Skip to navigation – Site map

HomeOther ContributionsBook Reviews2018Divya Vaid. Uneven Odds: Social M...


Divya Vaid. Uneven Odds: Social Mobility in Contemporary India

Mathieu Ferry
Bibliographical reference

Vaid, Divya. 2018. Uneven Odds: Social Mobility in Contemporary India. New Delhi: Oxford University Press. 356 pages.

Full text

1Despite the existence of a large body of literature on intergenerational social mobility within Indian society, UnEven Odds by sociologist Divya Vaid offers a unique contribution to the field. The book is drawn from her Doctoral thesis and subsequent publications over the years (notably Vaid and Heath 2010, Vaid 2012, 2014a, 2014b, 2016), updated with more recent data. It is the first comprehensive study of the whole country that theoretically questions and quantitatively tests the links between class of origin and destination, as well as caste and gender identities, and rural-urban locations. With a few exceptions, previous studies have largely adopted a more localized approach, focusing on particular areas (Chaudhuri 1987) or certain castes (Chekki 1970). Further, many of these studies have relied on ethnographic and qualitative methods, and while their approach makes an important contribution to our understanding of the experiences of mobility (for instance see Naudet 2018), it tells us little about the degree (or lack) of mobility in Indian society.

2A graduate of Nuffield College, Oxford University, Divya Vaid adopts the class schematic approach to social mobility, a conceptualization internationally popularized by John H. Goldthorpe and his colleagues (Erikson and Goldthorpe 1992). This neo-Weberian approach defines social class as similar positions embedded in market and work situations, which characterize common life chances. A widely used class schema, known as the “EGP class schema” (after the three initial contributors to this typology: Robert Erikson, John H. Goldthorpe and Lucienne Portocarero) has hence been conceptualized and validated over the years in industrialized countries. Kumar, Heath, and Heath (2002a, 2002b) have been largely inspired by this approach to initiate social mobility studies in India and Divya Vaid has followed their lead, given the lack of standardized class schemas in Indian research. The schema is adapted to the idiosyncrasies of the Indian case with respect to its industrial development. Indeed, even though there has been substantial economic growth over the past thirty years, and particularly in the 1990s in the era of liberalization, the agricultural sector remains the predominant employer (about 50 per cent). Most occupations belong to the informal sector (83 per cent according to Harriss-White 2004) and are characterized by low income, lack of employment contracts and almost no social security benefits. Hence, Vaid’s 11-class schema (and a 5-class schema in a collapsed version for sample size limitations) emphasizes distinctions among agricultural occupations based on land ownership and the grouping of occupations reflects the degree of job security and permanency.

3This formalization also highlights a clear distinction between class and caste. Indeed, the existence of one form of stratification does not preclude another, and caste and class are not a continuum. Social mobility studies in India sometimes associate these categories, merging concepts of class, status, prestige and economic situation. But the author draws again from Weber’s concepts and identifies caste with “status” groups, which are separated on the basis of lifestyle, and the honor and prestige they hold in a community. Caste can be considered an extreme form of status where distinctions are guaranteed by rituals. This does not mean that caste does not affect material well-being, but whereas class affects it directly through the market, “status affects material well-being indirectly, through the ways that categories of social honor underwrite various coercive mechanisms” (Saunders 1990:5). Thus caste is an ascribed characteristic that allows for the control of jobs through the mobilization of specific resources, rather than jati corresponding directly to certain occupations (Vaid 2014a).

4The whole book is driven by the ambition to test the modernization thesis in India. Within this framework, also called the liberal theory of industrialism (Treiman 1970), social mobility is supposed to increase over time, that is to say one should observe a distancing of origin (in particular, father’s class) from destination (one’s class). There is a twofold explanation for this trend. First, one should observe an increase in absolute mobility, or more people in upward rather than downward mobility, due to structural changes. As industrialization occurs, there is a decline in the number of people engaged in agriculture and a shift from manual to non-manual occupations, along with the emergence of new occupations. Second, relative mobility, also called social fluidity, should increase. This concept captures the degree of inequality of opportunities within a society because it is measured “net” of the structural effects. This increase is supported by the growing importance of educational attainment in class destination and a decline of the salience of ascribed characteristics such as gender and caste. In short, industrialized societies would become more “meritocratic.” The author nonetheless reminds us that although structural mobility has indeed increased in industrialized countries, a “constant flux” remains within and across countries in terms of relative social mobility rather than a constant increase in relative mobility over time (Erikson and Goldthorpe 1992).

  • 1 In particular, the lack of consistent variable collection in the surveys since their inception (196 (...)

5Locating Indian society within this framework requires representative surveys that would permit the social scientist to build mobility tables, the core tool of such analyses. Individual, and not household, units of analysis are required; Vaid makes it a point to distinguish male and female patterns. These surveys need to include fathers’ occupations, individual occupations, caste information and educational attainment levels. Ideally, the surveys should be repeated at different points in time to measure the evolution of social mobility. Since the National Sample Survey Office, India’s public statistical institute, does not offer such detailed surveys, Divya Vaid mobilizes the National Election Studies (NES) produced by the Center for the Study of Developing Societies. These data concentrate on political behavior, the opinion and attitudes of the Indian electorate, but they provide all the required variables for a social mobility study, albeit not perfectly.1 Whereas fairly basic statistical tools, such as row frequencies and dissimilarity indices, measure absolute mobility, relative mobility requires more refined statistics such as odds ratios and log-linear models. While these are classical tools in the sociology of social mobility, the unfamiliar reader can only appreciate Vaid’s didactic summaries in each chapters and her step-by-step argumentation.

6To briefly summarize Vaid’s work, Indian society is remarkable for its intergenerational class stability, as developed in the empirical Chapters 4 to 6 and in Appendix B. The association between class origin and class destination is very strong, in other words people’s class is largely dependent upon their father’s class. But this statement should be qualified. Women experience higher stability, particularly in rural India, where over time they tend to be clustered in low agricultural occupations or non-permanent contract jobs. One observes a slight increase in absolute mobility for younger birth cohorts, with more upward than downward mobility, and more social fluidity for men, but the results are inconclusive for women. Hence, gender inequalities on the labor market seem to result more from opportunity structures available to women and men, rather than gender differences in fluidity patterns. The lack of social fluidity both men and women face, shows how privileged families manage to preserve their prerogatives for both women and men. Urban India appears more fluid, which is hardly surprising when Vaid finds high levels of “stickiness” (lack of relative mobility) in the agricultural workforce and little crossover between manual and non-manual occupations over generations. Overall, the picture of intergenerational class mobility in India is hence one of “continuity and change” (Vaid 2018:166), with high stability, but steady increases in absolute mobility, and changes in relative mobility for men.

7Further, even though the author does not measure a clear association between caste and class with her statistical categories, a certain congruence does exist. In particular, caste clustering is identified at the extremes of the class schema, both when looking at class origin and destination. More specifically, higher castes (jatis that are neither reserved groups nor peasant castes) are more often found in higher classes, especially in white collar and “clean work” occupations, whereas Scheduled Castes are more commonly associated with manual work and low agricultural positions. Muslims show an interesting concentration in the business classes, but they are underrepresented in other higher classes such as amongst the professionals. Overall, caste-class congruence diminishes when looking at younger birth cohorts, but it is still present. Nonetheless, the impact of caste should not be overestimated in class destination. Indeed, the effect of class origins appears to be far more important than the caste effect when measuring the odds of belonging to the professional class on the basis of class of origin and caste (using a logistic regression). These results are consistent over the different birth cohorts, with the caste effect diminishing slightly over time. Finally, contrasting intergenerational class mobility by caste shows that whereas Scheduled Castes find it harder than other groups to take advantage of upward mobility, upper castes are not protected from downward mobility. But in general, families use class along with caste to protect their children from downward mobility. While they are not always successful in this, higher castes experience a clear advantage.

8Divya Vaid then moves on to question whether education tempers the barriers to social mobility faced by certain groups. She clearly shows a strong difference between individuals located in urban and rural areas. In urban areas education is far more salient in class destination, and this applies to women too. Hence urban areas seem to provide opportunities to move beyond ascribed characteristics, although she underscores that women’s access to the urban remains restricted. In general, higher levels of education are associated with higher absolute and relative mobility, especially upward mobility. Interestingly, when modelling the odds of belonging to the professional class, she again highlights the significance of class origin, but here, she outlines the educational attainment effect on class destination, which seems to weaken and even cancel the caste effect. This key result indicates that rather than directly influencing occupational position, caste acts indirectly through the potential for educational attainment.

9While considering the differences of mobility between men and women, the author discusses and addresses a crucial limitation of her study: the workforce participation rate for women has been consistently below 30 per cent in the last four decades and has even declined in the last two NSSO surveys (2009–2010 and 2011–2012). Without entering into a debate on the possible flaws in these indicators, she nonetheless acknowledges that focusing solely on women’s mobility on the labor market might result in an underestimation of the degree of potential mobility for women in India. Therefore she suggests measuring their mobility on the marital market, as already studied in industrialized societies, notably by Glenn, Ross, and Tully (1974). In this framework, class destination is not based on employment occupations but on the husbands’ occupation. One can regret that this complementary analysis only appears in the Appendix even though she refers to it in her conclusion. In India, where arranged marriages are extremely common, one expects to observe a high level of female intergenerational stability. At the same time, hypergamy, in “closely related rather than widely separated ranks” (Davis 1941:381) occurs within caste endogamous marriages. Since this hypergamy affects caste along with class, one might expect women from lower castes to be more upwardly mobile than other women in terms of class. But the analysis does not support this view; marriage is not an avenue of upward mobility for women. On the contrary, mobility on the marriage market for women is lower than mobility for men on the labor market. Consequently, for high caste families the marriage market for their daughters is more efficient than the labor market for their sons in terms of maintaining their class position.

10Social mobility studies are nearly solely available in industrialized countries, and the present study is an important step to move beyond this research gap. India appears more stable, less fluid, and seems to have higher degrees of “stickiness” than Britain, France, Poland or Ireland (for which data are presented). Compared to these industrialized countries, the class origins of men and women are more dissimilar, because Indian women engaged in the workforce are “selected” and tend to come from lower classes. On the contrary, the class destination of men and women is more similar in India than in other countries, because even though women are more concentrated in less privileged classes, there is less sex-segregation on the labor market in India. Overall, Divya Vaid shows how the modernization thesis developed in the West is not adaptable to the Indian case, as she concludes: “‘Western models of modernity and modernization are not necessarily useful concepts for understanding contemporary India’ (Vaid and Heath 2010:159). What one needs is a more nuanced look at the experiences specific to a large diverse country undergoing economic change which may be benefitting some, but which is also leaving quite a few behind” (Vaid 2018:253). In that regard, the lack of a more thorough analysis of reservation policies as a possible driver of social change is surprising.

11In her conclusion, the author evokes Grusky and Sørensen’s (1998) suggestion to study social mobility at a more disaggregated level. Their argument suggests that occupations rather than big classes would better capture “social closure.” This reference is somewhat surprising as these authors intend to respond to the postmodernist discourse of the diminished class membership influence on social choice. But if anything, Vaid’s empirical results show that class has been a fairly underestimated variable in Indian research and that it has quite strong effects. Rather than disentangling her fruitful class conceptualization, one would prefer to extend the empirical premises of class validity as suggested in Appendix A. In particular, focusing only on “primary occupation,” that is to say the occupation from which people earn the most, or in which they are employed most of the time, might be a bit narrow to allocate individuals into classes. Since Vaid’s class schema aims to reflect job security and permanency, the paradox is then that she identifies class stability using a schema where lower classes are themselves characterized by high job instability. It is not an issue as long as people are unstable within their own class. For instance, if people are agricultural laborers during harvests and rely on poultry farming the rest of the time, they remain “low agriculturists” according to her schema. But if, instead, they migrate seasonally and are at times bricklayers, then they become “semi and unskilled manual workers.” Or if they just rely on their small shop the rest of the year, they join the entrepreneurial “petty business” class. Since “pluri-activity” (Jodhka 2018) is an important, growing phenomenon in rural India, it needs to be captured using adequate quantitative tools. The Indian Human Development Surveys, jointly organized by researchers from the University of Maryland and the National Council of Applied Economic Research, precisely aims to capture this diversity of economic resources at the individual and household unit level. As described in survey manuals: “The combination of information from the different activity streams provides a holistic picture of the types of economic activities undertaken by all individuals in the households in the preceding year” (Desai and Vanneman 2015:17). “Pluri-activity” also means pluri-occupations but it does not imply pluri-class, it is rather a defining characteristic of Vaid’s class schema that should be taken into account in its statistical categories. Luckily, these surveys also allow for an intergenerational mobility analysis with data collection on the head of the household or father’s occupation and educational attainment. Mobilizing these data could then reinforce the operationalization of the class schema.

  • 2 The NES 2014 mobilized in this chapter surveyed 21,967 individuals.
  • 3 The attribution of jatis to eight categories is not detailed but one can refer to the chapter appen (...)

12Given that chapter 3 adequately summarized “Class debates and class schemas,” one would have expected a similar literature review on caste in a subsequent chapter. But the author chooses to review the role of caste on the labor market in chapter 5 (based on Vaid 2014a) with only a brief discussion of its statistical operationalization. The caste variable, with its eight categories called “communities” because Muslims and other religious minorities are included as two separate entities, consists essentially of groups who benefit from the reservation policies (Other Backward Classes, OBC, Scheduled Castes, SC and Scheduled Tribes, ST). As explained in the Appendix to the chapter, the “creamy layer” forms a subcategory of the OBC category and the “peasant high castes” separate the General from the other “high castes.” It is perfectly understandable that the analysis is based on big categories of caste and not jatis because of sample size issues.2 Groups benefiting from institutionalized reservations are often mobilized to operationalize caste because it is the only source of information about caste in the surveys (see for instance Desai and Dubey 2012 or Ferry, Naudet, and Roueff 2018), even though it is an imperfect proxy. But as Vaid clarifies: “The information on community or caste was gathered from question Z5 in the survey: ‘What is your Caste/Jati-biradari/Tribe name? And your sub caste?’ Responses to this question were coded into the master list of over 240 names obtained from Lokniti, Centre for the Study of Developing Societies (CSDS)” (Vaid 2018:205).3 Hence, the author could have chosen categories other than the reserved groups, all the more so because quota policy evaluation is not a focus of the book. Considering that in her introduction she conceptualized caste as status following Weber, these categories certainly appear as a proxy for measuring it and overcoming the geographic diversity of caste on the subcontinent. But just as class categories require conceptual justification for their operationalization, one might expect the same for caste. In fact, as there is no standardized class schema in India, we do not really know whether the standardized caste schema commonly applied to reserved groups really captures what we intend it to. Therefore it might be time for quantitative sociologists to think of validity tests for caste schemas, just as they do for class (Evans 1992, as carried out by the author in Appendix A).

13Overall, UnEven Odds is not only remarkable because it is the first book to really analyze class mobility in India, it is also a model of thorough quantitative analysis. It definitely opens the way for further quantitative research in social stratification as indicated in the book’s conclusion. It offers an important theoretical conceptualization of social class and achieves an ambitious empirical analysis of social mobility in India. Discussing sociological literature not specific to India, it contributes to debates in sociology beyond the Indian case. At the same time, both qualitative and quantitative scholars working on the subcontinent will certainly find it useful in their own field surveys.

Top of page


Chaudhuri, Maitreyi. 1987. “Class Analysis in a Village Society: A Cluster Theoretic Approach.” Economic and Political Weekly 22(49):2121–25.

Chekki, Dan A. 1970. “Social Stratification and Trends of Social Mobility in Modern India.” Sociologus 20(2):146–63.

Davis, Kingsley. 1941. “Intermarriage in Caste Societies.” American Anthropologist 43(3):376–95.

Desai, Sonalde and Amaresh Dubey. 2012. “Caste in 21st Century India: Competing Narratives.” Economic and Political Weekly 46:40–49.

Desai, Sonalde and Reeve Vanneman. 2015. “India Human Development Survey-II (IHDS-II), 2011–12.”

Erikson, Robert and John H. Goldthorpe. 1992. The Constant Flux: A Study of Class Mobility in Industrial Societies. Oxford: Clarendon Press.

Evans, Geoffrey. 1992. “Testing the Validity of the Goldthorpe Class Schema.” European Sociological Review 8(3):211–32.

Ferry, Mathieu, Jules Naudet, and Olivier Roueff. 2018. “Seeking the Indian Social Space: A Multidimensional Portrait of the Stratifications of Indian Society.” SAMAJ, Freestanding Articles. Retrieved October 10, 2018 (

Glenn, Norval D., Adreain A. Ross, and Judy Corder Tully. 1974. “Patterns of Intergenerational Mobility of Females Through Marriage.” American Sociological Review 39(5):683–99.

Grusky, David B., and Jesper B. Sørensen. 1998. “Can Class Analysis Be Salvaged?” American Journal of Sociology 103(5):1187–34.

Harriss-White, Barbara. 2004. India Working: Essays on Society and Economy. South Asian Edition. Cambridge: Cambridge University Press.

Jodhka, Surinder. 2018. “Rural Change in Times of ‘Distress.’” Economic and Political Weekly 53(26–27):5–7.

Kumar, Sanjay, Anthony Heath, and Oliver Heath. 2002a. “Changing Patterns of Social Mobility: Some Trends over Time.” Economic and Political Weekly 37(40):4091–96.

Kumar, Sanjay, Anthony Heath, and Oliver Heath. 2002b. “Determinants of Social Mobility in India.” Economic and Political Weekly 37(29):2983–87.

Naudet, Jules. 2018. Stepping into the Elite: Trajectories of Social Achievement in India, France, and the United States. S.I.: OUP INDIA.

Saunders, Peter. 1990. Social Class and Stratification. London; New York: Routledge.

Treiman, Donald J. 1970. “Industrialization and Social Stratification.” Sociological Inquiry 40(2):207–34.

Vaid, Divya. 2014a. “Caste in Contemporary India: Flexibility and Persistence.” Annual Review of Sociology 40(1):391–410.

Vaid, Divya. 2014b. “Class and Social Mobility.” Seminar 663.

Vaid, Divya. 2016. “Patterns of Social Mobility and the Role of Education in India.” Contemporary South Asia 1–28.

Vaid, Divya. 2012. “The Caste-Class Association in India: An Empirical Analysis.” Asian Survey 52(2):395–422.

Vaid, Divya and Anthony Heath. 2010. “Unequal Opportunities: Class, Caste and Social Mobility.” Pp. 129–64 in Diversity and Change in Modern India: Economic, Social and Political Approaches, edited by A. F. Heath and R. Jeffery. Oxford: Oxford University Press.

Top of page


1 In particular, the lack of consistent variable collection in the surveys since their inception (1967, and then 1996, 1998, 1999, 2004, 2009 and 2014) leads her to focus on the last two periods for her analysis, and to adopt a birth-cohort approach to study the evolution of mobility rates over time.

2 The NES 2014 mobilized in this chapter surveyed 21,967 individuals.

3 The attribution of jatis to eight categories is not detailed but one can refer to the chapter appendix in Vaid and Heath (2010) which used a similar variable.

Top of page


Electronic reference

Mathieu Ferry, Divya Vaid. Uneven Odds: Social Mobility in Contemporary IndiaSouth Asia Multidisciplinary Academic Journal [Online], Book Reviews, Online since 10 October 2018, connection on 18 May 2024. URL:; DOI:

Top of page

About the author

Mathieu Ferry

SciencesPo, Observatory for Social Change

By this author

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search