Caste in 21st Century India:

Competing Narratives
Sonalde Desai

India Human Development Survey

Working Paper No. 12

Caste in 21st Century India:

Competing Narratives
Sonalde Desai
University of Maryland College Park
sdesai@socy.umd.edu

Version:
December 2008

India Human Development Survey

Working Paper No. 12

Views presented in this paper are authors personal views and do not reflect institutional
opinions.
The results reported in this paper are based primarily on India Human Development Survey,
2005. This survey was jointly organized by researchers at University of Maryland and the
National Council of Applied Economic Research. The data collection was funded by grants
R01HD041455 and R01HD046166 from the National Institutes of Health to University of
Maryland. Part of the sample represents a resurvey of households initially surveyed by NCAER
in 1993-94.

Caste in 21st Century India:

Competing Narratives

Abstract:
Much of what is known about caste differences in India is based on studies conducted in the early
and middle part of the 20th century. Relatively little research focuses on caste differences in modern India,
a gap addressed by this paper. Using data from a nationally representative survey of 41,554 households
conducted in 2005, this paper examines the relationship between caste and status hierarchies as well as
caste and material inequalities. Based on these results, this paper argues that a vision of caste hierarchies
based on a traditional model of social stratification in which symbolic dimensions of social status
dominate has little support in 21st century India. However, caste-based inequalities in material conditions
are widespread. This suggests that instead of seeing Indian caste systems as a special form of social
stratification, research on caste-based inequalities can be used to illustrate the nature of social
reproduction during an era of social change.

Caste in 21st Century India:

Competing Narratives
Introduction:
Although research on social inequality remains at the heart of the sociological enterprise,
texts on social stratification tend to ignore traditional forms of social inequality as being
outside the purview of the modern stratification system. Introductory stratification textbooks
often begin with a reference to the Indian caste structure as an archetypical example of a
stratification system based on religious precepts and then move on to discuss modern forms of
stratification based on economic inequality. A primer on class and stratification provides an
interesting example (Crompton 1998); on the first page of a 255-page volume it states,
[sometimes] inequalities have been viewed as deriving from the divinely ordained structuring
of society, as in the Hindu caste system in classical India. In this system social rank
corresponded to religious (ritual) purity. Lower castes polluted the higher, and as a consequence,
a series of restrictions were placed on low-caste individuals (p.1-2). It goes on to focus on class
as an attribute of modern societies and pays little attention to caste in the subsequent 254 pages.
This textbook is not unique; a variety of stratification textbooks and studies of status hierarchies
tend to classify caste as a form of stratification rooted in religion and universally accepted by the
dominant as well as subordinate groups in the hierarchy (Grusky, Ku, and Szelenyi 2008; Gould
2002).
As a result of sociologys tendency to relegate caste to the realm of the traditional, most
of the debates on caste have been carried out in the specialized arena of South Asian studies.
This disjunction has had a negative impact on both literature streams. The stratification literature

continues to uncritically accept the vision of a hierarchical caste system built on the principles of
karma (action) and dharma (religion), unaware of strongly dissenting voices in South Asian
studies literature. The South Asian studies literature, in turn, frequently continues to debate a
straw man while ignoring material foundations to caste-based inequalities in modern India.
Absence of data on caste has further limited this discourse. Much of what we know about caste
inequalities comes either from the census of 1931, the last time caste data was collected and
tabulated, or from village ethnographies with limited regional focus.
This paper seeks to integrate these two separate streams of literature and provide
empirical contours to it by (1) reviewing the recent debates on caste in the field of South Asian
studies; (2) reframing them in the form of testable hypotheses about inequalities in status and
material conditions, which are then examined using survey data from a nationally representative
survey of 41,554 households; and (3) linking these results to broader literature on social
inequalities in changing societies.

Caste and Status Hierarchies:

One of the reasons caste has excited sociological imagination is because it is seen as a
representation of pure status, based on religious and ideological grounds (Milner 1994; Dumont 1980;
Weber 1958) with class inequalities being epiphenomenal to caste. This disjunction between the sacred
and the profane gives the Indian caste system a sociological character that sets it apart from other forms
of social inequality based on material resources. Weber lays out the essential characterization of the caste
system as opposed to affinity groups that has undergirded much of the sociological discourse on caste:
A status segregation grown into caste differs in its structure from a mere ethnic segregation: the
caste structure transforms the horizontal and unconnected coexistences of ethnically segregated
groups into a vertical social system of superordination and subordination . ethnic coexistences

condition a mutual repulsion and disdain but allow each ethnic community to consider its own
honor as the highest one; the caste structure brings out a social subordination and an
acknowledgement of more honor in favor of privileged castes and status groups. (Gerth and
Mills 1946, p. 189)
While Weber largely relied on writings by colonial bureaucrats in the Indian Civil Services
(acting as amateur anthropologists) for data on Indian society, anthropological villages studies of the 20th
century by Indian as well as Western scholars provided a foundation for Louis Dumonts work (Dumont
1980). With the publication of Homo Hierarchicus in 1966, Dumont presented a canonical formulation
that has framed the conversation about caste over the past four decades and provided a rationale for status
hierarchy. For Dumont, the origin of the Indian caste system is located in vedic ideology articulated in
ancient scriptures like Dharmasastras (circa 200 B.C.), which enumerate the rights and obligations of
four varnas ranging from the highest to the lowest status Brahmin (Priest), Kshatriya (Warrior),
Vaishya (Merchant) and Shudra (Servants). Brahmins, Kshatriya and Vaishya are considered higher in
status than Shudras because adolescent males from these varnas undergo a thread ceremony making them
twice born, a privilege not available to the Shudras. In the post-vedic period untouchables (now referred
to as dalits) and tribals (now referred to as adivasis) were added to the bottom of the pyramid. Brahmins
are conceptualized as being at the top of the stratification system, superseding the Kshatriyas in spite of
the latters worldly power and wealth as the ruling caste. Dumont follows the vedic texts in assigning
supremacy to Brahmins based on the location on a purity/pollution scale where they gain higher status
due to their abstention from polluting food (meat), polluting activities (farming, leatherworking and
scavenging) and polluting sexual behavior (remarriage of a widow).
Although a focus on varna is useful in a pan-India analysis or in examining the vedic texts, the
canonical varna system is less salient to daily lives of most Indians than jati. Thousands of jatis castes
have been enumerated all over India, which have some rough correspondence with the varna system, but
are far more concrete in day-to-day lives of individuals defining an endogamous group that forms the core
4

of their salient social network. However, in Dumonts formulation, jatis are purported to retain the
hierarchical nature of the varna systems and draw their ranking within the hierarchy from the same
ideological principles, particularly the disjunction between purity and pollution. This implies that in a
typical north Indian village, a Rajput farmer is inferior to a Joshi priest but superior to the Chamar
leatherworker, since a farmer is polluted by dealing with earthly substances but less polluted than the
leatherworker who deals with the skin of dead animals.
The preceding paragraphs briefly summarize Dumonts formulation of caste hierarchy. Arguably
it is the best-known description of caste as a system of stratification, but in emphasizing the ideological
over the material his formulation has much in common with his predecessors (Weber 1958) and
successors (Milner 1994). This narrative of caste has excited tremendous passions from diverse groups
with wide-ranging critiques (Appadurai 1986; Gupta 2000; Kolenda 1973; Marriott 1969; Srinivas 1996;
Berreman 1991).
Status theories of caste hierarchies have a tendency to focus on ageless and timeless India as
represented in Sanskritic traditions, partly because they draw upon the religious foundations of caste,
which are remarkably stable over time. This focus often ignores modern India, particularly urban India, in
which concepts like purity and pollution are difficult to implement in day-to-day life. As noted Indian
social anthropologist Beteille (Dumont and Beteille 1987) remarks in an acerbic exchange with Dumont,
Dumonts lack of ease with modern India is writ large in his work, although it does not shine as brightly
as his enthusiasm for traditional India, which is partly an India of his own construction. . Modern India,
in Dumonts construction, is not made of whole cloth, it is a thing of shreds and patches. (Beteille, p.
675 in Dumont and Beteill 1987). While there seems to be a general agreement regarding social
differentiation between castes based on visibly recognizable symbols, including rituals, dress, tonsorial
styles and a host of other behavioral markers, whether this differentiation translates into social hierarchies
in modern India is far from clear (Gupta 2000) and some intriguing studies have documented declining
salience of caste over time even in rituals and food habits (Mayer 1997).
5

The second set of critiques originates in debates arguing whether Brahmins are at the pinnacle of
the caste system or whether they were placed there by colonial imagination either because of a
fascination with the exotic or as a way of creating a social order that could be exploited in colonial
governance (Dirks 2001). This critique suggests that the Brahmin superiority is a creation of an orientalist
discourse that privileged Brahmins over Kshatriyas (Dirks 2001; Raheja 1988) and ignored the claims to
ideological superiority from lower castes (Khare 1991; Chatterjee 1992; Das 1982). By focusing on
alternate sources of legitimacy and hierarchy, these studies question the notion that the Brahminical
values are internalized by all subordinate castes and that caste hierarchies are accepted without
interrogation.
The foregoing discussion suggests a need to empirically examine the extent to which Brahminical
ideology holds sway in modern India in a way that castes align along a spectrum of behaviors in
descending order of rank. Identification of these behaviors and their empirical examination is one of the
central goals of this paper, and I return to it in a later section.

Caste and Economic Hierarchies:

While there is no intrinsic reason why higher castes may not have greater material power, it is the
presumed disjunction between status and material resources that has given rise to such interest in the
Indian caste system. Assumptions about the disjunction between caste and economic status rest on two
distinct arguments: (1) Lower economic status of Brahmin priests compared to Kshatriya kings; and (2)
Declining salience of caste/occupational nexus in modern India, particularly in the context of affirmative
action favoring the lower castes.
One explanation of the presumed higher status of Brahmin priests against Kshatriya kings has
been to postulate (Dumont 1980) that status which rests on performance of religious duties (dharma) is
more valuable than wealth and power (artha), and some commentators go as far as claiming that
preoccupation with status and lifestyle limits and handicaps the acquisition of wealth and political

power (Milner 1994, p. 147).

This is a particularly tricky issue for research on caste. Most village

studies have documented greater power and wealth among upper castes than lower castes which would
suggest that status and wealth in India go hand in hand. It is only when the Brahmins are compared to
other twice-born castes that one expects to see a disjunction between wealth and caste status.
However, whether Brahmins in modern India are poorer than the middle castes is far from clear.
Historically, Brahmins served as interpreters of India to the West (Das 2003) and in the process managed
to consolidate their power on Western education and government jobs. An interesting analysis of the
background of elite bureaucrats in the Indian Administrative Services (IAS), the successor to colonial
Indian Civil Services, shows that as late as 1985, about 37% of the IAS officers self-identified as being
Brahmin (Goyal 1989), a disproportionately large number since Brahmins form only about 5% of the
population. Moreover, since a substantial proportion of the applicants do not provide caste information,
this number is undoubtedly an underestimate. Thus, while Brahmins of pre-colonial India may or may not
have surpassed the Kshatriya kings in material wealth, there is little reason to believe that the modern
Indian Brahmin is a poor village priest, and the notion that caste hierarchies reflect differences in status
rather than wealth deserves greater empirical scrutiny.
Economic growth and the changing nature of occupational structure as well as strong Indian
affirmative policies have provided a second justification for the assumption that caste is increasingly
unconnected to material resources. In addition to a religious dimension to caste, caste also contained an
occupational dimension. The varna schema even named the varnas based on the activities they performed,
and research in early 20th century documents that economic exchanges between different castes were
governed by strict rules of exchange defined in jajmani system (Wiser 1979). However, it has been
argued that the relationship between caste and occupation has been much misrepresented. It is
doubtful that there was at any time a complete correspondence between the two. At any rate, even before
independence many castes, and probably most, had more than half their working members in occupations
other than those specifically associated with their caste (Beteille 1992, p.40).
7

Regardless of the link between caste and occupation in colonial India, in an independent India
this link has weakened considerably. The jajmani system has all but vanished, allowing for market-based
pricing for services rendered by the workers (Commander 1983), and a variety of forces are believed to
have disrupted the link between caste and occupation. Land reforms transferred land ownership to many
former sharecroppers, most of whom belong to the middle castes (Dantwala 1950); declining incomes of
artisans and influx of mass-produced goods have led to declining caste-based occupations among potters,
weavers and other artisans who must now rely on manual labor for subsistence (Bayly 1999); and
increased requirements for education among modern professions has led to influx of people from a variety
of castes into modern occupations (Sharma 1999). All of these trends would suggest that the link between
caste and economic status in modern India is marginal at best. In an analysis of the numerically
preponderant dominant castes in South India, noted anthropologist M.A. Srinivas found that certain
peasant castes enjoy numerical superiority as well as political and economic power, although they remain
middle castes by the varna schema (Srinivas 1987). Politics of affirmative action has further
strengthened the power of lower and middle castes by allocating quotas sometimes as much as 50% in
college admissions, government jobs and local governance (Beteille 1992). Since government jobs
provide far greater incomes than private-sector jobs, increasingly the upper castes have begun to feel
overshadowed by the opportunities available to the lower castes (Ghuyen 2004), particularly in areas
dominated by anti-Brahmin movements.
Empirical research in this area began with a focus on the lowest castes and tribes (Mendelsohn
and Vicziany 1998; Deshpande 2000) with recent collection of survey data allowing for exploration of
inequality between the middle castes and others (Deshpande 2007; Mohanty 2006). Curiously, these
empirical studies continue to document continued caste inequalities in material resources without making
a dent in the discourse of declining salience of caste in modern India. Part of this incongruence may be
that empirical research is limited by the available data and consequently relegated to combining all
upper castes into one category, an issue discussed below.

Debates without Data:

Our brief discussion of the debates surrounding caste hierarchies in India and the competing
theories about the nature and magnitude of upper caste control over ideological and material resources
clearly suggests a need for empirical research. However, many of these debates have been conducted in a
sterile environment devoid of empirical substantiation using recent data. India continues to rely on the
1931 census i for detailed caste information (Sundar 1999); the present censuses only collect information
about whether the respondent is dalit or adivasi (Scheduled Caste or Scheduled Tribe). Large nationwide
household surveys continue to follow this practice, although recent surveys have added a category, Other
Backward Classes (castes) or OBC. ii Each of these categories is based on a government- defined list (or
schedules) of castes eligible for affirmative action benefits. Since many debates are associated with the
nature of Brahminical behavior and privilege, leaving the upper castes undifferentiated is not
particularly helpful. While ethnographic studies continue to collect detailed caste data, frequently the
range of castes living in a single village tends to be limited and does not allow comparisons across a
diverse spectrum of social groups. Moreover, many ethnographic studies tend to focus on villages, for
practical reasons, with resultant theoretical bias in favor of rural rather than urban experiences. In theory,
data from household surveys can fill this vacuum; however, most household surveys in India are
conducted by economists or demographers and tend to be restrictive in scope both in collecting
information on caste identification and in range of social behaviors being examined.
We are fortunate to have access to a newly collected nationally representative household survey
of 41,554 households. India Human Development Survey 2005 (IHDS) surveyed urban and rural
households in 1503 villages and 971 urban blocks spread across all Indian states and territories, with the
exception of the islands of Andaman Nicobar and Lakshadweep. Detailed social and economic
information was collected in face-to-face interviews from one knowledgeable respondent usually the
head of the household about household social and economic status, while 33,482 ever married women
aged 15-49 were interviewed about gender roles, health and fertility. The English and Hindi

questionnaires were translated in additional eleven Indian languages and were administered by male and
female interviewers fluent in local language and customs, with female respondents being typically
interviewed by female interviewers. While this was a multitopic, multipurpose survey, it contains one
unique piece of information that makes it possible to test some of the ideas discussed above: unlike other
household surveys, it distinguishes between Brahmins and other forward castes.

Categorizing Caste:
Collecting data on caste in a nationwide survey or census is one of the most complex tasks faced
by any analyst. As Veena Das noted in the context of colonial data collection projects, The process of
recording objectified a particular definition and to freeze the ongoing processes by which caste came to be
solidified in the official imagination It also generated the conception of community as an enumerative
community which had a strong influence on processes of political representation (Das 2003, p.10).
While caste categories have concrete and specific meanings in local contexts, these meanings do not
transcend geographical boundaries. Consequently most surveys have stayed with categorizing caste into
the lowest castes (dalits or Scheduled Castes); tribes (adivasi or Scheduled Tribe); and, more recently,
Other Backward Classes (castes).
This has been a practical step that was taken for very good reasons. In independent India,
affirmative action policies are geared towards dalits, adivasis and Other Backward Classes (OBC).
Consequently, there are clearly demarcated lists of who belongs in these categories. Castes classified as
Scheduled Caste, or dalit, are so designated because they suffered from social exclusion and were at the
bottom of the caste pyramid. The OBC category contains castes that are just above the dalits and largely
belong to the upper sections of Shudra category, with a few lower-ranking Kshatriya jatis being included.
It is important to note that this is an imprecise definition and does not fully correspond with the varna
definition, but dalits and OBCs together reflect the lowest-ranking castes in most regions. Since
households need caste certificates to obtain affirmative action benefits, most are aware of their
membership of these defined caste categories.
10

Differentiating between the middle and the upper castes is far more difficult if one wants to
obtain general categories that are comparable across India. However, given the location of Brahmins in
the Hindu religion, most jatis and households seem fairly clear as to whether they belong to the Brahmin
category or not. Consequently, a category was added to the IHDS survey, allowing a household to
categorize itself as Brahmin. This distinction between Brahmins and middle castes allows us the analytic
power to examine the dispersion of markers of status and wealth along the caste hierarchy at both the
upper and lower ends of the spectrum.
[Figure 1 about here]
In addition to caste-based division among Hindus, the Indian population also contains about 13%
Muslims and about 6% members of other minority religions, including Christians, Jains, Sikhs and
Buddhists. Figure 1 indicates the distribution of households in these various categories in the India
Human Development Survey (IHDS). This analysis focuses on the following categories:
Hindu:

Brahmin

High caste This group includes non-Brahmin Hindus who are not categorized as OBC,
Dalit or Adivasi. Typically castes that fall into broad Vaishya (merchant) or Kshatriya
(warrior) varnas.

Other Backward Classes (castes) Middle castes typically belonging to lower groups of
Kshatriya or upper group of Shudras who are formally defined as being backward and
deserving of affirmative action.

Dalit Lowest castes containing lower category of Shudras and castes formerly termed
untouchables, who engage in such polluting activities as cleaning toilets and
scavenging and hauling carcasses, that they are considered among the lowest in Hindu

11

hierarchy. Since these castes are listed on a schedule for affirmative action, they also are
called Scheduled Castes.

Adivasi Literally translates to the original inhabitants of the land. These tribal
populations remain outside of mainstream society and follow a variety of religions. Here
I focus only on Scheduled Tribe households who are Hindu. Adivasis who are non-Hindu
form about 1.7% of the total sample and are combined with other religions.

Muslim:

While some Muslims claim to have a caste based on their hereditary occupation, given
our focus on Hindu hierarchy I combine all Muslims. Less than 5% of the Muslim
households claim to have a caste and all of these are OBCs.

Christians/Jains/Sikhs/Buddhists and other religious groups:

The remaining religious groups are combined in this category. While minority religious
families seem to have a caste identification, for the purpose of this analysis I focus only
on Hindus and combine all other religious groups into one category for comparison,
while noting that converts to Christianity and Buddhism, in particular, often came from
lower segments of Hindu caste structure who sought to improve their status via
conversion.

12

Research Questions and Methods:

Three central questions motivate this paper:
1. Does caste structure in modern India continue to reflect status hierarchies?
2. Does caste structure in modern India reflect material hierarchies?
3. If either of these is evident, what can we conjecture about the processes that generate these
relationships?
Vast quantities of social stratification literature tell us that education and incomes are the crucial
material resources dividing different strata of the modern society. Similarly, a large number of studies in
India and elsewhere have noted that land ownership is another crucial marker of economic resource in
agrarian societies (Basu 1994). However, what are the indicators of social status or prestige that are
relevant if we want to study caste inequalities in social status? Here we must draw upon the literature that
has focused on different dimensions along which castes are hierarchically ranked; to do otherwise would
risk tilting against a straw man of our own creation.
A focus on caste hierarchy with Brahmin at the apex often rests on the assumption that Brahmins
follow a way of life that confers high status on them. Other castes may emulate this lifestyle to try to raise
their status but may face a variety of obstacles, and how high they rise in the caste hierarchy depends on
their ability to Sanskritize (Srinivas 1956), or move higher up along the purity spectrum (Dumont 1980).
Thus, we expect that caste position in hierarchy will depend on how high castes are along this
Brahminical way of life or, conversely, Brahmins, other high castes, middle castes and low castes are so
ranked because of their differential ability to achieve this way of life.
What is the Brahminical way of life? Here a host of variables are postulated to characterize the
Brahminical lifestyle, but vegetarianism is perhaps most frequently mentioned, followed by marriage
patterns (Dumont 1980; Kolenda 1997; Srinivas 1956; Milner 1994). Eating meat is seen as being
polluting, and castes that eat meat have a lower social status than those that do not.

13

Marriage patterns and associated patterns of gender relations play an important role in
determining the status of caste in several ways. Historically control of female sexuality was one avenue
through which upper castes retained their purity (Dube 2001). This was done by arranging marriage
before puberty, forbidding remarriage of widows, clear demarcations of male and female roles, and
immurement of women (Srinivas 1977). Research on gender has also highlighted the association between
performance of gender roles and higher caste status, and the way in which hierarchies of caste are
articulated through gender (Dube 1996; Liddle and Joshi 1989).
In addition to greater gender segregation, marriage arrangements also serve as a tool of caste
mobility. While in Hindu scriptures marriage within the same caste is preferred, marriage of an uppercaste male to a lower-caste woman (anuloma) is acceptable but that of a lower-caste male to an uppercaste woman (pratiloma) is forbidden. Marriage of daughters into upper-caste households is one of the
ways in which families and castes may gain higher status. This desire for hypergamy creates considerable
demand for a bridegroom from upper caste and results in their being treated as near deities (Milner
1994).
These observations suggest that if the symbolic dimension of caste hierarchy remains relevant in
modern India, we should see it reflected in vegetarianism, marriage arrangements and gender segregation,
with middle castes striving to get closer to the Brahmins and a greater difference between Brahmins and
the lower castes. In contrast, if the ideological dimension is no longer relevant to the caste hierarchy and
simply reflects a Brahminical view of caste (Berreman 1991), this hierarchy will not withstand an
empirical test.
These three markers vegetarianism, gender segregation and high status of upper caste
bridegrooms are operationalized using the data from India Human Development Survey 2005 in the
following way:

14

Meat Consumption: The IHDS asked about household consumption of various items including
meat and fish, quantities consumed, whether these items were home produced or purchased, and
amount spent. These items are recoded to create a variable that takes the value 0 if no meat or
fish was consumed by the household in the month preceding the survey and 1 if it was
consumed regardless of whether meat was purchased or produced from family-owned livestock
or poultry. Information for this variable is available for all households.

Gender Segregation: Gender segregation is measured using responses of ever married women
aged 15-49 to a question about whether they practiced purdah or ghunghat. Purdah, a term
typically used by Muslims, and ghunghat, a term typically used by Hindus, refers to the practice
of fully or partially veiling ones face in the company of men from within or outside the family. A
substantial number of studies in South Asia have identified veiling as a key dimension of
womens seclusion (Mandelbaum 1988; Papanek 1973) and other research with data from IHDS
notes that practice of ghunghat or purdah is highly correlated with other markers of gender
segregation (Identifying Reference). Information for this variable is available for 33,482
households with eligible women.

Demand for Bridegroom: Demand for bridegrooms is measured by the expenditure a grooms
family is typically expected to undertake in the course of a wedding in the respondents caste.
This includes expenditure for the wedding ceremony, reception, jewelry for the bride and other
gifts. I expect that if upper-caste bridegrooms are treated as near deities, their families must
undertake very little expenditure, with most of the expenditure undertaken by the brides family.
The IHDS asked respondents about the typical wedding expenditure in their jati, differentiating
between expenditure from the brides and grooms side. Almost universally, the brides family is
expected to spend more money than the grooms family. But the expenditure for grooms is not
negligible. The present paper analyzes this expenditure to see if, ceteris paribus, upper-caste
grooms spend less than lower-caste grooms. Information for this variable is available for all
households. The exchange rate in 2005 was 44.5 Indian rupees per one U.S. dollar.
15

Our measures of material conditions borrow from past research on stratification and include the
following:

Per Capita Consumption Expenditure: In developing country research, given the difficulties in
measuring agricultural incomes, researchers frequently rely on consumption expenditure, which is
seen as a marker of permanent income, free of seasonal fluctuations (Grosh and Glewwe 2000).
The IHDS used a brief questionnaire measuring consumption of items in various categories, using
a technique used by the National Sample Surveys Employment Supplement. The total household
expenditure is divided by the number of residents, to give a per capita consumption expenditure,
and is available for all survey households.

Amount of Land Owned: Amount of land owned in local units was collected from all families
and is converted into acres for the present analysis. Households without land are coded 0.

Educational Attainment: The IHDS collected information in educational attainment for all
members of the surveyed households. This analysis focuses on educational attainment of males
aged 25-49, reflecting the composition of the present labor force. The analytic sample for this
portion of the analysis includes 35,259 men

iii

Salaried Employment/Business Ownership: Having a regular job with a monthly salary instead
of casual wage work is an important avenue for higher incomes in India (as in many developing
countries). Since daily wage work is seasonal and minimum wage is rarely enforced, casual wage
workers are among the poorest in India. Ownership of a business, such as a small grocery store, a
cobblers shop or a bicycle repair shop, is another avenue through which individuals can gain a
somewhat secure livelihood. This portion of the analysis focuses on employment in a salaried job
or in a family-owned business for men aged 25-49. The analytic sample includes 35,259 males.
Descriptive statistics for these variables are provided in Appendix Table 1. Each of these

dependent variables is examined in multivariate analyses, controlling for urban/rural residence, state of
residence, and highest educational level attained by household adults. In equations for all markers of
16

status, logged consumption expenditure per capita is also included in order to control for the economic
status of the household. The two continuous dependent variables, consumption expenditure per capita and
wedding expenses, are logged and analyzed using linear regression as is educational attainment; the rest
are estimated using logistic regression. While results from full regression are presented in Appendix
Tables 2-4, for presentational clarity, the text contains predicted values from each of these regressions,
setting the value for the control variables to mean.

Does Caste Confer Status?

A variety of scholars has noted differentiation between castes in external behaviors such as food
preferences, clothing and religious rituals (for a detailed discussion see (Gupta 2000)); these differences
are not in contention. It is the notion of system of hierarchy and complicity within this system by the
lower orders that is subject to debate. For a true hierarchy to exist, empirical data must demonstrate that
castes are arrayed on different dimensions of status in an order that is consistent with caste hierarchies.
Detailed regressions estimating caste differentials in different markers of status are presented in Appendix
Table 2. These control for urban/rural residence, state of residence, highest education attained by
household adults and the households per capita expenditure. However, in order to interpret substantive
importance of these coefficients, Table 1 shows predicted values of these outcomes of interest for
different caste/religious groups while holding other factors such as urban residence, state of residence,
education level and household economic status at their mean values. The predicted values in Table 1 tend
to support the notion that at least for the markers I examined, caste differentiation does not appear to be
hierarchical in the way I hypothesized.
[Table 1 about here]
The first column of Table 1 shows the caste differences in meat consumption, a classic point of
differentiation along the purity/pollution spectrum. Proponents of the hierarchical view of caste based on
purity and pollution argue that vegetarianism is associated with higher caste status. The results indicate

17

that the Brahmins (reference category) are the least likely to consume any meat. However, the following
two caste groupings, forward and backward castes, are more or less indistinguishable in their propensity
to eat meat. If renouncing meat consumption truly were a significant marker of caste status, we would
expect to see greater differentiation between the forward and backward castes. The forward caste group
contains mostly Kshatriyas and Vaishyas from the varna scheme, while the other backward castes (OBC)
contain Shudras from the varna scheme. If there is little difference in meat avoidance between these
vastly unequal groups, it casts doubts on either the existence of status hierarchy or vegetarianism as a
marker of status. However, the other two groups, dalits and adivasis, are the most likely to consume meat.
The second variable, not adhering to purdah, is even more ambiguous. Here Brahmin women are
most likely to practice purdah and other forward caste women are the least likely to practice purdah, but
the lower groups consisting of OBC and dalit women are indistinguishable from Brahmin women. This
suggests that while castes have distinctive characteristics with respect to womens practice of purdah, it is
difficult to place them in a hierarchical scale on this marker of status.
The third variable, typical wedding expenditure by the grooms family, seems to be positively
rather than negatively associated with the presumed caste status. Brahmin grooms engage in the higher
wedding expenditure, with other castes progressively spending less. This relationship is strong with or
without inclusion of controls for education and economic standing of the family. A theory of status
hierarchy would suggest that grooms in great demand, those at the upper end of caste hierarchy who are
treated as near deities (Milner 1994), should be able to transfer all burdens of wedding celebrations
onto the brides family. We see little evidence of this in the empirical data presented in this paper. iv
To recapitulate the results presented in Table 2, I find little relationship between these
hypothesized markers of status and caste hierarchy. Only for one variable, meat consumption, does the
data indicate an increasing consumption down the caste ladder. For the second variable, practice of
female seclusion, the relationship is curvilinear, and for the third variable, wedding expenditure for

18

grooms, the relationship is reversed. These empirical results suggest considerable justification for the
following observation about Dumonts classical formulation of caste hierarchies (Berreman 1991):
Dumont relies heavily on some Sanskrit textsthe result is that he conveys a view of caste
which is artificial, stiff, stereotypical and idealized. It is a view which confirms rather closely the
high-caste ideal of what the caste system of Hindu India ought to be like according to those who
value it positively: it conforms well to the theory of caste purveyed in learned Brahminical texts.
(P.91)
It may well be that the markers of status hierarchy used above are less relevant than some other
markers; however, given the paucity of recent empirical work showing status hierarchy on any other
markers, and an overwhelming chorus from South Asia specialists arguing against simplistic assumptions
about caste hierarchies (Fuller 1997; Gupta 1991; Srinivas 1996), it seems reasonable to suggest that it is
time to reconsider what caste means in modern India.

Is Caste Still Relevant in Modern India?

If modern scholarship on India is united in its chorus against a religion-based hierarchy in Indian
caste structure, it fractures into a cacophony of sounds when it comes to defining the nature of caste in
modern India. While few scholars claim a demise of caste, there is an increasing tendency to view caste as
a form of ethnicity in which castes compete with each other for power and proudly brandish their own
narratives of origin, with even the lowest castes claiming a place in the national history characterized by
valor and accomplishment (Narayan 2004). Dissenting from an orientalist viewpoint, which saw the
subjugated as being acquiescent in their own subjugation, recent research has tended to focus on the
agency and power of the oppressed (Khare 1991). This dissent has been bolstered by political
developments in modern India. As the era of grand ideology came to an end in India as elsewhere,
regional political parties have emerged as power brokers (Brass 1990; Frankel 2000). Many of these
parties rely on numerically large middle and lower castes for support; consequently, there is a perception

19

that the varna structure has transformed itself into an ethnic jati structure, where small endogamous jatis
compete with each other for social and political power. The politics of affirmative action sharpened this
competition and led to increased attempts to capture the support of the state. Consequently, it is argued
that the middle and lower castes have staked out a claim to economic and political power that is on par
with, if not greater than, the Brahmins and other upper castes (Gupta 2005).
It would be an oversimplification to say that this narrative of ethnicity is unquestionably accepted
by researchers working on India. A few empirical studies continue to point to caste-based inequalities in
economic circumstances (Deshpande 2007) between OBCs and higher castes. Moreover, starting from the
perspective of lowest castes and tribes the subaltern a number of studies of the lowest castes have
observed that the control of higher castes over material resources seems clear, control over ideological
resources seems debatable and a preoccupation with ideology seems to obfuscate on-the-ground
inequalities (Panini 1997; Mendelsohn and Vicziany 1998; Berman 1974; Omvedt 2006).
In spite of these dissenting voices, the narrative of ethnicity seems to prevail, possibly because
modern Indian intelligentsia continues to define caste through the eyes of a religious discourse that they
find increasingly irrelevant. Moreover, urban India has enthusiastically embraced a global discourse of
equality where anyone who speaks against equality in public is bound to lose his audience (Beteille
1991). As Chris Fuller (1997, p.13) in the introduction to a volume titled Caste Today notes, The
elimination of hierarchical values from legitimate public discourse accounts, for the claim that there is
no caste left. That caste hierarchy can no longer be legitimately defended in public has itself contributed
to the emergence of a more or less acceptable public discourse about status coded as cultural difference.
Because people cannot openly speak of castes as unequal, they describe them as different.
Unfortunately, this complacency about vanishing caste-based social inequalities in a rapidly
globalizing India has received little empirical scrutiny, a gap the present paper seeks to fill. Have
inequalities in material conditions between castes disappeared with the disappearance of traditional modes
of economic exchange and advent of new occupations? Figure 2 shows differences in annual per capita
consumption expenditure for different caste categories. The boxes show distribution at 25th, 50th and 75th
20

percentile respectively while the whiskers indicate the range (with outliers omitted). The results very
clearly document that median expenditure levels for Brahmins is Rs. 11,810 compared to Rs. 10,472 for
other high castes, Rs.7,586 for OBCs, Rs.6,301 for dalits and Rs.4,754 for adivasis. While minority
religions are not a focus of this paper, their expenditures, presented for comparison purposes, show Rs.
7,046 per capita expenditure for Muslims and Rs. 8,988 for Jains, Sikhs, Christians and other religions.
What is even more interesting is to note the distribution of expenditure across different social groups. The
75th percentile in Brahmin/High Caste groups is considerably higher than the 75th percentile for the lower
castes and the bottom 25th percentile in these groups is also higher than that of other castes. This suggests
that at the upper end, the achievement potential for Brahmins and other high castes is considerably higher
than that for lower castes, resulting in higher mean expenditures for Brahmins (Rs. 16015), other high
castes (Rs. 14254) compared to the middle castes (Rs. 10015), lowest castes (Rs. 8102) and tribes (Rs.
6142).
[Table 2 about here]
A substantial proportion of this difference is due to higher concentration of Brahmins in cities and
that of adivasis (Scheduled Tribes) in remote areas of the country; they face significant differences in
economic opportunities. However, even controlling for urban residence and state of residence, Table 2
indicates substantial disparities in per capita expenditure between different castes. Regression results
presented in Appendix Table 3 show that these differences are statistically significant.

Pathways to Income Inequality:

How do we account for this continued inequality? I suggest that three mechanisms may play an
important role in shaping caste-based income inequalities: (1) greater land ownership; (2) higher levels of
educational attainment; and (3) greater access to better-paying jobs or self-employment in business.
The greater the amount of land a household owns, the greater the possibility of obtaining higher
incomes through economies of scale, greater access to credit and greater ability to engage in crop rotation.
Research in India and elsewhere has consistently identified the size of the landholding as an important
21

marker of economic well-being. Land ownership in colonial India was limited largely to upper-caste
households, with lower castes working for them as sharecroppers or wage laborers. Following
independence in 1947, government embarked on a series of land reforms that continued well into the
1960s. These involved land ceiling, with land from large landowners appropriated by the government for
redistribution to the landless, transfer of land titles to tenant farmers and distribution of land owned by the
government to the landless (Dantwala 1950; Bardhan 1984; Deshpande 2003). While these reforms have
successfully broken up large landholdings, results presented in Appendix Table 4 point to continued
caste-based disparities in the size of landholding, with Brahmins and other high castes owning more land
than OBCs, dalits and adivasis. The predicted values in Table 3 indicate that holding residence constant,
Brahmins and other upper castes own about 2 acres of land, compared to about 1.69 acres for OBCs and
1.27 acres for dalits.
[Table 3 about here]
Addition of landholding size to the consumption expenditure regression (Table 2) serves to
reduce caste-based inequalities in per capita income somewhat. For example, the difference between
Brahmins and OBCs (the middle castes) declines by Rs. 281 and that between Brahmins and dalits
declines by Rs. 548. However, it is important to note that given the low productivity of Indian agriculture
and the small sizes of most holdings (Basu 2007), unequal access to land may not be a primary
determinant of income inequality and is likely to become even less important as urbanization progresses.
A variety of studies has demonstrated persistence of caste-based educational inequalities in India
(Desai and Kulkarni 2008; Govinda 2002). Appendix Table 4 shows regressions for males aged 25-49,
with their years of completed education as the dependent variable, and these results are presented in the
form of predicted probability in Table 3. These results suggest that holding place of residence constant,
Brahmin males obtain nearly 11 years of education, compared to 8.78 years for other high-caste males,
with all other castes showing lower levels of education. These results clearly indicate that education is
one of the largest areas of Brahmin advantage. The Indian educational system is organized in such a way
that 10 and 12 years of education reflect two important thresholds. High school completion typically
22

means passing of the 10th standard examination, and junior college completion involves passing the 12th
standard examination, with college typically requiring an additional 3 years of education. Thus,
completion of 11 years of education on average makes individuals eligible for a much greater range of
jobs than dropping out before the 10th standard.
Changes in caste coefficients after addition of education to the regression for consumption
expenditure in Table 2 further highlight the importance of education. When we control for the highest
level of education attained by adults in a household, the difference between various caste groups narrows
substantially over panel B, which only controlled for residence and landholding. The difference between
Brahmins and other high castes declines by Rs. 927 and that between Brahmins and OBCs, dalits and
adivasis declines by more than Rs. 1,500. This suggests that a substantial portion of caste differences is
associated with higher education levels of Brahmins and other upper castes, a topic to which I return in
the concluding section.
The final pathway I explore is access to salaried jobs and ownership of a family business. One of
the reasons for higher material status of upper castes could be due to long-term caste association with
certain activities, such as greater entrepreneurship or greater likelihood of being in formal sector
employment. Appendix Table 4 and predicted probabilities in Table 3 explore the relationship between
caste and salaried jobs/owning a family business. The results support the contention that caste is
associated with certain types of employment. Holding education constant, Brahmins are far more likely to
be in salaried jobs or business than other caste groups, with high castes and OBCs falling in the middle
and dalits and adivasis at the bottom.
Contrary to my expectation, however, inclusion of a households access to salaried or business
incomes does little to reduce caste-based inequalities in consumption expenditure. Access to
salaried/business incomes increases household consumption expenditure but has a relatively small impact
on caste differentials, with caste differentials in panel D of Table 2 declining by less than Rs. 100. This is
an intriguing finding, suggesting that while different castes may revert to familiar activities for example,
a young man from a community in which trading is common may choose to operate a shop), he is
23

unlikely to do so if farming is more remunerative. This would suggest that caste/occupation linkage does
not have substantial economic implications.
While this observation dovetails with some of the literature cited above, suggesting less than
perfect correlation between caste and occupation, it is important to be cautious about any generalization
based on the regressions presented in this paper. This paper has concentrated on the employment sector
ownership of a business or employment in salaried jobs and does not have much to contribute to
discussions about specific occupations. A focus on salaried employment and business ownership is
relevant to economic status of a household, given the vast differences in income between salaried
workers, owners of petty businesses and manual laborers; however, these controls may not accurately
reflect the relationship between caste and occupation. Given the affirmative action programs, dalits may
manage to find employment in government jobs but these could be low-level jobs rather than high-paying
jobs held by Brahmins. There is ample evidence to suggest that many Brahmins are employed in highstatus Indian administrative services (Goyal 1989), while the ranks of janitorial staff in government
services contain many dalit workers (Desai and Kulkarni 2008).
In spite of the three pathways this paper controls for, the final model in Table 2 continues to
demonstrate substantial inequalities between different caste groups in expenditure. After controlling for
all of these factors, the difference between annual per capita expenditure between Brahmins and other
high castes is Rs. 537, that between Brahmins and OBCs is Ra. 1,442, and that between Brahmins and
adivasis is Rs. 3,300. Since the average poverty line in India in 2004-5 was about Rs. 4,500 for rural areas
and Rs. 6,000 for urban areas, this is a substantial difference. These results, based on broad caste
categorization, do not discount the possibility that in some villages middle castes may have substantial
economic and political power and in others, Brahmins may indeed be poor village priests relying on
patron gifts for subsistence. However, these broad patterns are striking in their documentation of material
inequalities. Its implications for future research on caste and how caste is incorporated in broader social
stratification literature are discussed below.

24

Caste in Modern India:

The results presented above show that if status hierarchy based on Hindu scriptures has little
empirical support in modern India, economic disparities between large caste groupings continue to
flourish. What are the implications of these observations for research on caste in 21st century India, as
well as for the linkages between research on caste and broader stratification research? Caste in India has a
historically specific nature that has been further reified by a variety of public policies. However, once
stripped of its religious and ideological trope, it offers one of the most interesting examples of
consolidation of material resources in hands of certain groups.
Another such example is offered by the comparative historical research on societies in Central
and Eastern Europe, where transition from socialism to capitalism has led to the capture of economic
resources by certain groups (Eyal, Szeleny, and Townsley 1998; Rona-Tas 1998). In reconciling
scholarship on post-socialist transition with broader stratification research, Borocz (Borocz 1997) argues
for a middle space between the societies organized around the logic of estate where the individuals
position is determined by ascription based on membership in a group, conferred mainly by birth and
logic of status in which the basis of the individuals position is exclusively achievement portrayed as
returns on human capital endowments valorized in a wide open, virtually closureless terrain. His
advocacy, middle-ground of social reproduction is the terrain where new logics of distinction are created
as enclosures in the social field, resonates with the observations offered in this paper. Instead of hard
distinctions between caste, estate and class (e.g., Svalastoga 1965) based on varying degrees of closure
within each form of stratification, we will gain more analytical power by focusing on ways in which a
socially closed system like caste adapts and manipulates emerging class inequalities in a society
undergoing economic transformation.
The IHDS documents that caste and kin remain at the center of Indian civic life, with nearly 95%
of the female respondents reporting getting married within their own caste. This suggests that caste
structure remains largely defined by social closure, while the economic opportunities are increasingly
more open, creating new arenas in which castes must operate if they are to maintain their distinction.
25

Seeing Indian caste structure through the lens of Bourdieus notions of social reproduction (Bourdieu
1984) we begin to see a variety of ways in which castes manage to shape the access to social, political and
cultural capital to their members. Much has been written about the political power of various castes
(Gupta 2005) with a particular focus on the middle castes. However, the way in which upper castes
manage to establish and exercise their dominance over the opportunities to their members has received
little attention.
As education and entrepreneurship emerge as twin pillars of advancement in modern India,
historically wealthy castes play an interesting, often covert, role in shaping opportunities. Caste
associations organize private schools, colleges and charitable trusts through which members obtain
scholarships and loans for higher education. While these schools are ostensibly open to all, members of
the caste that established the school often receive priority. Scholarships are given based on
recommendations from members of the caste-based governing body. For rural students, educational
opportunities in cities are governed by their ability to obtain subsidized dormitory accommodations. A
search of hostels in Mumbai turns up dormitories with such identifiable caste names as Lad Baniya and
Modh Baniya. In addition to caste-based educational opportunities, many caste organizations have also
managed to set up cooperative banks, initially set up to serve caste members, and where caste members
continue to retain considerable clout. The tiny caste of Saraswat Brahmins is associated with a
surprisingly large number of banks. This access to capital is reflected in the fact that when IHDS
enumerated the source of loans obtained in the preceding five years, of the households that took out a
loan, 39% of the Brahmins borrowed from a bank or credit society while only 18% of the dalit households
did so. Among dalits, an overwhelming number resorted to private moneylenders, paying a considerably
higher rate of interest than if they had been able to borrow from a bank. These are just a few examples of
the way in which castes manage to parlay their historical privileges into opportunities for their members
and often use the vehicles such as tax-deductible status provided by the Indian state.
While the examples offered above are unique to India, they have much in common with other
studies of elite capture of state resources (Stark and Bruszt 1998) and ways in which social and cultural
26

institutions are manipulated to create and sustain inequalities (Arrow 2000; Bowles and Gintis 1976;
Bourdieu and Passeron 1990). With that recognition, it makes sense for students of social stratification to
stop treating India as a system representing an archetypical traditional form of social inequality and
begin to see it as an interesting and unique opportunity to study the way in which elites react to the
process of social transformation to retain their economic, social, cultural and symbolic power as well as
the way in which marginalized groups contest this hegemony.

27

References:
Appadurai, Arjun. 1986. Is Homo Hierarchicus? American Ethnologist 13 (4):745-761.
Arrow, Kenneth Samuel Bowles and Steven Durlauf. 2000. Meritocracy and Economic Ineqaulity.
Princeton: Princeton University Press.
Bardhan, Pranab. 1984. Land, Labor and Rural Poverty. New York: Columbia University Press.
Basu, Kaushik. 1994. Agrarian Questions. Delhi: Oxford University Press.
, ed. 2007. Oxford Companion to Economics in India. New Delhi: Oxford University Press.
Bayly, Susan. 1999. Caste, Society and Politics in India from the Eighteenth Century to the Modern Age,
New Cambridge History of India. Cambridge: Cambridge University Press.
Berman, Jan. 1974. Patronage and Exploitation: Changing Agrarian Relations in South Gujarat, India.
Berkelely, California: University of California Press.
Berreman, Gerald D. 1991. The Brahmanical view of Caste. In Social Stratification, edited by D. Gupta.
New Delhi: Oxford University Press.
Beteille, Andre. 1991. Society and Politics in India: Essays in a Comparative Perspective. London:
Athlone Press.
. 1992. The Backward Classes in Contemporary India. Delhi: Oxford University Press.
Borocz, Jozsef. 1997. Stand Reconstructed: Contingent Closure and Institutional Change. Sociological
Theory 15 (3):215-248.
Bourdieu, Pierre. 1984. Distinction: A Social Critique of the Judgement of Taste. Translated by R. Nice.
London: Routeledge. Original edition, 1979.
Bourdieu, Pierre;, and Jean-Claude Passeron. 1990. Reproduction in Education, Society and Culture.
Translated by R. Nice. Beverly Hills, CA: Sage. Original edition, 1977.
Bowles, Samuel, and Herbart Gintis. 1976. Schooling in Capitalist America: Educational Reforms and
Contradictions of American Life. New York: Basic Books.
Brass, Paul. 1990. The Politics of India since Independence, New Cambridge History of India.
Cambridge: Cambridge University Press.
Chatterjee, Partha. 1992. Caste and Subaltern Consciousness. In Subaltern Sutides: Writings on South
Asian History and Society, edited by R. Guha. Delhi: Oxford University Press.
Commander, Simon. 1983. The Jajmani System in North India; An Examination of Its Logic and Status
across Two Centuries. Modern Asian Studies 17 (2):283-311.
Crompton, Rosemary. 1998. Class and Stratification: An Introduction to Current Debates. 2nd ed.
Malden, MA: Blackwell Publishers.
Dantwala, M. L. 1950. India's Progress in Agrarian Reforms. Far Eastern Survey 19 (22).
Das, Veena. 1982. Structure and Cognition: Aspects of Hindu Caste and Rituals. Bombay: Manohar
Das, Veena 2003. Social Science and the Publics. In Oxford Companion to Sociology and Social
Anthropology, edited by V. Das. New Delhi: Oxford University Press.
Desai, Sonalde, and Veena Kulkarni. 2008. Changing educational inequalities in India: In the context of
affirmative action. Demography 45 (2):245-270.
Deshpande, Ashwini. 2000. Does Caste Still Define Disparity? A Look at Inequality in Kerala, India. The
American Economic Review 90 (2):322-325.
. 2007. Overlapping Identities under Liberalization: Gender and Caste in India. Economic
Development and Cultural Change 55 (4):735-760.
Deshpande, R. S. 2003. Current Land Policy Issues in India. Geneva: Food and Agricultural Organisation.
Dirks, Nicholas. 2001. Castes of Mind: Colonialism and the Making of Modern India. Princeton, NJ:
Princeton University Press.
Dube, Leela. 1996. Caste and Women. In Caste: Its Twentieth Century Avatar, edited by M. N. Srinivas.
New Delhi: Penguin Publishers.
. 2001. Anthropological Explorations in Gender: Intersecting Fields. New Delhi:: Sage
Publications.
28

Dumont, Louis. 1980. Homo Hierarchicus: The Caste System and its Implications. Chicago: University of
Chicago Press.
Dumont, Louis, and Andre Beteille. 1987. On Individualism and Equality. Current Anthropology 28
(5):669-677.
Eyal, Gil, Ivan Szeleny, and Eleanor Townsley. 1998. Making Capitalism without Capitalists: Class
Formation and Elite Struggles in Post-Communist Central Europe. London: Verso.
Frankel, Francine Zoya Hasan Rajeev Bhargava and Balveer Arora. 2000. Transforming India: Social and
Political Dynamics of Democracy. New Delhi: Oxford University Press.
Fuller, C. J. 1997. Caste Today. New Delhi: Oxford University Press.
Gerth, Hans;, and C. Wright Mills. 1946. From Max Weber: Essays in Sociology. New York.
Ghuyen, Gilles, ed. 2004. Who is a Brahmin? The Politics of Identity in India. New Delhi: Manohar
Books.
Gould, Roger V. 2002. The Origins of Status Hierarchies: A Formal Theory and Empirical Test.
American Journal of Sociology 107 (5):1143-1178.
Govinda, R. 2002. India Education Report. New Delhi: Oxford University Press.
Goyal, Santosh. 1989. Social Background of Officers in the Indian Administrative Service. In Dominance
and State Power in Modern India, edited by F. Frankel and M. S. A. Rao. Delhi: Oxford
University Press.
Grosh, Margaret, and Paul Glewwe. 2000. Designing Household Survey Questionnaires for Developing
Countries: Lessons from 15 Years of the Living Standards Measurement Study. Washington,
D.C.: The World Bank.
Grusky, David B., Manwai C Ku, and Szonja Szelenyi, eds. 2008. Social Startification: Class, Race and
Gender in Socilogical Perspective. Boulder: Westview Press.
Gupta, Dipankar. 1991. Social Stratification. New Delhi: Oxford University Press.
. 2000. Interrogating Caste: Understanding Hierarchy and Difference in Indian Society. New
Delhi: Penguin Books.
. 2005. Caste and Politics: Identity Over System. Annual Review of Anthropology 34:409-427
Khare, R. S. 1991. The Untouchable's Version: Evaluating an Ideal Ascetic. In Social Stratification,
edited by D. Gupta. New Delhi: Oxford University Press.
Kolenda, Pauline. 1973. Book Review: Homo Hierarchicus and Religion, Politics and History in India.
Journal of the American Oriental Society 93 (1):120-124.
. 1997. Caste in Contemporary India: Beyond Organic Solidarity. Jaipur, India: Rawat
Publications.
Liddle, Joanne, and Rama Joshi. 1989. Daughters of Independence: Gender, Caste and Class in India.
New Brunswick: Rutgers University Press.
Mandelbaum, David G. 1988. Women's Seclusion and Men's Honor: Sex Roles in North India,
Bangladesh and Pakistan. Tuscon: University of Arizona Press.
Marriott, McKim. 1969. Review: Homo Hierarchicus by Louis Dumont. American Anthropologist 71
(6):1166-1175.
Mayer, Adrian. 1997. Caste in an Indian Village: Change and Continuity 1954-1992. In Caste Today,
edited by C. J. Fuller. New Delhi: Oxford University Press.
Mendelsohn, Oliver, and Marika Vicziany. 1998. The Untouchables: Subordination, Poverty and the
State in Modern India. Cambridge: Cambridge University Press.
Milner, Murray, Jr. 1994. Status and Sacredness: A General Theory of Status Relations and an Analysis
of Indian Culture. New York: Oxford University Press.
Mohanty, Mritiunjoy. 2006. Social Inequality, Labour Market Dynamics and Reservation. Economic and
Political Weekly 41:3777-3789.
Narayan, Badri. 2004. Inventing Case History: Dalit Mobilisation and Nationalist Past. In Caste in
Question: Identity or Hierarchy?, edited by D. Gupta. New Delhi: Sage Publications.
Omvedt, Gail. 2006. Dalit Visions : The Anti-Caste Movement and the Construction of an Indian Identity.
New Delhi: Orient Longman.
29

Panini, M. N. 1997. The Political Economy of Caste. In Caste: Its Twentieth Century Avatar, edited by
M. N. Srinivas. New Delhi: Penguin Books.
Papanek, Hanna. 1973. Purdah: Separate Worlds and Symbolic Shelter. Comparative Studies in Society
and History 15 (3):289-325.
Raheja, Gloria Goodwin. 1988. India: Caste, Kingship, and Dominance Reconsidered. Annual Review of
Anthropology 17:497-522.
Rona-Tas, Akos. 1998. Path Dependence and Capital Theory: Sociology of the Post-Communist
Economic Transformation. East European Politics and Society 12 (1):107-31.
Sharma, K. L. 1999. Social Inequality in India: Profiles of Caste, Class and Social Mobility. Jaipur:
Rawat Publicastions.
Srinivas, M. N. 1956. A Note on Sanskritization and Westernization. The Far Eastern Quarterly 15
(4):481-496.
. 1977. The Changing Position of Indian Women. Man 12 (2):221-238.
. 1987. The Dominant Caste and Other Essays. New Delhi: Oxford India Paperbacks.
. 1996. Caste: Its Twentieth Century Avatar. New Delhi: Penguin Books.
Stark, David, and Laszlo Bruszt. 1998. Postsocialist Pathways: Transforming Politics and Property in
East Central Europe. Cambridge: Cambridge University Press.
Sundar, Nandini. 1999. The Indian Census, Identity and Inequality. In Institutions and Inequalities:
Essays in Honour of Andre Beteille, edited by R. Guha and J. Perry. New Delhi: Oxford
University Press.
Weber, Max. 1958. The Religion of India: The Sociology of Hinduism and Buddhism. Translated by H.
M. Gerth, Don. Gelncoe, Ill.: Freepress.
Wiser, William Henricks. 1979. The Hindu jajmani system : a socio-economic system interrelating
members of a Hindu village community in services. New York: AMS Press.

While caste information was also collected in 1941, it was not tabulated due to wartime resource shortages.
While the formal term for OBC is Other Backward Classes, it actually refers to castes, and OBC lists for various
states contain a list of castes that are considered backward for that state. Sometimes these lists differ from district to
district.
iii
While not presented here, the analysis was also carried out for women aged 25-49 with similar caste-based
differences in their educational attainment.
iv
A similar analysis was performed for expenditure by brides family and ratio of grooms and brides family
expenditure. Results show declining expenditure for brides family by caste status and little relationship between
caste and the ratio of grooms and brides expenditure. This suggests that the caste pattern of expenditure is related
to conspicuous consumption by richer castes rather than a preconceived hierarchy based on ritual status.
ii

30

Fig. 1: Caste and Religious Distribution

of IHDS Households
High Caste
15%

Brahmin
5%

OBC
36%

Other Relig.
6%
Muslim
12%

Dalit
20%
Adivasi
6%

Table 1: Predicted Values on Status Indicators for Different Caste/Religious Groups

Eat
Parctice
Meat Purdah/Ghunghat
Brahmin
High Caste
OBC
Dalit
Adivasi
Muslim
Other Relig.

19%
50%
53%
70%
67%
90%
65%

51%
44%
52%
54%
39%
89%
28%

Avg. Wedding
Exp. For Grooms
In Rs.
56532
53970
47308
38020
33135
41599
38450

* The caste/religious group differences are significant at .001 level in a test of joint significance
for all outcome variables

Table 2: Predicted Annual Per Capita Expenditure

Adjusted
For Residence

Brahmin
High Caste
OBC
Dalit
Adivasi
Muslim
Other Relig.

11959
10435
8678
7210
6391
7491
9136

Landholding
adjustments
added

Education
adjustments
added

Emp. Sector
adjustments
added

11649
10163
8649
7448
6435
7612
9182

10161
9604
8698
7752
6812
8055
9094

10122
9603
8680
7810
6822
8017
9115

* The caste/religious group differences are significant at .001 level in a test of joint significance
for all outcome variables

Table 3: Predicted Values for Land Holding, Education and Salaried/Business Employment
Land Owned
in Acres

Brahmin
High Caste
OBC
Dalit
Adivasi
Muslim
Other Relig.

2.01
2.02
1.69
1.27
1.55
1.44
1.59

For Men 25-49

Years of
Salaried/
Education
Business
Employment
10.82
8.78
7.28
5.93
5.06
5.39
7.33

46%
40%
41%
34%
33%
47%
39%

* The caste/religious group differences are significant at .001 level in a test of joint significance
for all outcome variables

App Tab 1: Distribution of Status Indicators and Material Conditions

Brahmin
High Caste
OBC
Dalit
Adivasi
Muslim
Other Relig.
All India

Eat
ParcticeAvg. Wedding
Meat Ghunghat/
Exp. For
Purdah
Groom
Rs.
24%
63%
107226
55%
47%
85012
57%
52%
57855
69%
57%
42266
61%
53%
31276
88%
84%
55752
64%
23%
63603
62%

55%

59908

Household
Males 25-49
Consump.
Land
EducationIn Salaried
Exp.
Owned Years Comp.
Emp./
per capita
Acres
Business
Rs.
16015
2.23
10.92
60%
14254
2.27
8.89
49%
10015
1.71
6.84
37%
8102
0.67
5.35
28%
6142
1.95
4.15
20%
9253
0.70
5.48
43%
12817
1.30
7.77
40%
10441

1.49

6.85

38%

App Tab 2: Coefficients from Multivariate Analyses: Status Indicators

Meat Consumed in Home
in last month
Coefficient
t-statistic
Caste/Religion (Brahmin omitted)
High Caste
OBC
Dalit
Adivasi
Muslim
Other Relig.
Residence (Metro city omitted)
Other Urban
Developed village
Less Developed Village
Maximum adult education (none omitted)
1-4 Years
5-9 Years
10-11 Years
12-Some College
College degree
Log Per Capita Consumption Exp.
Constant
Observations
R-squared
Chi-Square (36 df)

1.47
1.56
2.32
2.18
3.65
2.08
-0.27
-0.38
-0.28
0.07
0.11
-0.05
-0.16
-0.21
0.33
-4.79

15.85
17.35
23.77
18.40
31.31
18.12

Women practice
Ghunghat/Pardah
Coeff. t-statistic
**
**
**
**
**
**

-0.26
0.04
0.14
-0.49
2.10
-0.96

-3.11
0.49
1.62
-4.36
18.59
-8.68

-3.68 **
-5.06 **
-3.61 **
0.81
2.02
-0.69
-2.27
-2.73
10.18
-15.13

*
*
**
**
**

** p <= 0.01, * p<= 0.05

**
**
**

-0.05
-0.18
-0.40
-0.53
-0.31
-0.39

-1.64
-6.19
-12.77
-14.70
-9.71
-11.12

-0.01
0.80
0.72

-0.07
8.77 **
8.57 **

-0.13
-0.17
-0.36

-7.83 **
-8.99 **
-18.75 **

0.04
0.24
0.15
0.23
0.00
-0.05
1.42

0.47
4.26
2.19
2.98
0.02
-1.44
4.64

41456

33307

5520.53

7068.88

State of residence dummies included in each regression but not reported here.

Log Wedding Expen.

for Grooms
Coefficient t-statistic
**

**
*
**

**

0.14
0.23
0.38
0.50
0.62
0.22
9.17
41190
0.33

6.02
13.07
16.69
21.96
24.76
20.29
85.12

**
**
**
**
**

**
**
**
**
**
**
**

App Tab 3: Coefficients from Multivariate Regression for Per Capita Consumption Expenditure
Residence
Controls only
Coefficientt-statistic
Caste/Religion (Brahmin omitted)
High Caste
-0.14
-6.30
OBC
-0.32 -15.47
Dalit
-0.51 -23.84
Adivasi
-0.63 -24.60
Muslim
-0.47 -21.16
Other Relig.
-0.27 -10.13
Residence (Metro city omitted)
Other Urban
-0.20 -12.27
Developed village
-0.54 -33.04
Less Developed Village
-0.61 -38.76
Log of Land Owned by Household
Maximum household adult education (none omitted)
1-4 Years
5-9 Years
10-11 Years
12-Some College
College degree
Anyone in household in salaried job/business
Constant
9.71 368.90
Observations
R-squared

41514
0.24

Add
Land Size Control
Coefficientt-statistic

Add Salaried job

or Business Control
Coefficient t-statistic

**
**
**
**
**
**

-0.14
-0.30
-0.45
-0.59
-0.43
-0.24

-6.30
-14.33
-20.95
-23.50
-19.22
-8.93

**
**
**
**
**
**

-0.06
-0.16
-0.27
-0.40
-0.23
-0.11

-2.74
-7.73
-13.10
-16.05
-10.84
-4.32

**
**
**
**
**
**

-0.05
-0.15
-0.26
-0.39
-0.23
-0.11

-2.58
-7.72
-12.68
-16.01
-10.99
-4.12

*
**
**
**
**
**

**
**
**

-0.21
-0.62
-0.71
0.13

-13.06
-37.19
-43.34
20.61

**
**
**
**

-0.18
-0.47
-0.53
0.09

-11.82
-29.17
-33.09
13.82

**
**
**
**

-0.17
-0.43
-0.49
0.10

-11.40
-26.93
-30.24
15.58

**
**
**
**

0.03
0.09
0.24
0.34
0.59

2.18
7.80
16.75
18.98
37.25

*
**
**
**
**

0.03
0.07
0.20
0.29
0.53
0.13
9.20

1.68
5.90
14.01
16.15
32.80
14.51
326.50

**
**
**
**
**
**

**

9.68
41356
0.25

State of residence dummies included in each regression but not reported here.
** p <= 0.01, * p<= 0.05

Add
Education Control
Coefficientt-statistic

368.40 **

9.26
41298
0.31

336.20 **

41298
0.32

App Tab 4: Coefficients from Multivariate Analyses: Land Ownership, Education and Salaried Job or Business Employment

Caste/Religion (Brahmin omitted)

High Caste
OBC
Dalit
Adivasi
Muslim
Other Relig.
Residence (Metro city omitted)
Other Urban
Developed village
Less Developed Village
Years of Completed Education
Constant

R-squared
Chi Square (31df)

Acres of Land Owned

by the Household

Years of Education
For Males 24-49

Coefficient t-statistic

Coefficientt-statistic

0.00
-0.17
-0.46
-0.26
-0.34
-0.24

**
**
**
**
**

-2.04
-3.54
-4.89
-5.76
-5.43
-3.49

-15.47
-28.15
-36.26
-33.10
-38.39
-21.33

**
**
**
**
**
**

-0.27
-0.24
-0.54
-0.56
0.01
-0.32

-3.31
-3.00
-6.39
-4.86
0.13
-3.28

**
**
**
**

0.10
0.65
0.74

8.67 **
46.45 **
54.15 **

-0.54
-2.58
-3.19

-4.62 **
-20.80 **
-26.64 **

8.28 **

12.76

72.26 **

-5.90
-21.77
-27.51
32.77
0.00

**
**
**
**

0.24

-0.38
-1.56
-1.91
0.13
0.00

41356
0.22

0.18
-6.71
-17.98
-8.19
-13.01
-8.15

Emp. In Salaried Job

or own business
For Males 24-49
Coefficient t-statistic

35259
0.19

35259
4273

**