Rural Urban Migration

Southeast Asian Studies, Vol. 49, No.
1, June 2011
Employment Structure and Rural-Urban Migration

in a Tamil Nadu Village:
Focusing on Differences by Economic Class
SATO Keiko*
Abstract
Since the mid-1990s, migration of workforces from rural to urban areas has accelerated in south India
accompanied by remarkable urban-based economic development. To investigate the nature of such rural-
urban migration in detail, especially any differences influenced by economic class, a study village was
selected from the Madurai District. The detailed analysis found the existence of clear inter-class difference
in terms of the shift to non-agricultural occupations; that is, the wealthier class tended to find more remu-
nerable non-agricultural jobs, such as white-collar jobs. The most striking finding was that the traditional
class structure in rural India based on ownership of farmland was basically unchanged even after non-
agricultural jobs became much more important. This was because of the huge expenditure for education
necessary to acquire remunerable jobs and the differential access to credit markets among the different
classes.
Keywords: rural-urban migration, employment structure, white-collar job, economic class, caste
Introduction
Since the mid-1990s, migration of workforces from rural to urban areas has accelerated in India, espe-
cially south India, with the remarkable urban-based economic development.1) The income which can
be earned in the non-agricultural sectors, especially by white-collar workers, is much higher than in the
agricultural sector. Thus, by shifting occupation from agriculture to non-agriculture, people can raise
their income substantially and mitigate poverty [Lanjouw and Murgai 2009; UNDP 2009]. Accordingly,
within the agricultural sector itself, there has been a surge in the wage rate and a shift from cereal
* 佐藤慶子，Ph.D. candidate, Graduate School of Asian and African Area Studies, Kyoto University, 46 Shimoadachi-
cho, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan
e-mail: [email protected]
1) According to Census of India 2011 Provisional Population Totals, the total population of Tamil Nadu stands 72.14
million whereas the rural population 37.19 million and the urban population 34.95 million respectively. Its
decadal growth rate during 2001–11 in rural and urban areas was 6.5% and 27.2% respectively [GOI 2011].
22
SATO K. : Employment Structure and Rural-Urban Migration in a Tamil Nadu Village
production to high value-added production such as livestock and horticulture.

Several studies have sought to understand the nature of such accelerated rural-urban migration in
India, including those analyzing government statistics, such as the National Sample Survey [Singh 1986;
Pal and Kynch 2000; Usami 2002; Mitra and Murayama 2008] and studies based on information obtained
through field surveys at the village level [Paul 1989; Walker and Ryan 1990; Landy 1992]. But due to
inherent difficulties in scrutinizing the process of rural-urban migration [Banerjee 1981; De Haan 1997],
especially in collecting information on migrants staying in other areas from the study village [Misra
2009], the details of this phenomenon are not yet fully understood.
The meaning of the term “details” here is at least twofold. One is how to capture the diversity of
non-agricultural jobs. Even within employed jobs (other than self-employed jobs), we need to at least
discriminate among white-collar jobs, blue-collar jobs and daily labor jobs, since labor conditions are
totally different among the three. The blue-collar job category is especially important in the case of
south India, including Tamil Nadu, where not only opportunities for white-collar jobs and daily labor jobs
but also a wide variety of blue-collar jobs have expanded rapidly. Blue-collar jobs include both regular
and non-regular working opportunities in factories, hotels, restaurants, workshops, bus service com-
panies and car rental companies, for instance.
The other meaning of “details” here is how to grasp the multiple steps of migration, that is, how
rural households transfer their various resources from their village to urban areas. First of all, higher
education (at least up to the 12th standard and increasingly up to the college/university or even the gradu-
ate school level) is becoming indispensable for acquiring remunerable non-agricultural jobs, not only
for white-collar jobs but also for some of the blue-collar jobs [Lanjouw and Murgai 2009]. The process
requires huge expenditure in education for extended periods of time, and how to finance it has become
a big question for all rural households [Lanjouw and Sharief 2004; Oka 2006]. In fact, parents engaging
in non-agricultural jobs are more active in spending income on their children’s education [Gidwani and
Sivaramakrishnan 2003]. On the other hand, parents in rural areas often have to sell/mortgage their
assets (such as land or livestock) to cover this expenditure. Furthermore, financing from relatives who
have already got remunerable non-agricultural jobs, particularly white-collar jobs, sometimes becomes
crucial [ibid.]. It can also be even observed that all the household members move to an urban area in
order to facilitate the higher education of their children. But note here that first, even if children finish
higher education, it is not easy for them to land remunerable jobs, especially in the government sector,
without having social networks already established within urban areas [Banerjee 1983]. Second, after
children acquire remunerable jobs in urban areas, questions arise regarding whether and how much they
remit to their parents. In some cases they start to assist higher education of their brothers and sisters.
23
東南アジア研究 49 巻 1 号
Even parents sell off their assets such as livestock and land then later on buy them back. Third, there
are different strategies employed after children get married and they (especially sons) form independent
households, remaining in urban areas. What kind of relationship they keep with the parents’ household
in the original village depends on various factors. Fourth, after retirement if sons do not return to the
village, and often call their old parent(s) to their household in the urban area, the whole rural-urban
migration process becomes completed. Otherwise, the process will be transferred to the next genera-
tion. In the former case, implications to the rural economy will differ depending upon how they deal
with their assets (especially land and houses) in their village.
Now, the key questions are as follows. Given the traditional, highly stratified rural society based
on unequal land ownership in India, 1) how economic class is related to the employment structure,
including self-employed agriculture, agricultural and other daily labor, self-employment in non-
agriculture, and white-collar and blue-collar jobs in non-agricultural sectors? 2) how households with
different economic class are adopting different strategy regarding the choice of jobs for members of the
younger generation, and thereby the attitudes toward the higher education of children? 3) to what
extent education is affecting the choice of jobs for the younger generation, and how various categories
of job differ in terms of labor conditions and economic return (income)? And finally 4) how much
money is necessary to finish higher education and what are the strategies employed to finance the
necessary education, which may differ among different economic classes?
In order to investigate the above questions, the author selected a study village in the Madurai
District, Tamil Nadu, one of the relatively successful states of India, and collected detailed data. The
composition of this article is as follows. The following chapter presents the outline of the study village.
Chapter II delineates the characteristics of households in the village in terms of ownership of farm-
land, durable goods and income by classifying them into five economic classes. Chapter III analyzes
the employment structure of the households, with special attention to migration to urban areas, by
classifying jobs into white-collar, blue-collar, and daily labor. Chapter IV focuses on the educational
requirements for non-agricultural jobs, and investigates the costs of higher education and the differ-
ences among the economic classes in financing such education costs. Finally, a summary and conclusion
follows.
I Outline of the Study Village
The study village (hereinafter, Si Village) is located approximately 34 kilometers southwest from
Madurai City via Thirumangalam Town (Fig. 1). It belongs to Sowdarpatti Gram Panchayat (hereinafter,
24
S GP), Thirumangalam Taluk of the Madurai District.2) S GP is one of the 52 GPs belonging to
Thirumangalam Taluk. According to the 2001 Population Census data, the population size of Madurai
City and Thirumangalam Town was 929 thousand and 44 thousand, respectively [GOTN 2009b].3) There
is a direct bus service 10 times a day from Si Village to Thirumangalam Town, which takes about 30
minutes. Another one hour is needed from Thirumangalam Town to Madurai City by other bus ser-
vices. However, a direct bus service from Si Village to Madurai City was started in 2009, enabling
villagers to reach Madurai City in less than 90 minutes.
Tamil Nadu State is famous for the development of manufacturing industries in India. Madurai
District is not an exception and, according to government statistics, there were 27,385 units of “micro
small medium enterprises (registered)” as of 31 March 2007 in the district [ibid.].4) The largest number
was classified as “Hosiery and Ready-made Garments” (24.8%), followed by “Food Products” (9.2%),
“Paper and Paper Products” (8.6%), “Metal Products and Parts” (6.6%), “Machinery and Parts except
Electrical” (5.3%), and “Rubber and Plastic Products” (4.3%). In and around Thirumangalam Town,
together with Madurai City, especially after the 1990s, there has been a rapid growth of industries such
as textile, match factories, as well as hand-made craft shops, which have grown to be a center for
attracting surplus rural labors [GOI 2001].5) The development of the service sectors was also observed
in the district.
S GP had 862 households and 3,131 residents in 2001 (Table 1). The table indicates that although
the number of households increased during the two decades between 1981 and 2001, the population
itself decreased6) as the average number of members per household declined from 4.37 to 3.63. Accord-
ing to the table, the percentage of workers in agriculture (farmers plus agricultural laborers) was still
more than 80% in 2001, but as argued later in detail, recently many villagers stay outside the village
(mostly in urban areas), engaging in various types of non-agricultural jobs, which are difficult to iden-
tify through the use of government statistics.
S GP is composed of 9 villages (hamlets) and divided into 2 revenue villages. Revenue Village I,
to which Si Village belongs, has an area of 1,677 hectares (4,144 acres) including 154 acres of wet
farmland (nañcai) in tank-command area ((ayacut) and 1,680 acres of dry farmland ((puñcai) (Fig. 1).
2) Gram Panchayatt is the lowest local administrative unit in rural India. In ascending order, the next administrative
unit is taluk and then district.
3) Madurai was the third largest city in Tamil Nadu next to Chennai (4,344 thousand) and Coimbatore (931 thou-
sand).
4) The number of units in the district was the sixth largest in Tamil Nadu, next to Chennai (56,912), Coimbatore
(54,471), Salem (44,245), Kancheepuram (37,531), and Erode (30,353).
5) Same are reported to the area closer to the Karur-Tirupur textile industry belt in the former Tiruchirapalli
district, Tamil Nadu [Djurfeldt et al. 2008].
6) The rural population in the Madurai District as a whole decreased by 6.2% during 1991 to 2001 [GOTN 2009c].
25
Fig. 1 Map of the Study Village

Source: Prepared by the author.
Table 1 Demographic Features of S Gram Panchayat
1981 1991 2001

No. of Households 795 868 862
Male 1,709 1,622 1,566
Population Female 1,769 1,666 1,565
Total 3,478 3,288 3,131
Average No. of members per household 4.37 3.79 3.63
Male 69 62 75
Literacy rate (%)
Female 41 37 54
Male 40 33 38
Cultivators among workers (%)
Female 24 35 31
Agricultural Laborers among Male 43 51 40

workers (%) Female 70 58 57
Male 17 16 22
Others among workers (%)
Female 6 7 12
Source: GOI [1981; 1991; 2001].
26
Table 2 Caste Wise Distribution of Households by the Size of Farmland Holdings in Si Village
No. of Size of Farmland (acre)

Caste (jati
( ) Category HH 0 0.01–0.60 0.61–1.80 1.81–3.60 3.61–6.00 6.01–9.00 9.01–18.00
Reddiyar BC 118 22 4 20 25 25 16 6
Naidu BC 1 1
Chettiyar BC 1 1
Asari MBC 8 3 2 3
Wannar MBC 5 2 1 1 1
Ampattar MBC 1 1
Source: Prepared by the author based on the field survey in 2008.
Notes: For the caste category see Oshikawa [1990], GOTN [2009a], and Ministry of Law and Justice [2009].
BC: Backward Caste, MBC: Most Backward Caste.
Fig. 2 Combination of Puñcai and Nañcai for All the Househols in Si Village
Source: Prepared by the author based on the field survey in 2008.
According to my household survey in 2008, Si Village had a total of 134 households and 421 persons,
with an average of 3.14 members per household. Out of the 134 households, 105 households (78.4%)
owned farmland; its caste ((jati) wise distribution is shown in Table 2. From the table, we can see that
although the village is categorized as a single-caste village dominated by the Reddiyar caste (who
occupied 87% of households and 92% of owned farmland), land distribution among the households was
fairly skewed. Fig. 2 shows the combination of nañcai and puñcai land for all the households. Only 58
households (43.3%) had nañcai land (mostly less than an acre), whereas 103 households (76.9%) owned
puñcai land, with large differences in their size (maximum 16 acres). Relatively few households owned
nañcai land, mainly due to its distance from the village.
27
The land lease market for nañcai land was active in the village. More than half (51.7%) of the
households which owned nañcai land leased-out all or a part of it. Out of 30 households that leased-out
nañcai land, 25 households and 7 households were under the contract of mortgage (otti) and tenancy
(
(saibogam), respectively.7) The major households that leased-in nañcai land were the Mooparr caste
living in V Village, one of the neighboring villages in S GP (Fig. 1). On the other hand, there was no
household which leased-in nañcai land among the villagers. Therefore, the number of households
cultivating nañcai land was only 28 (48% of total). In contrast, the land lease market for puñcai land
was inactive. Nine households (8.7% of puñcai land-owning households) leased-out puñcai land (under
mortgage contract) and 13 households leased-in puñcai land (under tenancy contract). As a result, 96
households (93% of total) cultivated their own puñcai land.
Fig. 3 shows the 25-year trend in cropped area of major crops in the Madurai District since the
1980. The cropped area of paddy, sorghum (cholam), and cotton has been decreasing whereas that of
maize has been increasing. The same trend can be observed in Si Village; the most popular crop was
maize (85 households), followed by sorghum (45 households) and cotton (41 households). Due to the
small cultivated nañcai land, paddy (28 households) was not so important. According to my interviews,
the cost for paddy production was relatively high compared to other crops,8) and also the risk of un-
germination due to the shortage of rainfall. Because of the tiny plot of nañcai land, lack of economies
of scale and the necessity for purchasing expensive well water made the villagers further avoid paddy
cultivation. On the other hand, in puñcai land, cotton cultivation, which was previously widespread, has
become less popular in recent years due to its labor-intensive nature; it requires lots of hired labor
especially for weeding and harvesting. Instead, a combination of maize cultivation and livestock rearing
(mainly goats) has become more popular, under which the villagers can give maize as a feed for goats
to minimize rearing costs [Sato 2011].9)
Note here that the major agricultural season in the study area was from September to March. April
to July was basically the agricultural lean season for both farmers and agricultural laborers, with the
exception of small areas under cotton cultivation. Therefore, during the dry season, villagers tried to
7) Two households made both tenancy and mortgage contracts. Tenancy can be divided into sharecropping and
lease. Under the sharecropping contract, both the production and the cost of current input are shared 50:50 by
tenants and landowners. Under the lease contract, usually Rs.1,000 per acre is paid to landowners as a land
rent. On the other hand, under the mortgage contract, Rs.10,000 is usually paid to obtain usufructuary rights of
an acre of land. The period is usually 3 years, when the principal money Rs.10,000 is repaid. But if landowners
wish to get land back by returning Rs.10,000, the contract can be terminated even before finishing the term.
Note that Rs.1–2 yen.
8) See also Table 8 in Jegadeesan and Fujita [2011] in this special issue.
9) Maize cultivation had another advantage of less labor-intensiveness, which is particularly convenient for farm
households with side jobs, such as temporary work at a factory or restaurant during the agricultural lean season.
28
Fig. 3 Trend of Sown Area of Major Crops in Madurai District

Source: Data from Department of Economics and Statistics (Tamil Nadu) [2008 (also from 1980–2005)].
Modified by the author.
Notes: Calculation is based on the following data: From 1980 to 1984 Madurai only, from 1984 to 1995 Madu-
rai and Anna (Dindigul), from 1996 to 2005 Madurai, Dindigul and Theni, respectively.
obtain off-farm jobs, working, for example, as waiters at restaurants or casual workers at factories in
Thirumangalam Town. As Sato [ibid.] showed, factory work has become more popular among the
villagers than goat rearing in recent years.
II Economic Class Structure in the Study Village
Using the wealth ranking method,10) we classified the total 134 households in the study village into five
economic classes: “upper” (15 households), “upper middle” (17 households), “middle” (37 households),
“lower middle” (39 households), and “lower” (26 households).11) The wealth-ranking method is a
technique to classify households by the subjective judgment of native villagers themselves, and is
therefore the most time-saving method for classification. Furthermore, its reliability is usually high,
10) Wealth ranking is a method of classifying village households into multiple economic classes by subjective judg-
ment of a villager who knows the village very well [Gardin 1988].
11) Wealth ranking was actually conducted by a 60-year old farmer who had 8 acres of farmland, belonging to the
Naidu caste. His wife, belonging to the Reddiyarr caste, was a leader of two Self-Help Groups (SHGs) in the
village. For details on SHGs, see Fujita and Sato [2011] in this special issue.
29
Table 3 Relation between Economic Class and Farmland Ownership
No. of Size of Farmland (acre)

Class HH 0 0.01–0.60 0.61–1.80 1.81–3.60 3.61–6.00 6.01–9.00 9.01–18.00
Upper 15 1 4 4 6
U. Middle 17 1 1 6 8 1
Middle 37 3 1 16 12 5
L. Middle 39 6 2 17 11 3
Lower 26 18 2 4 2
Total 134 29 5 22 29 25 17 7
Source: Fieldwork by the author in 2008.
Table 4 Relation between Economic Class and Poseession of Durable Goods
No. of Car/Motor Refrigerator DVD Radio Electric Phones Pucca

Class Armali
HH Bike /PC /CD Fan /Cell House
Upper 15 4 5 4 14 14 14 14
U. Middle 17 1 1 1 1 13 7 12 13
Middle 37 2 10 30 16 28 21
L. Middle 39 9 29 12 23 9
Lower 26 3 9 2 13 6
Total 134 5 6 7 23 95 51 90 63
Source: Fieldwork by the author in 2008.
since villagers themselves know each other fairly well. However, in order to persuade the reader, let
us examine the classification result by comparing it with some objective indicators.
First, we investigated the relation between economic class and asset ownership. Table 3 and Table 4
show the relationship between the economic class and the ownership of farmland12) and durable goods,13)
respectively. It revealed that wealthier households tended to own larger farmland and also more expen-
sive household appliances such as cars, motor bikes, and refrigerators. By contrast, almost all classes
of households except for the “lower” class owned low-priced home appliances such as armari (lockers)
and electric fans. The relatively wide diffusion of pucca (brick or concrete-made) houses can partly be
attributed to the IAY ((Indira Awas Yojona) program which subsidizes house-building for the poor.14)
12) As argued later, there observed a strong tendency that migration to urban areas was taking place from wealthier
households (same is reported by Greenwood [1971]), usually after selling their farmland. In fact, there were
some wealthy households with a very small size of farmland remaining who were going to migrate very soon.
Therefore, if classification is conducted based on the size of land holding it may sometimes be misleading.
13) When conducting a survey on holdings of durable goods in developing countries such as India, it is common to
check TV sets and bicycles [see Oshikawa 1996]. However, in the case of Tamil Nadu, TV sets were almost
exclusively diffused because of the government’s recent free distribution program. And a free bicycle was also
distributed to the households with students in the 9th standard or higher till FY 2010. Therefore, TV sets and
bicycles were not included in Table 4.
14) See Fujita [2011] in this special issue.
30
Table 5 Types of Major Jobs by Economic Class
Female Living in the Village (230 Persons) Living Outside (60 Persons)
No. of
Headed Non-
No. of Workers Blue- White- Blue- White-
Class Single Daily agricultural Daily
HH per Agriculture 1) collar collar collar collar
Income Laborers 2) Self Laborers 2)
HH Workers 3)
Workers 4)
Workers 3)
Workers 4)
HH Employment 5)
Upper 15 2.40 18 1 3 2 1 11
U. Middle 17 3 2.35 26 6 1 2 5
Middle 37 6 2.21 66 6 7 3
L. Middle 39 12 2.23 42 6 11 3 18 7
Lower 26 14 1.73 4 19 14 1 7
Total 134 35 2.16 156 25 38 4 7 1 38 21
Source: Fieldwork by the author in 2007–09.

Notes: 1) Including agricultural management and livestock rearing.
2)
Including agricultural laborers, wood chopping laborers, construction laborers and laborers hired by the
NREGA (Mahatma Gandhi National Rural Employment Guarantee Act) program.
3)
Including workers at mill and match factories, sewing factories, and restaurants, house cleaners, elec-
tricians, drivers of either truck or autorickshaw, conductors, guards, quarry managers, third class civil
engineers, midday meals cooks, metal workers, and washing workers.
4)
Including military officers, school teachers, nurses, accountants, office workers, company drivers, and
second class public servants.
5)
Including shop owners and merchants in the village.
Second, we investigated the relation between economic class and major jobs (and thereby income
level). Table 5 shows the types of major jobs according to economic class. Note here that the workers
“living outside the village” mean that they are “temporary migrant workers,” usually staying in urban
areas for work but often returning to the village to join their families. The table indicates that workers
from wealthier households are engaged more in white-collar jobs, rather than blue-collar jobs or daily
labor jobs, while workers in the “upper middle” and the “middle” classes are engaged more in self-
employed agriculture.
On the other hand, generally speaking, the type of job is the major determinant of income level
(Table 6). Monthly salary was more or less as follows: Rs.10,000–50,000 for white-collar workers,15)
around Rs.5,000 for blue-collar workers,16) around Rs.1,200–2,500 for daily laborers. Note also that
income from agriculture was lower than for white-collar workers, but usually higher than for blue-collar
workers or daily laborers.
15) However, a large wage difference existed among the white-collar workers. Workers in the IT sector with IT-
related college or graduate degrees in particular, earned a lot; education for this job requires more expensive
tuition fees (see for instance, the TNAU homepage [2010]). Indeed, most of the IT engineers working in the
US were graduates of the Indian Institute of Technology (or its affiliated colleges) and received almost the same
salary as US engineers. But there were no students or workers in the IT-related sectors in Si Village. So the
rough estimates of monthly salary (including stipend) of the white-collar workers in the village ranged from
Rs.10,000 to Rs.50,000.
16) Note that the permanently employed skilled factory workers in public sectors, although not found in the study
village, are classified here to blue-collar workers.
31
Table 6 Monthly Income Level of Various Occupations
Category Class Vocation and Its Salary1)

Army officers (12,000–50,000), School
I Higher grade professionals teachers (12,000–), Nurses (10,000–35,000),
Accountants
White-collar workers Office workers (including drivers, teleoperators
Lower grade professionals & store keepers at public/private firm:
II including administrators and 6,000–10,000), Public servants (first and
officials second class)
Managers at mill/match factories, Quarry
Routine non-manual employees supervisors, Third class civil engineers, Metal
I including managers, supervisors, workers, Sales persons at established shops
and business workers (4,500–6,000)
Blue-collar workers Workers at mill/match factories (temporary
4,000–5,000), Waiters at restaurants, Cleaning
persons (4,500–5,000), House servants,
II Routine manual employees Electricians (5,000), Autorickshaw or Truck
drivers, Midday meals organizer & cook
(1,300–2,500), Tailors (2,100), and Washer men.
Agriculture Farm management and livestock raring
Agricultural laborers (Male: 100–150, Female:
60–80), Wood chopping laborers (120),
Daily laborers Construction laborers (150), Laborers under
NREGA (80–100)
Non-agricultural self Glossary shopkeepers and Merchants
employment (agricultural product)
Notes: The categorization of the job is based on the modification of the class scheme presented on Erikson et al.
[1979]. The unit of income is in Indian Rupee (Rs.). Rs.1=2 yen.
Salaries are specified only where available.
1)
Daily wage in the case of daily laborers.
In sum, it can be concluded that the classification of households into the five economic classes by the
wealth ranking method was quite accurate, from the viewpoint of both asset holding and household income.
Fig. 4 shows the population distribution in the study village according to decadal age group. In the
figure, boxes in black indicate persons who stayed outside the village. More villagers, especially younger
generation in their 20s and 30s, were “temporary migrant workers,” who stayed mainly in urban areas
for non-agricultural jobs.
Fig. 5 is a map of the study village made by the author, in which all the households were classified
into the five economic classes. Major public facilities are also shown in the figure. The village had a
nursery17) and a primary-cum-middle school (1st–8th standard), a post office, a bus stop, a sub-health
center, a village pond, several temples, several grocery shops, a village hall, and a fair price shop under
the Public Distribution System (PDS). Compared to the neighboring villages, there were more pucca
17) Nursery was installed for the nutrition and pre-school education to pre-school kids by the government’s Tamil
Nadu Integrated Nutrition Program (TINP), which was funded by the Women and Child Development Project
(WCDP) of the World Bank. See Heaver [2002] for details.
32
Fig. 4 Population Distributions by Age Group in Si Village

Note: White indicates for those staying in the village, whereas black for those outside the village.
houses and paved roads18) which extended throughout the village.19)
III Job Structure and Migration Process of Households by Economic Class
III-1. Current Status

Table 5 in the previous chapter presented the type of major jobs for workers in the study village by
economic class. The table also indicated the existence of a large number (60 out of 290 workers or
20.7%) of “temporary migrant workers,” who emigrated alone to urban areas and engaging in non-
agricultural jobs. They came back to the village periodically and often supported their families. How-
ever, in addition to such temporary migrant workers, we also observed a non-negligible number (32) of
“independent migrant workers” who stayed outside the village with new independent households but
kept in close contact with their original families. They used to be “temporary migrant workers,” but
after they married they formed new households in urban areas and became “independent migrant
workers.” Finally, there were 30 households whose members used to stay in the village but by the time
of the survey in 2008, had left the village and settled in urban areas completely, returning only occasion-
ally for festivals and/or crop harvest. For convenience in this article, we term them “migrated-away
workers.” Some of them still had their houses in the village, which were mostly left empty. Typical
cases display the pattern of parents leaving the village to join their sons’ families in urban areas after
18) Some of the paved roads in the village were constructed by Sampoorna Grameen Rozgar Yojana (SGRY) program
(in FY 2007–2008). See GOI [2009] for details.
19) However, at the far end of the village, there were some streets with less extent of pavement where kucha
houses (made of mud and organic materials) were more dominant.
33
Fig. 5 Map of the Study Village

Source: Prepared by the author based on fieldwork in 2008.
34
the sons become “independent migrant workers.”

In sum, the typical process of migration of rural households to urban areas is as follows: first,
children (especially sons) move to urban areas to pursue higher education and then obtain some non-
agricultural jobs after graduation (temporary migrant workers); second, when they get married if they
form new households in urban areas they become “independent migrant workers”; third, after that if
they bring parent(s) to their residence to live together in order to take care of them they become
“migrated-away workers.” When parent(s) leave the village, they usually sell farmland and residential
houses, but some of them keep their assets for leasing-out or leave them unused. Therefore, the land
sales market was active in the study village. In fact, at the time of the survey in 2008, it was found that
51 households (38% of total) in the village answered that they had either sold or purchased farmland in
the past. The major direction of land transfer was from the wealthy to the less affluent, and from the
Reddiyarr of Si Village to the Mooparr or Ampattarr of the neighboring villages.
The following will now explain the details of the employment situation according to the category
of workers.
1) Workers staying in the village
As shown in Table 5, out of 230 workers who stayed in the village, the overwhelming number (156
persons; 67.8%) worked in self-employed agriculture, followed by blue-collar workers (16.5%), daily
laborers (10.9%), self-employed non-agricultural workers (3.0%), and white-collar workers (1.7%).
2) Temporary migrant workers
No one in this category was engaged in agriculture, and the largest number of workers could be
found among blue-collar workers (63.3%), followed by white-collar workers (35.0%) and daily
laborers (1.7%).
3) Independent migrant workers
Table 7 shows the number of households with “independent migrant workers” and the types of the
job they were engaged in. The places they stayed were as follows: 5 persons in Thirumangalam
Town, 3 in Madurai City, 2 in other major cities in Tamil Nadu, 10 in small rural towns and villages in
Tamil Nadu, and 4 in other states of India. The number of white-collar workers accounted for the
majority (50.0%), followed by blue-collar workers (37.5%) and self-employed non-agricultural workers
(12.5%).20) Each category included military, drivers and conductors in state railway company, various
20) Compared to the “temporary migrant workers,” the share of white collar workers was high. The interpretation
of it seems to be rather difficult. Deshingkar [2006] argued that “employment is the secondary motivation to
marriage” and suggested those who were lucky enough to get higher quality job have more desirable charac-
teristics for marriage and hence more easily become “independent migrant workers” (reported by Behrman et al.
[1995] as well). Anyway, the factors for explaining the phenomenon remain to be explored in the future.
35
Table 7 Migrated Relatives of the Sutdy Village
No. of HH Types of Job of Migrated Relative Household

No. of which Had
Class Blue-collar White-collar Non-agricultural
HH Migrated
Relatives Workers Workers Self-employment
Upper 15 5 4 1
U. Middle 17 6 1 5
Middle 37 9 1 5 3
L. Middle 39 9 7 2
Lower 26 3 3
Total 134 32 12 16 4
Notes: Each job category is explained as in Tables 5 and 6.
public servants, school teachers, engineers, a nurse (white-collar jobs), painters, tailors, mechanics
(blue-collar jobs), a photocopy shop owner, an agent at a real estate business, a news stand owner,
and metal workers/jewelry processer-cum-moneylender (self-employed non-agricultural jobs).
4) Migrated-away workers
According to my survey, the number of migrated-away workers was 30. Generally speaking, they
were highly educated and engaged in white-collar jobs in urban areas. In some cases, they still
owned fairly large farmland in the village. If we exclude the 3 households with unknown residential
arrangements, the places they stayed were: 13 households in Thirumangalam Town, 2 in another
major city in Tamil Nadu, and 12 in various towns and villages in Tamil Nadu. Eighteen households
(60%) had workers engaging in white-collar jobs; teachers (4 persons), military/policemen (6
persons), public servants in the power corporation, telecommunications, the public works depart-
ment (PWD) and the panchayatt union (6 persons), a medical doctor (1 person), and a nurse (1
person). In addition, there was a migrant worker in Dubai (1 person), 2 workers employed at
spinning mills, and a waiter at restaurants.21) Recently, the number of migrated-away workers from
the study village has been increasing, and they tend to send their children to a private school, well-
known for its English-based education from the primary level onwards, in urban areas.22)
The next summarizes salient features regarding the differences in employment structure according
to the five economic classes in the study village.
First, the “upper” class household members were most engaged in white-collar jobs among others:
21) He owned a tractor and conducted a tractor rental business.

22) Besides seeking better white-collar jobs in urban areas, seeking better educational opportunities for children
was becoming a major reason for migration from rural areas.
36
there were 3 persons from the first category (workers staying in the village), 11 persons from the
second category (temporary migrant workers), and 4 persons from the third category (independent
migrant workers). At the same time, 2 persons had a self-employed business in the village and 18
persons were engaged in self-employed agriculture. However, because of large farm size, they usually
did only managerial work and depended on agricultural laborers for their farming. Moreover, out of 18
farmers, 7 were people retired from their white-collar jobs in urban areas. After retirement they came
back to the village and embarked on farm management. Many of them were receiving pensions, and
they also played an important role in the village society as informal leaders.23) In the future, it seems
that at least some of the 11 temporary migrant workers now working in urban areas as white-collar
workers will eventually come back to the village after retirement.24)
Second, the “upper middle” and the “middle” class household members were engaged mostly in
self-employed agriculture; 92 persons (86% of all the workers in these classes) were from the first
category (workers staying in the village). They owned large farmland, although smaller than the “upper”
class, and were also active in leasing-in farmland. Compared to other classes, goat rearing was the most
popular activity among them [Sato 2011]. Also, many members were working as blue-collar workers
(12 persons each from the first and the second categories, plus 2 persons from the third category) and
white-collar workers (1 person, 3 persons, and 10 persons from the first, second, and third categories,
respectively). Compared to the “upper” class, they were more actively involved in farming and tended
to engage more in blue-collar jobs.
Third, the “lower middle” class household members were also engaged mainly in self-employed
agriculture; 42 persons (68% of all workers) were from the first category. However, their farm size
was smaller and thus they needed to find some other jobs to supplement their income. Blue-collar jobs
were dominant among them, although not a few people were engaged in daily labor and white-collar
jobs as well. In fact, the numbers of blue-collar workers were 11, 18, and 7 from the first, second, and
third category, respectively. In contrast, the number of daily laborers was 6 (from the first category),
whereas white-collar workers was 7 (from the second category) and 2 (from the third category),
respectively.
23) In the study village, there was an informal local autonomous body, the activities of which were supported by a
common fund. To manage this local body is one of the major tasks of the informal leaders. The major revenue
source of the fund was a tax imposed on sellers of various products (mainly crops and livestock) in the village.
The product-wise tax rates are determined in an annual meeting and the right to collect taxes is sold to the
highest bidders. The system is locally called “magemai. ” The common fund is spent mostly for village festivals,
which are celebrated usually once in 2 years. See Sato [2008] for details.
24) There were two distinctive groups in this class: one in which both parents and children were engaged in white-
collar jobs and the other in which parents were engaged in blue-collar jobs but children were engaged in white-
collar jobs.
37
Fourth, the “lower” class household members were mainly daily laborers; 19 persons (51% of all
workers) were from the first category and 1 person from the second category.25) However, there were
many blue-collar workers also; 14 persons, 7 persons, and 3 persons were from the first, second, and
third categories, respectively. Also notable was the fact that 4 persons were engaged in self-employed
agriculture, although their farm size was minimal.
It should be pointed out here that as shown in Table 5 a large number of households in the “lower-
middle” (30.8%) and the “lower” (53.8%) classes were the so-called female-headed households. They
were the poorest in the village, as can be observed extensively in rural India.26)
Although the number of daily laborers in the study village was small (Table 5), if we include workers
whose main jobs were something else but occasionally worked as daily laborers for supplementary
income, almost 100 such laborers were found in the whole village. Especially among the farm house-
holds with small farm size, there were many agricultural laborers working 10–20 days to earn around
Rs.1,200–2,500 per month. Therefore, daily labor was still a very important source of income in the
village, especially among the poorer households.
There are primarily two types of systems for hiring laborers: first, laborers are employed indi-
vidually when they were needed by farmers and paid on a daily basis and second, laborers are managed
by a labor group leader 27) working on a seasonal basis. For example, 5–7 laborers were recruited by a
group leader for weeding some plots and the wages were paid through the leader. In general, the pre-
vailing wage rate for the daily laborers was Rs.100–150 for men and Rs.50–60 for women, but for some
hard tasks requiring physical strength such as wood chopping, the rate was higher. During the agricul-
tural peak season (such as crop harvesting period) the wage rate tended to be higher than usual. Also,
under the NREGA, a 100-day employment guarantee program for the registered laborers in the GP
office, started in 2006 in the study village, the daily wage rate of Rs.80 (or Rs.100 since FY 2008–09)
was paid regardless of the gender of the laborers, which benefitted women more than men. An interview
with the clerk of S GP in January 2009 revealed that 134 persons from the study village were registered
for the scheme; most of them were females and either elders, newly married young females, or agri-
cultural laborers (from the four classes except the “upper”).
III-2. Some Historical Background

So far, the article has discussed a typical migration path, in which mainly wealthier Reddiyarr caste
25) See Mosse et al. [2005] for the case of the poor migrant daily workers for construction.
26) See Yagi [1999] for instance.
27) Leaders of labor groups were found in S, P, and M Villages in S GP (Fig. 1).
38
people migrated from the village to urban areas. Let us now consider this phenomenon in a historical
perspective. Table 8 contains a summary of the socio-economic structural change occurred in S GP,
based on my interviews. The number of households of S GP in 2008 is a rough estimate based on an
interview at the GP office,28) whereas that of Si Village is the actual number based on my household
survey in 2008.
As shown in the table, the socio-economic structure of S GP in the 1910s was characterized by the
dominance of the Brahmin29) who occupied most of the farmland. The castes of Reddiyar/Naidu,30)
Maravar/Kallar,31) and Vallayarr (Moopar)32) leased-in land from the Brahmin and cultivated it while
depending on Pallarr33) people, who were employed as agricultural laborers. In addition, there were a
variety of artisan caste people such as Chettiyarr34) and Asari,35) and service caste people such as Wannar
and Ampattar.36) However, since the 1920s or 1930s, the Brahmin people started to migrate to urban
areas for higher education and then obtained white-collar jobs such as lawyers, medical doctors, and
other public servants created under the British colonial regime.37) They sold their farmland mainly to
Reddiyarr and Kallar caste people who used to work as their tenants. According to a Brahmin still living
in S GP (interviewed in February 2007), there were around 60 Brahmin households for the whole GP
28) Total number of households was 1,360, although the Population Census data showed that it was 862 in 2001
(Table1). There is a big gap between the two, but the reason remained unclear.
29) Originally they were sent to the study area from a temple in Tanjore and provided with the nañcai land attached
to Sowdarpatti and Allaparachelli tanks, as salary for their works to perform pujya (ritual ceremony) and became
large landowners.
30) They are one of the dominant agricultural castes in south India known as “kshatriya” and are characterized by
active acquisition of farmland [Srinivas 1989]. The Reddiyarr people in S GP used to work as attached laborers
earlier, but later became tenants for the Brahmin households.
31) They are the dominant caste originally staying in Thirumangalam Town and Madurai City, and used to work as
attached laborers for the Brahmin people throughout the British regime.
32) They originally stayed in Ramanathapuram and used to represent 5% of the total population in Tamil Nadu
(Population Census 1971). They might have jajmani relations with service castes [Setty 1990]. In fact, in Si
Village, they worked for Reddiyarr caste people as their tenants, but later purchased land from the Reddiyar
people when they migrated to urban areas.
33) They were the rulers of Tamil during the 14th–15th century and later became cultivators of wet land. Though
not classified as SC in Andra Pradesh, in Tamil Nadu they were ranked the highest SC among all SCs, and were
discriminated by the other higher castes, while looking down and employing other SC people to work for them
[Ramaiah 2004].
34) They are a merchant caste originally from Pudukottai, a city between Madurai and Tanjore, but have migrated
overseas (especially to Malaysia and Burma) and become successful.
35) There were two types in Si Village: blacksmiths (cum moneylenders) and carpenters.
36) Wannarr and Ampattarr were originally untouchable castes, who were traditionally engaged in washing and hair
cutting, respectively. But depending on states, now some of them are categorized as the most backward caste
(MBC), and receive less merit under the reservation scheme [Oshikawa 1990]. The “migrated-away workers”
to urban areas had tended to be engaged in traditional caste jobs but are now engaged more in white-collar jobs,
especially among the younger generations.
37) This phenomenon was commonly observed in Tamil Nadu. See Yanagisawa [1996] for details.
39
40
Table 8 Summary of Historical Change in Rural Socio-Economic Structure in S Gram Panchayat
Occupational Trend in Agriculture and Non-agriculture by Caste in 1910s and 2010s

Caste (Community) No. of HH in 2007
1910s 2010s
Non-agri. Non-agri.
Administrative Caste (jati
( in Traditional S Gram Si Land Owner Agri Land Owner Agri
Classification Category General Names) Occupation Panchayat Village Owner Cultivator Tenant Labor Owner Cultivator Tenant Labor
Rural Urban Rural Urban
Priest FC Brahmin Priest 2 × × ×
BC Reddiyar/Naidu 400 119 × × × × ×
Cultivator BC Maravar/Kallar Cultivator 300 × × × × ×
MBC Vallayar (Moopar) 500 × × × ×
BC Chettiyar Merchant 10 1 × × ×
Craftsmen
MBC Asari Metal worker 10 8 × × ×
MBC Wannar Washer 10 5 × × ×

Service MBC Ampattar Haircutter 10 1 × × ×
SC Pallar Agri laborer 160 × × × ×
Source: Fieldwork by the author in 2007–09, Oshikawa [1990]; GOTN [2009a]; Ministry of Law and Justice [2009].
Notes: × denotes major occupational trend by respective caste in 1910s and 2010s, based on the broad and general observation and hearings.
Caste (jati
( in general names) appear is used in Madurai, Tamil Nadu. Under the reservation scheme by government, more quotas for the entry examina-
tion for the tertiary education and the placements for government jobs were introduced to SC/ST and other backward caste categories to mitigate the
economic and social dominance by upper caste people in India. See Oshikawa [1990].
The administrative categories used here are those used in Tamil Nadu. FC=Forward Castes, BC=Backward Castes, MBC=Most Backward Castes,
SC=Scheduled Castes, ST=Scheduled Tribes.
in the 1930s but the number decreased to around 20–25 by the 1980s, and finally declined to only 2
households in 2007.
As shown in Table 8, at the time of my survey, the dominant inhabitants in S GP were Reddiyar,
Maravar/Kallar, and Vallayar as owner-cultivators and Pallarr as agricultural laborers. However, as
already noted, a highly skewed land distribution was observed even among the same caste households
(e.g., Reddiyarr in Si Village) and many labor households were observed among the Reddiyarr households
as well. The “upper” class people of Reddiyarr started to migrate to urban areas for non-agricultural jobs
(mainly white-collar jobs). And they also started to sell their land to the next highest caste people and
also to the poorer households in the same caste. Thus, the farmland continued to be transferred from
the higher to lower caste/class people, and the structure of caste-based land ownership is not as rigid
as it used to be.
IV Education, Its Cost and Source of Financing
IV-1. Occupation and Educational Requirement

At present, one of the most important requisites for getting white-collar jobs, and thereby stable and
high income in the context of India, is to attain at least an HSC/ITI (finished up to the 12th standard, as
explained later) or even a college/university degree [Lanjouw and Sharief 2004; Oka 2006]. To confirm
this point in the study village, the years of education among village workforces, especially younger
generations, was investigated and the result for non-household head members38) is shown in Table 9 by
types of job and by economic class.
Note here that the education system in Tamil Nadu is as follows: after 5 years of primary education
(2 years of pre-school often precede), students move on to 3 years of middle school, followed by 4 years
of secondary school. The secondary school is divided into a 2-year curriculum leading to the Secondary
School Leaving Certificate (SSLC), and, later, another 2 years of higher secondary education for the
qualification exam, the Higher Secondary Examination (if one passes it he/she is granted the Higher
Secondary Certificate, or HSC). Once an SSLC is obtained, one can choose to take a 1 to 3-year voca-
tional course at Industrial Training Institutes (ITI).39) On the other hand, an HSC is necessary for
applying to college/university, which is usually a 3–4 year degree program. The entrance to college
38) One of the major reasons for dealing only with non-household heads is that in the case of household heads, there
were many aged farmers who had retired from white-collar jobs, especially among the “upper” class.
39) ITI is the government or private owned training organization that provides post-school technical training for
the persons who passed the SSLC.
41
Table 9 Average Years of Education for Non-household Heads in Si Village
Workers Staying in the Village Temporary Migrant Workers

(114 persons) (52 persons)
Class Non-
Blue- White- agricultural Blue- White-
Agriculture Laborers collar collar Laborers collar collar
Self-
Workers Workers Employment Workers Workers
Upper 8.9 8.0 10.0 11.9

U. Middle 7.4 9.4 12.0 8.0 10.2
Middle 6.1 7.3 11.2 12.3
L. Middle 5.3 6.0 8.2 10.0 8.9 12.3
Lower 6.2 6.5 6.1 9.0 8.0
Average 6.4 6.4 7.6 12.0 9.0 9.0 9.4 12.1
depends on the mark attained in the HSE but the condition for the placement (especially at state
universities) differs depending upon the caste categories.40)
As Table 9 shows, in order to obtain white-collar jobs, at least 12 years of education (HSC/ITI
holder) is necessary. We should note, however, that even blue-collar workers, especially the younger
generations working as “temporary migrant workers,” had had 10 or more years’ educational back-
ground. By contrast, for the workers in self-employed agriculture and daily laborers from poorer classes,
the average years of education ranged between 5 to 6 years, indicating either primary school graduates
or middle-school dropouts.
In the study village a primary-cum-middle school for 8-year education was established during the
1990s, and almost all the children started to finish at least up to the middle school level. As shown in
Table 10, the level of education increased year by year in the village, which was accelerated after the
1980s (for males) and the 1990s (for females). In the late part of the 2000s, almost 100% of children
obtained an SSLC (10th standard). Note that increasing consciousness of the need for higher education
is a nationwide phenomenon. In fact, in order to join the National Defense Academy (NDA), for example,
the minimum educational requirement for applicants was raised from 10th standard to 12th standard (the
so called “plus 2”) in 2007.
Let me now focus on the “temporary migrant workers” and “independent migrant workers” aged
40) According to the government reservation policy in Tamil Nadu, the minimum score for placement at state
universities is as follows: 60% for open competition including students belonging to Forward Castes (FC), 55%
for students belonging to Backward Castes (BC), 50% for students belonging to Most Backward Castes (MBC)
and “Pass” (regardless of marks) for students belonging to Scheduled Castes/Scheduled Tribes (SC/ST) etc.
See the TNAU homepage [2010] for detail.
42
Table 10 Percentage of SSLC Holders in Si Village by Age Group and Sex
early 20s late 20s 30s 40s 50s 60s 70s
Male 96% 76% 67% 15% 41% 18% 14%

Female 96% 72% 27% 24% 13% 3% 0%
Notes: For the age group of 70s, ESLC (= 8 years completion) and 4th Form, both of which are regarded
the same level as SSLC, are included.
between 15 and 30 to examine their educational background (Table 11). Nearly half of them (47% of
males and 46% of females) attained 12 years or more of education. Even in the case of workers with-
out an SSLC (less than 10th standard), most of them were found to complete 8 years of education. In
fact, among young people aged more than 20, higher education has become much more popular than it
appears in the table. But if we focus on the difference by economic class, poorer households show
lower educational attainment. Among the “lower” class households there was no one who entered
college/university.
IV-2. Cost of Education and Source of Financing

The cost of education has been surging in recent years in Tamil Nadu, including the study village.41) For
instance, as noted before, a private English school, offering courses up to the higher secondary level
(matriculation school)42) and located in Thirumangalam Town, attracted many parents in wealthy house-
holds who were concerned for their children’s education. In fact, all 10 students who went to the private
English-based school from Si Village (including 2 who had already migrated-away in 2010) belonged to
either the “upper” or the “upper-middle” class. They commuted to the school from pre-school or
primary school level. The entrance fee and the annual tuition fee of the pre-school43) were Rs.2,000
and Rs.6,000 respectively, equivalent to a few months’ wage for an agricultural laborer.
On the other hand, if children go to public schools, the tuition fee for primary education was free,
together with text books and lunch.44) However, payment of a tuition fee was necessary afterwards,
41) See Education Cost India [2010].

42) It is a kind of higher secondary school, with a curriculum which is aimed at enabling students to pass the HSC
with a high grade for entry into competitive placements at colleges/universities. According to Minamino [2003],
more than 80% marks on qualification exams such as the HSE are required to apply for the affiliated colleges
to the Indian Institute of Technology (IIT) and the Indian Institute of Management (IIM), the most competitive
ones in India.
43) These figures are the fee structure at St. Francis Higher Secondary Matriculation School in Thirumangalam
Town in which students of Si Village prefer to attend. See GOTN [2010a] for the list of the fee structure of all
the private school in Madurai District.
44) See GOTN [2010b] for details of the welfare schemes implemented for school education in Tamil Nadu.
43
Table 11 Educational Attainment of Young People Aged from 15 to 30 in Si Village
Male Female
No. of
Class No. of Below HSC College Graduate No. of Below HSC College Graduate
HH SSLC SSLC
Persons SSLC or ITI* /Univ.* School* Persons SSLC or ITI* /Univ.* School*
Upper 15 9 1 2 2 2 2 10 1 2 3 2 2
U. Middle 17 9 4 3 2 9 2 4 1 2
Middle 37 23 6 5 9 3 23 7 3 11 2
L. Middle 39 22 8 4 8 2 19 6 7 4 2
Lower 26 7 6 1 16 12 2 2
Total 134 70 21 16 22 9 2 77 26 16 24 7 4

Notes: Including temporary migrant workers and independent migrant workers.
* includes those who have not completed yet at the time of the survey. Also for the Graduate School,
BA+diploma of education is included.
amounting to Rs.500–650 and Rs.700–900 annually in the middle level (6th–8th standard) and the higher
secondary level (9th–12th standard), respectively. The annual tuition fee at the college/university has
become much more expensive, ranging from Rs.60,000 for Bachelor of Science to Rs.260,000 for
Bachelor of Information Technology (IT) for 4 years.45) But note that expenditures other than tuition
fees, such as living expenses, are relatively small compared to the tuition fees, since many students
stay in their relatives’ houses or in university hostels.
Given that the monthly income of blue-collar workers and daily laborers in the study village was
around Rs.5,000 and around Rs.1,200–2,500, respectively, there was no room left for saving money to
cover the cost of higher education for their children. Even for white-collar workers with monthly
incomes exceeding Rs.10,000, educational expenditure is a heavy burden on their household economy.
Furthermore, after graduating from college/university, in order to obtain white-collar jobs especially in
the government sector, a recommendation from a person (generally a relative or friend of parents) who
has special connection is indispensable, and a commission of roughly Rs.50,000 should be paid to
him/her as a custom in the study village.46) Therefore, access to credit sources is critical for households
across all classes.
45) See the TNAU homepage [2010]. However, the cost for college tuition differs, depending on type of public or
private institutes.
46) However according to Banerjee [1984] during the 1980s no commission was needed to get a job in the govern-
ment sector. There has been a rapid increase in the number of college graduates all over India, although there
is only a small increase in job opportunities. According to GOI [2007], the annual rate of increase of placements
was –0.54% and 0.92% in the public and private sectors during 1994–2006, compared to 1.54% and 0.44% during
1982–1994. In urban areas, more college graduates were obliged to take low-paid jobs in the private sectors
and continued to study through distant learning. Once they obtain higher degrees, they change their job for
better salary.
44
Table 12 Major Source of Financing the Cost for Education
No. of HH Source of Fund

No. of
Class with Valid Selling Bank Loan from
HH Savings
Answer Goat Loan Relatives
Upper 15 3 1 1 1
U. Middle 17 16 3 3 10
Middle 37 13 1 5 4 3
L. Middle 39 7 1 5 1
Lower 26 2 2
Total 134 41 6 13 8 14
Table 12 shows the sources of financing for expenditure on higher education by economic class.
Because of the small number of valid answers in the case of the “upper” class, the author gathered more
information by interviewing the “upper” class villagers. The result was that most of them did not face
serious problems in financing their children’s education because they were white-collar workers in
urban areas and could draw on their own salaries or savings. However, for the “upper middle” class
people, cases of education financed by their own savings were limited, and many of them depended on
relatives. They borrowed money from relatives who were working as blue-collar workers.
An example of a household in the “middle” class is shown in Fig. 6. In the case of this household,
there were two sons and a daughter and all of them studied up to college level. The education cost for
the first son was provided by relatives who worked at factories (as shown in (1)). When the first son
finished college, he got a job in Saudi Arabia and started to support the education of his brother and a
cousin on his mother’s side (both are shown in (2)). The second son also received support from a rela-
tive working as a small businessman, in the form of accommodations when he studied at a college in V
City (as shown in (3)). The youngest child, a daughter, studied at a nearby college, and therefore only
the tuition fee was needed. To finance three children’s college education, they additionally borrowed
Rs.50,000 from a jewelry shop in the urban area by providing jewelry as collateral and Rs.10,000 from
a bank through the Self-Help Group (SHG). Later, all the loans were repaid within five years by the
first son through remittance from Saudi Arabia.
On the other hand, as shown in Table 12, the poorer households depended more on sales revenue
of goats for financing higher education costs. If they sell 10 goats, for instance, they can get Rs.30,000–
40,000, although it is insufficient to cover all the expenses. Other sources of funding, although not
listed in the table, were farm income, “jewelry loans” from moneylenders outside the village (as men-
tioned above), and borrowing from banks through SHG. But it is important to note that although the
poorer households came to borrow from banks through SHG, as clarified by Fujita and Sato [2011], the
45
Fig. 6 Mutual Help among Relatives for Higher Education: A Case of a “Middle” Class Household
Source: Field work by the author in 2009.

Notes: (1) is loan from relatives who work at mills to the eldest son, who is shown in bracket , to support his
cost to study at college.
(2) is a support (money giving) from the eldest son, who is shown in bracket , who graduated from
college and now works in Saudi Arabia, to those who are/were college students, shown in dotted
bracket .
(3) is a support (lodging) from a relative who moved to V city to the second son, when he was studying
in V City.
limited amount of the loan (Rs.10,000) was far from sufficient to cover the costs of higher education for
their children.
Summary and Conclusion
In India, especially south India, the expansion of non-agricultural jobs for rural people has accelerated
since the mid-1990s, inducing a large-scale migration of workforces from rural to urban areas. To
investigate the nature of such rural-urban migration (especially in terms of differences by economic
46
class) and its impact on rural economies, a detailed case study was conducted in a village in the Madurai
District, Tamil Nadu.
Although the village in question was a single-caste village with nearly 90% of households belonging
to Reddiyar, the cultivator caste, there was a large economic disparity among the households and eco-
nomic disparity was observed not only in asset holdings (including farmland) but also in income level.
Out of a 290-member workforce in the village with 134 households, 60 (20.7%) stayed outside the
village, mostly in urban areas, working as “temporary migrant workers.” They were still single and
often came back to their parents’ residences in the village. Besides, however, there were 32 workers
who had already formed an independent household in urban areas after marriage. Furthermore, 30
households had already “migrated-away” from the village, and came back only occasionally for festivals
and/or crop harvest (in cases of owning farmland). Some of them still had houses in the village, which
were mostly left empty. Typical cases displayed the pattern of parents leaving the village to join the
families of their sons in urban areas.
Out of 60 “temporary migrant workers,” 21 were working as white-collar workers and 38 as blue-
collar workers. Similarly, out of 32 “independent migrant workers,” 16 were white-collar workers and
12 were blue-collar workers. In the case of the “migrated-away workers,” the share of white-collar
workers was 60%. By contrast, out of 230 workers staying in the village, 156 (67.8%) were engaged in
self-employed agriculture and another 25 (10.9%) were daily laborers, including agricultural laborers.
There were only 4 white-collar workers and 38 blue-collar workers. Since the study village was located
in a typically pure rural area, most of the non-agricultural workforce was obliged to stay in urban areas.
The distinction between white-collar workers and blue-collar workers is important in terms of income
level and relative stability of employment. Roughly speaking, the monthly salary observed among the
workforce from the study village was Rs.10,000–50,000 for white-collar workers, around Rs.5,000 for
blue-collar workers, and Rs.1,200–2,500 for daily laborers. Income from self-employed agriculture was
typically lower than for white-collar workers, but usually higher than for blue-collar workers or daily
laborers.
The total 134 households in the study village were classified into five economic classes; “upper,”
“upper middle,” “middle,” “lower middle,” and “lower.” The study found that the “upper” households
owned the largest farmland and were engaged in agriculture, but many of them were white-collar
retirees who were receiving pensions and also working as informal village leaders. Their children
usually attained higher education and studied or worked in urban areas as white-collar workers. The
“upper middle” and the “middle” class households, on the other hand, were the most active farmers in
the village, with a fairly large farmland of their own. They were also active in leasing-in land and goat
47
rearing was most popular among them. The education level of the parents generation was usually not
high, but they were also enthusiastic in their intention to provide higher education for their children.
However, the share of children who obtained white-collar jobs was much smaller than in the “upper”
class. So far, blue-collar workers were dominant in these households.
Although marginal farmers and daily laborers were still dominant in the “lower middle” and the
“lower” class households in the village, with the increasing education level among the younger gen-
eration in recent years, they have also started to work as blue-collar workers. At the same time,
however, a substantial share (53.8% and 30.8% of the “lower” and the “lower middle” households,
respectively) were female-headed households, the poorest of the poor, whose major problem was the
lack of male workers.
In sum, the most striking finding was that the rural class structure in the study village still largely
corresponded with the size of farmland ownership, in spite of the increased importance of non-agricultural
employment today. The key to understanding this phenomenon is the high cost of education necessary
to acquire white-collar jobs (or even some blue-collar jobs) and inter-class disparity in the access to
credit (including relatives who have already obtained remunerable non-agricultural employment) for
financing an expensive education. In other words, the high cost of higher education (especially in the
college/university level) is reproducing (or even expanding) the traditional structure of economic
disparity among households based on ownership of farmland.
However, contrary to the situation during the 1920s or 1930s, when Brahmin people migrated from
rural to urban areas in Tamil Nadu, rural-urban migration after the mid-1990s is much more widespread,
influencing almost all the economic classes, with the exception of some female-headed households, in
terms of increase in both white-collar and blue-collar jobs, and the impact of this change is much more
fundamental. One example is the labor shortage in agriculture, best exemplified by the rapid diffusion
of combine-harvesters for rice harvesting since the mid-2000s. Another important difference is that
the scheduled caste people, after receiving farmland through the central government’s policy to trans-
fer farmland on a large scale during the 1970s,47) were able to gain an advantage in seeking higher
education and government jobs by the reservation policy.
Acknowledgements
I would like to thank the following funding bodies which made the field work for this article possible during 2007–10:
Kyoto University Miryoku Program, Kyoto University Global COE Program “In Search of Sustainable Humanosphere
in Asia and Africa,” represented by Professor Kaoru Sugihara, Center for Southeast Asian Studies, Kyoto University,
47) This movement is called ““Bhoodhan movement.” See Sato [2011] in this special issue.
48
the Mishima Kaiun Memorial Foundation, and the Suntory Foundation. Valuable information was provided by Mr.
Alagarsamy and his family, and Mr. Raja and his family. Valuable comments were provided by anonymous referees.
Special thanks are extended to Professor Koichi Fujita, Center for Southeast Asian Studies, Kyoto University, who
intermittently supervised me during the whole process of the research.
References
Banerjee, B. 1981. Rural-Urban Migration and Family Ties: An Analysis of Family Considerations in Migration
Behavior in India. Oxford Bulletin of Economics and Statistics 43(4): 321–355.
―. 1983. Social Networks in the Migration Process: Empirical Evidence on Chain Migration in India. The
Journal of Developing Areas 17 (2): 185–196.
―. 1984. Information Flow, Expectations and Job Search: Rural-to-Urban Migration Process in India. Journal
of Development Economics 15(1–3): 239–257.
Behrman, J.; Birdsall, N.; and Delolikar, A. 1995. Marriage Markets, Labor Markets, and Unobserved Human
Capital: An Empirical Exploration for South-Central India. Economic Development and Cultural Change 43(3):
585–601.
De Haan, A. 1997. Rural-Urban Migration and Poverty: The Case of India. IDS Bulletin 28(2): 35–47. Institute of
Development Studies.
De Haan, A.; and Ben, R. 2002. Introduction: Migrant Workers and Their Role in Rural Change. Journal of Develop-
ment Studies 38(5): 1–14.
Department of Economics and Statistics (Tamil Nadu). 2008. Season and Crop Report of Tamil Nadu 2006–07. (also
from 1980 to 2005).
Deshingkar, P. 2006. Internal Migration, Poverty, and Development in Asia. Institute of Development Studies,
Sussex. Unpublished manuscript.
Djurfeldt, G.; Athreya, V.; Jayakumar, N.; Lindberg, S.; Rajagopar, A.; and Vidyasagar, R. 2008. Agrarian Change and
Social Mobility in Tamil Nadu. Economic and Political Weekly 43(45): 50–61.
Education Cost India. 2010. Cost of Education Rising in India for Common Man !!! Accessed on 1 December 2010
at: http://epravesh.co.in/?tag=education-cost-india.
Erikson, R.; Goldthorpe, J. H.; and Portocarero, L. 1979. Intergenerational Class Mobility and the Convergence
Thesis: England, France and Sweden. The British Journal of Sociology 30(4): 415–441.
Fujita, K. 2011. Introduction: Socio-Economic Dynamics in a Tank-Irrigated Rural Area in Contemporary Tamil
Nadu, India. Southeast Asian Studies 49(1): 3–21.
Fujita, K.; and Sato, K. 2011. Self-Help Groups and the Rural Financial Market in South India: A Case of a Tamil
Nadu Village. Southeast Asian Studies 49(1): 74–92.
Gardin, B. E. 1988. Wealth Ranking in Smallholder Communities: A Field Manual. Intermediate Technologies
Publication Ltd.
Gidwani, V.; and Sivaramakrishnan, K. 2003. Circular Migration and the Space of Cultural Assertion. Annals of the
Association of American Geographers 93(1)：186–213.
Government of India (GOI). 1981. Census of India 1981, Series 33 Tamil Nadu, District Census Handbook, Part-B
Madurai District, Village and Town-wise Primary Census Abstract, Government of India.
―. 1991. Census of India 1991, Series 33 Tamil Nadu, District Census Handbook, Part-B Madurai District,
Village and Town-wise Primary Census Abstract, Government of India.
―. 2001. Census of India 2001, Series 33 Tamil Nadu, District Census Handbook, Part-B Madurai District,
Village and Town-wise Primary Census Abstract, Government of India.
―. 2007. Economic Survey 2006-07. Oxford University Press.
―. 2008. Statistical Abstract India 2007. Central Statistical Organization, Ministry of Statistics and Program
Implementation, Government of India.
―. 2009. Annual Report 2007– 7 08 Bharat Nirman through Rural Development, Ministry of Rural Development,
Government of India. Accessed on 10 April 2011 at: http://rural.nic.in/annualrep0708/anualreport0708_eng.pdf.
―. 2011. Census of India 2011 Provisional Population Totals, Paper 2, Volume 1 of 2011 Rural-Urban Distri-
bution Tamil Nadu Series 34, Director of Census Operations Tamil Nadu, Government of India. Accessed on 10
49
April 2011 at: http://www.censusindia.gov.in/2011-prov-results/paper2/data_files/tamilnadu/Tamil%20nadu_

PPT2_volume1_2011.pdf.
Government of Tamil Nadu (GOTN). 2009a. Backward Classes, Most Backward Classes & Minorities Welfare Depart-
ment. Accessed on 10 November 2009 at: http://www.tn.gov.in/departments/bcmbc.html#web.
―. 2009b. Annual Statistical Abstract of Tamil Nadu 2007-08. Department of Economics and Statistics.
―. 2009c. Economic Appraisal 2006-07 & 2007-08. Evaluation and Applied Research Department.
―. 2010a. Private Schools Fee Determination Committee: District Wise Fee Particulars; Madurai. Accessed
on 1 December 2010 at: http://www.tn.gov.in/departments/sedu/feestructure/madurai.pdf.
―. 2010b. Welfare Schemes of the School Education Department. Accessed on 1 December 2010 at: http://
www.tn.gov.in/schemes/school_education.pdf
Greenwood, M. J. 1971. An Analysis of the Determinants of Internal Labor Mobility in India. The Annals of Regional
Science 5 (1): 137–151.
Heaver, R. 2002. India’s Tamil Nadu Nutrition Program: Lessons and Issues in Management and Capacity Develop-
ment. HNP Discussion Paper. The International Bank for Reconstruction and Development, The World Bank.
Accessed on 10 April 2011 at: http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/
Resources/281627-1095698140167/Heaver-IndiasTamil-whole.pdf.
Jegadeesan M.; and Fujita, K. 2011. Deterioration of Informal Tank Institution in Tamil Nadu: Caste-based Rural
Society and Rapid Economic Development in India. Southeast Asian Studies 49(1): 93–123.
Landy, F. 1992. To Migrate or to Stay in the Maiden: A Survey in Two Rural Systems in South Karnataka. Pondy
Papers in Social Sciences 10. French Institute Pondicherry.
Lanjouw, P.; and Murgai, R. 2009. Poverty Decline, Agricultural Wages, and Nonfarm Employment in Rural India:
1983-2004. Agricultural Economics 40: 243–263.
Lanjouw, P.; and Sharief, A. 2004. Rural Non-Farm Employment in India: Access, Income, and Poverty Impact.
Economic and Political Weekly 39(40): 4429–4446.
Minamino, T. 2003. Indo Noson ni okeru Toshika Kogyoka ni tomonau Kyoiku Kodo no Henka [Changes of Attitude
towards Education Caused by Urbanization and Industrialization in a Indian Village]. Hyogo Kyoiku Daigaku
Kiyo (in Japanese) No. 23: 45–55.
Ministry of Law and Justice. 2009. The Constitution of Scheduled Caste Order 1951 (C.O.1). Accessed on 10
November 2009 at: http://lawmin.nic.in/ld/subord/rule3a.htm.
Misra, H. 2009. Rural-Urban Migration in India. The Indian Journal of Social Work 70(1): 5–25.
Mitra, A.; and Murayama, M. 2008. Rural to Urban Migration: A District Level Analysis for India. IDE Discussion
Papers 137. IDE-JETRO.
Mosse, D.; Gupta, S.; and Shah, V. 2005. On the Margins in the City: Advasi Seasonal Labor Migration in Western
India. Economic and Political Weekly 40 (28): 9–15.
Oka, M. 2006. Indo Gujyarato Shu Chubu ni okeru Nogyo Rochin no Teiisei: Noson Infomaru Kinyu tono Kanren
ni Shoten o atete [Low Wages among Agricultural Laborers in Gujarat, India: With Special Reference to the
Relation with Rural Informal Finance] (in Japanese). Ajia Kenkyu 52(1): 44–61.
Oshikawa, F. 1990. Shakai Henka to Ryuho Seido [Social Change and Reservation System]. In Indo no Shakai Keizai
Hatten to Kasuto [Socio-Economic Development and Caste in India], edited by Oshikawa et al., pp. 3–25. (in
Japanese) JETRO-IDE.
―. 1996. Indo no Toshi ni okeru Shohi to Kazoku [Consumption and Family in Urban Area of India]. Minami
Ajia Shokoku no Keizai Kaikaku to Shakai Kaiso [Economic Reform and Social Strata of South Asian Countries],
edited by H.Sato et al., pp. 93–116. (in Japanese) JETRO–IDE.
Pal, S.; and Kynch, J. 2000. Determinants of Occupational Change and Mobility in Rural India. Applied Economics
32: 1559–1573.
Paul, R. R. 1989. Rural-Urban Migration in Punjab: An Economic Analysis. Himalaya Publishing House.
Ramaiah, A. 2004. Untouchability and Inter-Caste Relations in Rural India: The Case of Southern Tamil Villages.
Journal of Religious Culture No. 70. Institut fur Wissenschaftliche Irenik Johann Wolfgang Goethe-Universitat
Frankfurt am Main. Accessed on 1 December 2010 at: web.uni-frankfurt.de/irenik/relkultur70.pdf
Sato, K. 2008. Local Informal Autonomous Bodies and Gram Panchayat at Work: A Study from a Tank- based Villages
of Semiarid South India. Field Research Programs in 2008, Global COE Program “In Search of Sustainable
Humanosphere in Asia and Africa,”” CSEAS, Kyoto University. Accessed on 1 December 2010 at: http://www.
50
humanosphere.cseas.kyoto-u.ac.jp/mediagallery/media.php?s=20090409155541530.
―. 2011. Goat-Rearing Practices and the Limited Effects of the SHG Program in India: Evidence from a
Tamil Nadu Village. Southeast Asian Studies 49(1): 52–73.
Setty, E. D. 1990. The Valayar of South India, Vol. III: Sex and Divorce, pp. 19-46. Inter-India Publications,
Singh, J. P. 1986. Patterns of Rural-Urban Migration in India. Inter-India Publications.
Singh, S. N.; and Yadava, K. N. S. 1981. On Some Characteristics of Rural Out-Migration in Eastern Uttar Pradesh.
Society and Culture 12(1): 33–46.
Srinivas, M. S. 1989. The Cohesive Role of Sanskritization and Other Essays. Oxford University Press.
TNAU homepage. 2010. Admission to Undergraduate Programs Information Brochure 2010–2011. Accessed on 1
December 2010 at: http://www.tnau.ac.in/ugadmi/ugbroch10.pdf.
UNDP. 2009. Human Development Report 2009: Overcoming Barriers: Human Mobility and Development. Accessed
on 15 March 2011 at: http://hdr.undp.org/en/reports/global/hdr2009/
Usami, Y. 2002. Indo Noson ni okeru Shugyo Kozo no Tokucho to Henka [Characteristics of Employment Structure
and Its Change in Rural India]. In Gendai Minami Ajia 2: Keizai Jiyuka no Yukue [Contemporary South Asia 2:
After Economic Liberalization], edited by H. Esho et al. (in Japanese). University of Tokyo Press.
Walker, T. S.; and Ryan, J. G. 1990. Village and Household Economies in India’s Semi-arid Tropics. The Johns Hopkins
University Press.
Yagi, K. 1999. Kekkon, Kazoku, Jyosei: Kita Indo Noson Shakai no Henyo [Marriage, Family, and Women: Changes
in Rural Society in North India]. In Shakai Henyo to Jyosei: Jenda no Bunka Jinruigaku [Social Change and
Women: Anthropology on Gender], edited by S. Kubota and Y. Yagi, pp. 36–65. Nakanishiya Shuppan.
Yanagisawa, H. 1996. A Century of Change: Caste and Irrigated Lands in Tamil Nadu 1860s–1970s. Manohar.
51

Rural Urban Migration

Uploaded by

Copyright:

Available Formats

Rural Urban Migration

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rural Urban Migration

Uploaded by

Copyright:

Available Formats

Southeast Asian Studies, Vol. 49, No.

Employment Structure and Rural-Urban Migration

production to high value-added production such as livestock and horticulture.

I Outline of the Study Village

Fig. 1 Map of the Study Village

Table 1 Demographic Features of S Gram Panchayat

1981 1991 2001

Agricultural Laborers among Male 43 51 40

No. of Size of Farmland (acre)

Fig. 3 Trend of Sown Area of Major Crops in Madurai District

II Economic Class Structure in the Study Village

Table 3 Relation between Economic Class and Farmland Ownership

No. of Size of Farmland (acre)

Table 4 Relation between Economic Class and Poseession of Durable Goods

No. of Car/Motor Refrigerator DVD Radio Electric Phones Pucca

Table 5 Types of Major Jobs by Economic Class

Total 134 35 2.16 156 25 38 4 7 1 38 21

Source: Fieldwork by the author in 2007–09.

Table 6 Monthly Income Level of Various Occupations

Category Class Vocation and Its Salary1)

Fig. 4 Population Distributions by Age Group in Si Village

houses and paved roads18) which extended throughout the village.19)

III Job Structure and Migration Process of Households by Economic Class

III-1. Current Status

Fig. 5 Map of the Study Village

the sons become “independent migrant workers.”

Table 7 Migrated Relatives of the Sutdy Village

No. of HH Types of Job of Migrated Relative Household

21) He owned a tractor and conducted a tractor rental business.

III-2. Some Historical Background

Occupational Trend in Agriculture and Non-agriculture by Caste in 1910s and 2010s

Priest FC Brahmin Priest 2 × × ×

BC Reddiyar/Naidu 400 119 × × × × ×

Cultivator BC Maravar/Kallar Cultivator 300 × × × × ×

MBC Vallayar (Moopar) 500 × × × ×

MBC Wannar Washer 10 5 × × ×

Service MBC Ampattar Haircutter 10 1 × × ×

SC Pallar Agri laborer 160 × × × ×

IV Education, Its Cost and Source of Financing

IV-1. Occupation and Educational Requirement

Table 9 Average Years of Education for Non-household Heads in Si Village

Workers Staying in the Village Temporary Migrant Workers

Upper 8.9 8.0 10.0 11.9

Table 10 Percentage of SSLC Holders in Si Village by Age Group and Sex

early 20s late 20s 30s 40s 50s 60s 70s

Male 96% 76% 67% 15% 41% 18% 14%

IV-2. Cost of Education and Source of Financing

41) See Education Cost India [2010].

Table 11 Educational Attainment of Young People Aged from 15 to 30 in Si Village

Source: Fieldwork by the author in 2007–09.

Table 12 Major Source of Financing the Cost for Education

No. of HH Source of Fund

Source: Field work by the author in 2009.

Summary and Conclusion

April 2011 at: http://www.censusindia.gov.in/2011-prov-results/paper2/data_files/tamilnadu/Tamil%20nadu_

You might also like