013
Source: Singh (1999, p. 147).Note: Includes proprietary and partnership enterprises within the same household.
Source: Singh (1999, p. 147).Note: Includes proprietary and partnership enterprises within the same household.
Table 11.8.Value per Enterprise of Fixed Assets, Working Capital, Outstanding Loans, and Turnover of Smaller Trading Enterprises Run by Dalits in Punjab by Location, 1985–86
LocationEstimated Number

of Enterprises
Fixed

Assets
Working

Capital
Loans

Outstanding
Turnover

(Rs.)
RuralOATE 1106.64737,2942,1401,96565,388
Others913825--27,004
UrbanOATE5.4475,0731,80130234,780
Others30611,0889.346-90.037
Source: Singh (1999, p. 148).

OATE is an acronym for “own account trading enterprises,” referring to enterprises without hired labor.

Source: Singh (1999, p. 148).

OATE is an acronym for “own account trading enterprises,” referring to enterprises without hired labor.

Table 11.9.Value per Enterprise of Fixed Assets, Working Capital, Outstanding Loans, and Turnover of Smaller Trading Enterprises of All Castes by Location in Punjab, 1985–86
LocationEstimated Number

of Enterprises
Fixed

Assets
Working

Capital
Loans

Outstanding
Turnover (Rs.)
RuralOATE 1197,69822,7443,4721,68557,357
Others5,33230.47111,8241,181311,832
UrbanOATE117,25918,38915,6581,85052,670
Others23.11726,129108,40420,901880,537
Source: Singh (1999, p. 148).

OATE is an acronym for “own account trading enterprises,” referring to enterprises without hired labor.

Source: Singh (1999, p. 148).

OATE is an acronym for “own account trading enterprises,” referring to enterprises without hired labor.

I have argued elsewhere that such a peculiarly “communal” organization of basic economic activities, along religious and caste lines, in which agricultural land and organization of its cultivation is almost exclusively in the hands of one community whereas trading and industrial capital is owned primarily by the other community, may not be incidental to the nature and process of capital formation in the economy. Economic growth stemming from a particular socioeconomic scenario and benefiting different social groups or classes differently may lead to a reassertion of social identities and a more marked social differentiation than before. This social differentiation may also be more marked than the two-class laborversus- capital models suggest. Therefore, rather than breaking down social barriers to development, it may increase the social distance between the classes, which is often reflected in the spatial segregation of classes (Singh, 1993 and 1999). I argue here that this kind of economic growth may have strong implications for the labor market and its mobility and in the making of the new working class as industrialization progresses. I argue that rural-urban labor mobility (which may not always be rural to urban but may even be urban to rural) is unlikely to be a neat two-class (capitalist and pauperized), state-led affair. It will also not be a free movement of labor even if the state leaves it to the market, because the market process is likely to be mediated by the social institutions and identities mentioned above. In fact, the state may even use traditional social institutions to restrict or guide labor mobility, as China has used Hukou, the age-old system of household registration kept alive by successive dynasties and regimes to this day.

Social Structure and Labor Mobility

The role that institutions like Hukou and caste play in the industrial transformation of socially segmented economies like China and India needs to be acknowledged and understood. These institutions may also have affected the differential performance of the Chinese and Indian economies in recent times. In China’s standard story of reform—led development, focused more on foreign direct investment and international trade, the role of indigenous, local, and traditional institutions is inadequately considered, at least by the economists. In India, the economists have almost completely avoided issues such as caste and religion and have conveniently shielded themselves behind the rhetoric that “caste is class.” Hukou, which can be regarded more as an entrenched social institution than as a mere state instrument, obviously has had a direct bearing on labor mobility in contemporary China. Indeed, it seems to have been primarily used in recent times for regulating labor mobility.

China’s is not a simple case of massive migration and massive controls. The way Hukou has been used in China not to restrict but to guide labor mobility is striking. The state arranged cheap migrant labor for basic urban sectors like construction and infrastructure and for major manufacturing sectors. It liberalized or practiced Hukou selectively in ways that allowed labor to move whenever and into whichever regions and sectors cheap labor was needed (indeed, labor has actively been encouraged and organized by the state from different regions and provinces). This has been the extent of the liberalization, if that is the proper term, of Hukou. Like the earlier marketization of the grain sector, Hukou is one of those areas of China’s reforms where liberalization has taken longer to come. This is similar to the opening up of foreign investment in India—despite policies supporting it and claims to the contrary, foreign investment has not really taken off. However, it is important to note here that in India the state has almost no policies for regulating migrant labor.

China’s migrant labor policy successfully maintained huge reserves of labor, thus keeping wages down and creating the country’s celebrated low-cost advantage in the manufacturing sector. China even created a huge bureaucracy to guide the movement of migrants. Even recruiting is partly done by state agencies. This guided movement of labor from the rural/agricultural sector to the urban/industrial sector could be seen as an unusual way of primary accumulation from agriculture. With its vibrant foreign investment and exports, China’s industrial sector perhaps did not need a financial surplus from agriculture and a home market as much as it needed cheap labor. The Indian economy, on the other hand, has a more inward orientation; for this reason, a financial surplus from agriculture and a home market have been very important in India. China has been successful in mobilizing labor easily and cheaply. This success has probably been helped by a sense of nationalism inspired by the idea of building a new China. In India, industrial labor is comparatively expensive, highly unionized and politically affiliated, and widely seen as an active enemy of reforms.

In China, sectoral and spatial labor transfer is definitely and in a planned way linked with the so called Xiaokang project, which aims to create a reasonably well-off society by 2050 and to quadruple the GDP by 2020 from the 2000 level. Project planners have established 7 percent as the target for the annual growth rate and 1 percent as the target for the yearly shift of the labor force from agriculture to the nonagricultural sectors. This does not seem to be impossible, for the Chinese have done it before—in the 20-year period preceding 2000. Interestingly, there are no such targets or targeted policies in India directly related to shifting labor from agriculture. It is expected that as agriculture develops and the urban industrial sector expands its employment-generating capacity, labor from agriculture will move out.

Through its use of Hukou in regulating labor mobility in the country, China has forced a particular kind of working class to be formed and kept the peasantry more or less back. (In addition to Hukou, China has accomplished this through “rurbanisation” and a rural industry led by the so-called Township and Village Enterprises, which have, however, grown weaker of late.) Because of these effects, Hukou might have perpetuated, if not sharpened, the deep-rooted rural-urban distinction in a social structure traditionally and historically dominated by the urban gentry. This particular mode of primary accumulation from agriculture might have had a lot to do with social mobility (or immobility) of classes, too, just as caste affects social mobility in India. Rural-urban division itself is very caste-like and extremely deep-rooted both in China and India.

Of course, Hukou and caste are two different things. Yet in some sense, Hukou is caste-like. Both are forms of social exclusion. Both have a hierarchy and an identity aspect. Both eventually control the social mobility of classes. Caste may well be performing Hukou-like functions in restricting or guiding labor mobility. In India, everybody knows who is whom and where people are from (if people don’t know these things about others, they try to find out!). The Indians do not have to keep a Hukou to identify and exclude the ruralites or migrants. Name, surname, caste, religion, racial features, and provincial identity give one away. Whereas the Mingongs of China can be recognized by their appearance also, in India the migrant identity is mixed with other identities too, such as caste, occupational, provincial, and communal identities. (Note the social invisibility of Mingongs in China: they are hardly noticed in everyday urban life despite their pervasive presence. This is somewhat like the social attitude toward poverty in India.) In India, there are landed and nonlanded castes, menial castes, artisanal castes, merchant castes, and so on, all of whom are identified with different regions. Of course, caste is much more pervasive and far-reaching in its social and economic role than Hukou. However, Hukou may just be one of the many social institutions associated with the divisions and segmentations that exist in Chinese society. It may be that the social class divisions in China are subtler and less visible than in India’s caste system. This could be a result of, among other things, the different nature of occupational status in China, the blurred distinction between the state and society, and the scholarly preoccupation with the state and the party.

The traditional urban gentry of China, in whose hands politics, bureaucracy, and even military continue to rest are like the Brahmins of the Middle Kingdom. The caste-like distinction between the gentry and commoners in China has persisted through the years regardless of the political regime in power. This distinction may have been perpetuated in recent times by keeping the commoners back, spatially and occupationally, through Hukou and through Danwei. The distinction may also have been reinforced through controls on higher education, which have helped form the educated elite. It would be an injustice to the interesting and complex social structure of China to ignore social distance between classes and to insist that the division is only between the rich and the poor. This would be akin to the argument that “caste is class” in India.

It is hard to imagine an India with Hukou (more difficult, perhaps, than imagining a China without Hukou). What would Mumbai, Kolkata, or Delhi be if there had been a Hukou to restrict the influx of outsiders? Or what would Delhi be without the in-migration of Bengalis, Punjabis, and Biharis, which are only three of the major groups of “outsiders” who are now numerically strong and comfortably claim to be Delhites. (The migration of these groups to Delhi is similar in ways to the privileged interregional Han movement and large-scale migration into Manchuria from other parts of China during the early twentieth century.) In Delhi’s case, Bengalis’ in-migration is associated with the early twentieth century shift of the British from Calcutta to Delhi; the British brought along the Bhadralok Bengali bureaucracy with them. Punjabis’ influx is associated with the Partition of India and the creation of Pakistan in 1947 and later to other sociopolitical factors and movements in Punjab. The case of the Punjabi migration to Delhi is perhaps somewhat different because of the Punjab’s physical proximity to Delhi; in fact, Delhi once was a part of the Greater Punjab. The influx of the Biharis is a more recent phenomenon—from the past 20 or 25 years. Occupationally, Punjabis are more likely to be found in the service, industry, and white-collar sectors, whereas Biharis hold a mixture of manual and informal service sector and white-collar jobs. It seems that India has had a much longer and perhaps a more interesting history of market-led internal labor mobility, of all types, including rural-urban. Spurts of economic development in different regions, social upheavals, and catastrophes have also periodically led to large—scale population movements of working people, both blue- and white-collar.

Yet the seeming “free-for-all” labor market in India might be deceptive. Just because there is no Hukou, or household registration system as such, in India, Indian labor is not necessarily free to move. In fact, social structure does restrict or hasten labor movement in India. In addition, factor markets in India (as also in other developing economies such as China) are far from integrated, and the dictum that labor—most obviously manual labor—moves from comparatively less developed areas to the more developed ones holds true in India, as well. Ethnic, religious, caste, cultural, and linguistic divisions and their corresponding identities (market imperfections of their own kind, one could say) may effectively restrict labor movement or even encourage some other kind of movement. Labor from certain areas may be more or less welcome in certain other areas. The state may not interfere directly (as the state in China does) and may uphold the idea of the free market and democracy, yet the demands of social structure would dictate who would leave the countryside and how they would be welcomed or “settled” in the cities (for the mostly poor and lower-caste migrants, survival in the market environment is already disadvantaged). Once in the cities, it is not uncommon in India for migrants to organize themselves around their caste and community identities, as much to seek and protect their rights in the city as to let their community organizations serve as networks for future incoming migrants. (The idea of political rights is apparently meaningless in China. In India, even the poorest of the poor migrants get their voting rights almost immediately upon arrival in the city, and it becomes politically not only incorrect but undesirable to evict them.) No wonder labor in India can be observed to be moving in caste and community groups. That is true as much of agricultural as of nonagricultural labor, of Bihari farm laborers in Punjab or Malayalee 11 Stenos and nurses across the country. The phenomenon is helped by community-based identities and networks on the one hand and by caste and regional discrimination in the job market on the other hand.

It is not surprising that migrant manual labor in Punjab has come from outside and from a distant, impoverished, and itself a violently casteridden region like Bihar (and some areas of Eastern Uttar Pradesh near Bihar). Punjab’s and Bihar’s economies are interesting contrasts. One is a classic Green-Revolution, rapidly grown economy; the other is “backward.” One is known for semi-feudalism and the other for enterprising peasant proprietors. The labor migration that has occurred between them represents the most pronounced and the best-known case of interprovincial rural labor migration in India. The rural laborers from Bihar were economically deprived and were members of the lowest castes; there was hardly demand for them in their own social and economic space. With few opportunities in the neighboring region, these illiterate and socially excluded laborers literally took the train to far-off Punjab, which was the most prosperous and booming province in the country at the time. In the process, they were not just meeting an economic demand for labor in Punjab’s agricultural and nonagricultural sectors but they were also filling a social space where local peasants and artisans refused to join the ranks of the proverbial industrial proletariat, contrary to expectations.

The peasants in Punjab have refused to become workers in the factories. Instead, they have aspired to industrial entrepreneurship or whitecollar jobs. They have not gotten access to the trade sector, which is in the hands of the traditional business community. Some have clung to their land; some have migrated abroad, legally and illegally. The decisions they have made regarding occupation should be understood not in terms of inflated aspirations or lack of industrial enterprise but in terms of the mobility of social classes. To be poor would still be preferable, or less demeaning, than being socially at the bottom of the hierarchy. The Jat Sikhs have been a dominant caste in Punjab. If they become factory workers, they slip down the social ladder. Sectoral labor mobility in terms of occupations is closely tied with social mobility of classes. The compulsions of social hierarchy may restrict or hasten sectoral, occupational, or spatial mobility 12.

As a whole, caste and religious discrimination is a fact of life in the Indian labor market. Workers belonging to certain castes or minority communities may not even expect to work for, or may prefer not to work for, nonstate employers of certain other communities or groups. In fact, entrepreneurs themselves go by strong regional, provincial, and ethnic identities, such as Marwari, Punjabi, and South Indian. It is not uncommon to find certain social groups flooding particular sectors of the job market, both white—and blue—collar. Certain communities dominate economic sectors such as trade and industry (such as the case of the Malayalee Stenos and nurses across India, as mentioned above, or the case of trade and industry in the hands of different social classes, as shown in the previous section). Labor (also capital) in India has a tendency to move together in their social groups both sectorally and spatially. When rural laborers move to the cities, they are not necessarily taking the jobs that the city dwellers do not want—the so-called leftover or undesirable jobs—as many researchers believe. The job choices they make are more likely to be influenced by the imperatives of the social structure—perhaps only those who have nothing to lose in their social status would be inclined to accept the jobs that no one else wants. Migrants often compete with the locals, and sooner or later some of them succeed. Migrants may work in the informal service sector because this sector may provide an entry point; the formal sector, especially the private sector, may be “protected,” with its jobs reserved for particular caste groups and communities. The informal service sector may also be attractive to migrant workers because it has a socially ambiguous status.

It is interesting to note that, despite social and cultural differences, such as those of caste and religion, or maybe because of these differences, India as a whole seems to be more receptive to internal migrant labor. Migrant labor exists on the fringes but is allowed and tolerated. This is true not only because there is no Hukou.

Of course, there are occasional protests in some provinces and cities against “outsiders.” In Punjab, certain Sikh groups wanted the rural migrant laborers to leave because they believed that the presence of the migrants, who are Hindu even if largely lower caste, would threaten the Sikhs’ numerical dominance in Punjab. In addition, there are periodic outbursts against outsiders in Mumbai for cultural, demographic, and economic reasons.

However, Indian cities seem to be absorbing migrants more permanently, more democratically, and into a wider range of occupations than Chinese cities are doing. Indian cities also give migrants more physical space, too. There are “migrant villages” both in Chinese and Indian cities, but migrants in Indian cities seem to be more spread out. Slums are only one part of the story, although these get more attention. This is not to deny, however, that industrialization projects in both China and India are using forced or hostage labor. The Indian economy has yet to effectively come out of the traditional, socially determined forms of labor organization. This is evident from the remnants of “attached” labor in Indian agriculture (“attached” through caste) and the persistence of “bonded” labor here and there in both rural and urban locations of India.

Bibliography

    AndorsStephen1977China’s Industrial Revolution: Politics Planning and Management 1949 to the Present (New York: Pantheon Books).

    • Search Google Scholar
    • Export Citation

    AuChun-Chung and VernonHenderson2002How Migration Restrictions Limit Agglomeration and Productivity in China,NBER Working Paper No. 8707 (CambridgeMassachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    BanerjeeBiswajit1983Social Networks in the Migration Process: Empirical Evidence on Chain Migration in India,Journal of Developing Areas Vol. 17 pp. 18596.

    • Search Google Scholar
    • Export Citation

    BianYanjie and JohnLogan1996Market Transition and the Persistence of Power: The Changing Stratification System in Urban China,American Sociological Review Vol. 61 No. 5.

    • Search Google Scholar
    • Export Citation

    CaiFang2000China’s Migration (Zhengzhou, China: Henan People’s Press).

    CaiFang2001Institutional Barriers in Two Processes of Rural Labor Migration in China,Working Paper No. 9 (Beijing: Institute of Population and Labor Economics, Chinese Academy of Social Sciences).

    • Search Google Scholar
    • Export Citation

    CaiFang2005The Rural-Urban Income Gap and Critical Points of Institutional Change,Social Sciences in China Vol. 26(Autumn), pp. 93111.

    • Search Google Scholar
    • Export Citation

    CaiFang and MeiyanWang2004Changing Labor Participation Rates in Urban China and the Policy Implications,Social Sciences in China Vol. 25 (Winter).

    • Search Google Scholar
    • Export Citation

    CostelloMichael A.1987Slums and Squatter Areas as Entrepots for Rural-Urban Migrants in a Less Developed Society,Social Forces Vol. 66 No. 2 pp. 42745.

    • Search Google Scholar
    • Export Citation

    DewarMargaret1956Labour Policy in the USSR 1917–1928 (London: Royal Institute of International Affairs).

    DongXiao-yuan and PaulBowles2002Segmentation and Discrimination in China’s Emerging Industrial Labor Market,China Economic Review Vol. 13 No. 2–3 pp. 17096.

    • Search Google Scholar
    • Export Citation

    “Don’t Let Delhi Go Mumbai Way: SC”2005Times of IndiaSeptember8.

    FaureDavid2002Town and Country in China: Identity and Perception (New York: Palgrave in association with St. Anthony’s College, Oxford).

    • Search Google Scholar
    • Export Citation

    FeiHsiao-tung1980China’s Gentry: Essays in Rural-Urban Relations (Chicago: University of Chicago Press).

    Ford FoundationLabor Mobility in China A Review of Ford Foundation Grantmaking 1997–2001 (Beijing: Ford Foundation, undated).

    Government of India’s National Sample Survey Organisation (NSSO1996)

    Government of India’s National Sample Survey Organisation (NSSO1989)

    GuChaolin and HaiyongLiu2002Social Polarization and Segregation in Beijing,in The New Chinese City: Globalization and Market Reformed. byJohnR. Logan (Oxford: Blackwell Publishers).

    • Search Google Scholar
    • Export Citation

    GuangLei and LuZheng2005Migration as the Second-Best Option: Local Power and Off-Farm Employment,China Quarterly Vol. 181 (March) pp. 2245.

    • Search Google Scholar
    • Export Citation

    HarrisJohn R. and Michael P.Todaro1970Migration, Unemployment, and Development: A Two-Sector Analysis,American Economic Review Vol. 60 No. 1 pp. 12642.

    • Search Google Scholar
    • Export Citation

    HintonWilliam1990The Great Reversal: The Privatization of China 1978–1989 (New York: Monthly Review Press).

    HoweChristopher1971Employment and Economic Growth in Urban China 1949–1957 (CambridgeUnited Kingdom: Cambridge University Press).

    • Search Google Scholar
    • Export Citation

    HughPatrick ed.1976Japanese Industrialization and Its Social Consequences (BerkeleyCalifornia: University of California Press).

    IredaleRobyn2000China’s Labor Migration since 1978,in Contemporary Developments and Issues in China’s Economic Transitioned. byCharlesHarvie (New York: St. Martin’s Press).

    • Search Google Scholar
    • Export Citation

    JasnyNaum1961Soviet Industrialization (Chicago: University of Chicago Press).

    JieFan and WolfgangTaubmann2002Migrant Enclaves in Large Chinese Cities,in The New Chinese City: Globalization and Market Reformed. byJohn R.Logan(Oxford: Blackwell Publishers).

    • Search Google Scholar
    • Export Citation

    JohnsonD. Gale2002Can Agricultural Labor Adjustment Occur Primarily Through Creation of Rural Non-farm Jobs in China?Urban Studies Vol. 39 No. 12 pp. 216374.

    • Search Google Scholar
    • Export Citation

    LewisA.1954Economic Development with Unlimited Supplies of Labour” (Manchester School of Economic and Social StudiesMay).

    LewisW. Arthur1955The Theory of Economic Growth (London: Allen and Unwin).

    LinJustin FangCai and ZhouLi1996The China Miracle: Development Strategy and Economic Reform (Hong KongSAR: Chinese University Press).

    • Search Google Scholar
    • Export Citation

    MazumdarDipak1983Segmented Labor Markets in LDCs,American Economic Review Papers and Proceedings Vol. 73 (May) pp. 25459.

    • Search Google Scholar
    • Export Citation

    MengXin and JunsenZhang2001The Two-Tier Labor Market in Urban China,Journal of Comparative Economics Vol. 29 (September) pp. 485504.

    • Search Google Scholar
    • Export Citation

    MooreBarrington1966Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World (Boston: Beacon Press).

    • Search Google Scholar
    • Export Citation

    NeeVictor1996The Emergence of a Market Society: Changing Mechanisms of Stratification in China,American Journal of Sociology Vol. 101 No. 4.

    • Search Google Scholar
    • Export Citation

    O’LearyGrey2000Labor Market Developments,in Contemporary Developments and Issues in China’s Economic Transitioned. byCharlesHarvie (New York: St. Martin’s Press).

    • Search Google Scholar
    • Export Citation

    PerryElizabeth and Xiaboeds.1997Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective (ArmonkNew York: M.E. Sharpe).

    • Search Google Scholar
    • Export Citation

    PingHuang1996Hukou System and Rural-Urban Migration in China,CERES the FAO Review.

    PingHuang and Frank N.Pieke2003China Migration Country Study,paper presented at the Regional Conference on Migration Development and Pro-Poor Policy Choices in Asia Dhaka BangladeshJune21–24.

    • Search Google Scholar
    • Export Citation

    SinghArvinder1993Peasants, Merchants and Accumulation: A Study of the ‘Awkward Classes’ in the Context of Rapid Agricultural Growth” (Trivandrum, Kerala, India: Centre for Development StudiesApril).

    • Search Google Scholar
    • Export Citation

    SinghArvinder1999Industrial Transition in an Agricultural Surplus Region: A Study of Punjab” (Ph.D. dissertation; Trivandrum, Kerala, India: Centre for Development StudiesMay).

    • Search Google Scholar
    • Export Citation

    SinghArvinder2003aChina Is Far Ahead: But Did India Run?” (unpublished; Beijing: China Centre for Economic Research, Peking University,January13).

    • Search Google Scholar
    • Export Citation

    SinghArvinder2003bChina: One Goal, Two Policies, Three Problems, Four Aspects,The Financial ExpressNew DelhiJune23.

    SinghArvinder2003cThe Xiaokang Pursuit: Will Agriculture Help or Obstruct?paper presented at an international conference on “China in the 21st Century: Chances and Challenges of Globalization” Institute of Far Eastern Studies Russian Academy of SciencesSeptember23–25.

    • Search Google Scholar
    • Export Citation

    SinghArvinder2004aMuch to Celebrate, Much to Take Stock,” China Report, Vol. 40 No. 4 pp. 35155.

    SinghArvinder2004bUna carrera con China?La Vanguardia Dossier Vol. 12 (July–September).

    SinghArvinder2005Comparisons Between China and India,China and the World Economy (May–June).

    SkinnerG. William1977Urban Development in Imperial China,in The City in Late Imperial Chinaed. bySkinner (Stanford, California: Stanford University Press).

    • Search Google Scholar
    • Export Citation

    SrivastavaRavi S.2005Bonded Labor in India: Its Incidence and Pattern,Working Paper No. 43 (GenevaInternational Labour Office,April).

    • Search Google Scholar
    • Export Citation

    SrivastavaRavi S. and S.K.Sasikumar2003An Overview of Migration in India, Its Impacts, and Key Issues,” paper presented at a Regional Conference on Migration, Development and Pro-Poor Policy Choices in Asia, Dhaka, Bangladesh,June21–24.

    • Search Google Scholar
    • Export Citation

    TawneyR.H.1935Land and Labor in China (Boston: Beacon Press).

    TodaroM.P.1969A Model of Labor Migration and Urban Unemployment in Less- Developed Countries,American Economic Review Vol. 59 pp. 13848.

    • Search Google Scholar
    • Export Citation

    TodaroM.P.1989Economic Development in the Third World (New York: Longman4th ed.).

    United Nations Educational Scientific and Cultural Organization (UNESCO)2004Together with Migrants,Research Action Project (BeijingJanuary).

    • Search Google Scholar
    • Export Citation

    WangFeng and XuejinZuo1999Inside China’s Cities: Institutional Barriers and Opportunities for Urban Migrants,American Economic Review Papers and Proceedings Vol. 89 (May) pp. 27680.

    • Search Google Scholar
    • Export Citation

    WeberMax1964The Religion of China: Confucianism and Taoism trans. anded. byHansH. Gerth (New York: MacMillan).

    YangMartin M.C.1969Chinese Social Structure: A Historical Study (Taipei: Eurasia Book Co.).

    ZhangZhongli1955The Chinese Gentry: Studies on Their Role in Nineteenth-Century Chinese Society (Seattle: University of Washington Press).

    • Search Google Scholar
    • Export Citation

    ZhangZhongli1962The Income of the Chinese Gentry (Seattle: University of Washington Press).

    ZhaoZhong2005Migration, Labor Market Flexibility, and Wage Determination in China: A Review,The Developing Economies Vol. 43 (June) pp. 285312.

    • Search Google Scholar
    • Export Citation
*

Arvinder Singh is a fellow at the Centre for the Study of Developing Societies, Delhi, India. An earlier version of this paper was presented at the China-India Roundtable at the Centre for Asian Studies, University of Hong Kong, September 22–23, 2005.

1

See the section entitled “Agricultural Surplus, Industrialisation and Social Structure” in Singh (1999, pp. 1–29) for a discussion of the centrality of price scissors in debates on industrialization in contemporary developing agrarian economies. The article also discusses the desirability, feasibility, and efficiency of this tool in extracting surplus resources from agriculture.

2

See “Agricultural Surplus, Industrialisation and Social Structure” in Singh (1999, pp. 1–29).

3

See “Agricultural Surplus, Industrialisation and Social Structure” in Singh (1999, pp.1–29). See also “Feasibility of High Accumulation in Peasant Agriculture” in Singh (1999, pp. 30–57).

4

See “Merchant Capital and Accumulation in Normal Commercialisation” in Singh (1999, pp. 58–82).

5

The data in Tables 11.1 and 11.2 were obtained from the census of 1981, the time around which the Green Revolution was at its height. Punjab’s population in 1981 was 16,788,915 and in 1991 it was 20,281,969. In 1981, 36.93% of the population were Hindus, 60.70% were Sikhs, and 2.37% were others. In 1991, Hindus were 34.46% of the population, Sikhs were 62.95%, and others were 2.59%. In the 1981 census, information on religion was collected according to the religion of the head of the household.

6

The information, both on religion and on workers engaged in trade and commerce in towns, was obtained from the 1991 census.

7

Otherwise, Tarn Taran in the district of Amritsar seems to be the only exception. In the Amritsar district/Majha, Sikhs have a more marked presence than they do in towns in Doaba, as in the districts of Jalandhar and Hoshiarpur.

8

Khatris have a strong taboo against working on iron or metals, which they consider to be the domain of the artisanal castes, mainly blacksmiths.

9

The data pertain to 1990–91 and were obtained from the Government of India’s National Sample Survey Organisation (NSSO, 1996). See Singh (1999, pp. 147–48).

10

The data pertain to 1985–86 and were obtained from the Government of India’s National Sample Survey Organisation (NSSO, 1989). See Singh (1999, p. 147).

11

Malayalee referes to people from the state of Kerala, who speak Malayalam.

12

Incidentally, Punjab’s case also shows that rural-urban labor mobility may not be as straight and simple as it appears. The laborers may traverse a long and complicated path, working in the rural sector or elsewhere before ending up in the urban industrial sector. Punjab attracted migrant labor both in the rural farm sector and in the manufacturing sector. However, a large section of the rural labor has “graduated” to the urban manufacturing sector over a generation. Also, the way migrants in Punjab have come to hold considerable political power in the province because of their numbers shows how considerations of electoral democracy affect internal migration in India.

    Other Resources Citing This Publication