The demand for a caste census has been widely raised by reserved quota communities in recent times, particularly by the Other Backward Classes (OBCs). It has now become a social justice agenda. For these groups, the census is not simply a headcount of community populations but is about identity, equity, resource distribution, and political participation. It is also believed that it would lay bare the historically embedded inequality, exclusion, and privilege of castes and communities—what Opposition leader Rahul Gandhi has called an X-ray of society.
A distinctive feature is the Composite Backwardness Index …a comprehensive measure based on 42 equally weighted parameters spanning various categories, and the first of its kind in the country.
Pressure from these quarters made the caste census an election agenda in the last General Election, with the Indian National Congress and other regional parties including it in their manifestos. Yet there has been ambivalence and apprehension about the outcomes of a caste census, particularly among the dominant castes, who fear it would lead to further division of society and the country. Another question is how far one can probe the everyday realities of caste, and whether it is possible to capture the deeper meanings of caste in daily life. The Telangana caste survey offers many takeaways for a nationwide caste survey.
Distinctiveness of the Telangana Survey
The caste survey conducted by the Government of Telangana between November and December 2024 had many surprises and covered 97% of the state’s population. Published as the Socio-Economic, Educational, Employment, Political and Caste (SEEEPC) Survey 2024, it documented the relative backwardness of 35 million people in Telangana, organised into 242 castes—of which 59 are Scheduled Castes (SCs), 32 Scheduled Tribes (STs), 134 Backward Classes (BCs), and 18 Other Castes (OCs).
The survey schedule included 57 questions and 75 subfields covering all aspects of caste, including discrimination. The government also constituted a nine-member Independent Expert Working Group (IEWG)—of which I was a member—to analyse this large dataset comprehensively and submit a report. The appointment of the IEWG was a new factor in the survey, making it unique among other states’ caste surveys.
Another distinctive feature is the Composite Backwardness Index (CBI), developed by the IEWG—a comprehensive measure based on 42 equally weighted parameters spanning various categories, and the first of its kind in the country. Even the Mandal Commission employed only 11 parameters, and not in the model developed by the IEWG.
The CBI differentiates between rural and urban living standards and uses a statistical quartile distribution to measure the gap between the most and least backward quartiles of castes for each parameter. A caste’s CBI score can range from 0 to 123, with 0 indicating the least backward and 123 the most backward. The higher a caste’s CBI score, the more backward it is relative to other castes in the state.
Indices of Backwardness
In Telangana, SCs emerge as the most backward social group, with a CBI score of 96. STs follow closely with a score of 95. The slight difference between the two may be due to STs having more landholdings than SCs. Together, SCs and STs are three times as backward as the dominant castes in the state. For comparison, the state’s overall weighted average CBI score is 81, indicating that both SCs and STs are nearly 15 points more backward than the average person in Telangana. BCs have a score of 86, also below the state average. In contrast, OCs have a significantly lower CBI score of 31, suggesting a much higher level of development and prosperity. It is important to note that a higher CBI score reflects greater backwardness.
The data also reveals that 99% of all ST individuals—belonging to 25 of the 32 recognised ST communities—are more backward than the state average. Similarly, 97% of SC individuals, from 41 of the 59 SC castes, fall below the state average. Among BCs, 71% of people from 69 castes are also more backward than average. On the other hand, all individuals from the 18 OC groups are more developed than the state average.
Economic Backwardness?
Going by this data, there is no economic backwardness among OCs in Telangana. The same could be the case in other parts of the country—in other words, there are no Economically Weaker Sections visible among OCs in Telangana. The question that then arises is: what percentage of OCs have an annual income below Rs. 8 lakh?
This also raises the question of how to understand the implementation of the 10% reservation for Economically Weaker Sections (EWS) of OCs in the state and the country. The EWS quota was introduced through the 103rd Constitutional Amendment in 2019, without any study or estimation of the economically weaker sections in the country.
Supporters of the EWS quota foregrounded economic disadvantage as the primary basis of backwardness in India, sidelining the role of caste.
The Supreme Court subsequently upheld it in November 2022, ruling that the quota does not violate the basic structure of the Constitution and, on the contrary, fulfils the Constitution’s promise of justice. The Court also clarified that the EWS quota does not infringe upon the rights of SC, ST, and OBC communities, and that the 50% cap on reservations does not apply to it.
Following the announcement of the EWS quota, a range of literature has been produced on its validity and efficacy. Supporters of the EWS quota foregrounded economic disadvantage as the primary basis of backwardness in India, sidelining the role of caste.
Central Indicator of Backwardness
However, the Telangana Caste Survey demonstrates that caste remains the central indicator of social and economic backwardness. Based on population share, 99% of ST members, 97% of SC members, and 71% of BC members are more backward than the state average. In contrast, 100% of those in the General Caste category are better off than the state average. Even among the very poor—those with an annual income of less than Rs 1 lakh—OC poor are less backward than SC and ST poor. Notably, 34% of poor OC children study in private schools, compared to just 5% for SCs and STs.
The poverty level among numerically larger castes and communities is very striking. Just four castes out of the 242 in Telangana—Madigas (SC), Lambadis (ST), and Mudirajs and Yadavas (both BCs)—make up nearly 40% of the 11 million people living in extreme poverty, defined as having an annual income of less than Rs 1 lakh. Although these four castes constitute 33% of the state’s population, their share of the extremely poor is disproportionately high, highlighting their economic vulnerability.
More broadly, BCs, SCs, and STs collectively account for a significantly larger share of the population living on incomes below Rs 1 lakh than the General Caste category. This stark imbalance underscores a well-established reality: caste oppression and poverty are deeply intertwined, and poverty is deeply rooted in caste.
In private sector professional jobs, SCs account for 30% while STs account for only 4%—a difference attributable to SCs having better educational access than STs.
The demographic data of castes also does not support EWS reservation in the state. Historically, Telangana is a land of Adivasi, pastoral nomadic, and service castes, and this is clearly reflected in the survey. The overall caste-wise distribution in the state shows 56.4% belonging to Backward Castes (including 10% Muslim Backward Castes), 17.4% to SCs, 10.4% to STs, and 11.9% to Other Castes (including 2.4% OC Muslims).
Of the 242 castes in Telangana, 56 have populations exceeding 50,000, and together they make up 90% of the state’s total population. When the “No Caste” and “Others” categories are included, this figure rises to 94%. Notably, OCs—including OC Muslims—constitute just 11.9% of the state’s population, yet receive 10% reservation under the EWS quota, a share that is slightly less than proportional representation.
In contrast, the BC and OBC category accounts for 56.4% of the total state population but has only 27% reservation. As has been widely discussed, the EWS reservation is not against the idea of justice enshrined in the Constitution, but it dilutes the spirit of the reservation system.
Disparities
The disparity between castes and communities also points to the caste-based dimension of poverty. Disparity emerges as a defining feature of the survey, largely shaped by the historical position of caste. One is rich because of caste, and one is poor because of caste—and this is glaring both between and within categories.
The disparities in land ownership between SCs and STs are very conspicuous, with STs owning more land than SCs. Most ST land is waste or fallow, but it is large in size. SCs stand out as the social category with the highest proportion of small landholdings (90.5%). In private sector professional jobs, SCs account for 30% while STs account for only 4%—a difference attributable to SCs having better educational access than STs. The geographical location of these social categories has also played a significant role in shaping disparities between them.
These disparities are further entrenched within social categories. Among STs, for instance, Nakkala—a small community—is the most backward, with a CBI of 112, followed by Chenchu (CBI 108), a Particularly Vulnerable Tribal Group (PVTG). The numerically major ST communities, however, cluster at similar levels: Lambadi (95), Gond (94), Koya (92), and Kolam (92).
Interestingly, Kolam—also a PVTG—ranks third in landholding among ST communities, after Koya and Gond, but is more backward across all parameters, particularly in education. Indeed, Kolam records the highest illiteracy, the highest school dropouts, the highest child labour, the highest proportion of agricultural labour, the highest number of households without toilets and electricity, and the highest proportion in the low-income group (below Rs 1 lakh per annum).
At the same time, Lambadi rank first in overall educational attainment among ST communities—whether in higher education, English-medium education, private education, or government jobs. Interestingly, the Yerukula, who rank very low on other parameters, have the highest proportion in English-medium education and the largest representation in private-sector jobs among STs. This data validates the demand for sub-grouping of STs in the state.
The survey reveals the scale of poverty among Muslims, strengthening the case for reservation based on religion.
The survey data also suggest that larger caste groups tend to experience greater levels of distress and disparity. As noted earlier, the four larger castes—Madiga, Mudiraj, Lambadi, and Yadava—who constitute 33% of the state’s population, show greater vulnerability. They rank below the state average in the CBI and account for about 40% of the extremely poor in the state, with annual incomes of less than Rs 1 lakh.
About 50% of Madigas survive on a daily wage, while among Lambadis, Mudirajs, and Yadavas the figure stands at nearly 40%. This is reflected in the state’s agricultural labour force, where 35% of Madigas and 30% of Lambadis, Mudirajs, and Yadavas depend on agricultural labour for survival. These four castes also rank lowest in educational attainment relative to the state average. However, together they own about 32% of the state’s agricultural land. After the Reddys, who rank first in land ownership, come Yadavas, Lambadis, Mudirajs, and Madigas—suggesting the level of disparity within each caste. Although a considerable number practise their own agriculture, a large share survives on daily wages.
The survey reveals the scale of poverty among Muslims, strengthening the case for reservation based on religion. Muslims constitute 12.56% of the state’s population, of whom 10.8% fall under the Backward Classes (Group E) category and 2.48% are classified as OCs. Among the 14 Muslim BC castes, 12 rank below the state average in backwardness. Their presence in land ownership is negligible—only Shaik Muslims, who make up 7.9% of the population, own about 2% of the land. Their educational attainment is also significantly lower than the state average. Though it is difficult to explain the reasons, they rank lowest in accessing government welfare schemes.
“No-Caste” Category?
One of the survey’s surprising findings is the emergence of a “No-Caste” category, which accounts for 3.4%—nearly 1.2 million people—of the state’s total population. When combined with those listed under the “Other” category, this figure rises to 3.9%. The survey included columns for No-Caste, No-Category, and No-Religion.
Who are the No-Caste people? They are predominantly urban, educated employees and professionals, a large proportion of them from the Greater Hyderabad Municipal Corporation and its surroundings. The “Other” category comprises smaller castes not included in the state’s official list, mostly migrant labour.
Interestingly, 43% of those who identified as No-Caste possessed a caste certificate, and 13.5% had availed themselves of reservation benefits—suggesting that nearly half of them actually belong to reserved categories. The No-Caste category accounts for 7.7% of government jobs and 13.3% of private jobs, compared to state averages of 2.8% and 7.4% respectively. They also rank least backward across all parameters, and represent, in effect, the state’s super-rich.
The resistance to a nationwide caste census stems largely from the apprehension among dominant castes that such an exercise would reveal the actual social structure—who holds numerical strength, property, power, and privilege across castes and communities.
How do we understand this tendency towards caste neutrality in an era of identity politics? It is a byproduct of educational and economic advancement, and of individual anxiety to escape from immediate identity. As noted in the survey report, this group likely comes from households that have made significant investments in education—possibly benefiting from historic OC privileges, or rising through BC, SC, or ST categories via intergenerational upward mobility.
In such contexts, adopting a caste-neutral identity may reflect both a desire to distance oneself from caste-based disadvantage and a strategic embrace of the meritocratic ideals promoted by elite state institutions. This tendency aligns with the idea of meritocracy, which advocates for reservations based on economic criteria rather than caste.
Nevertheless, the survey underscores that caste remains the principal determinant of backwardness—and this is the paradox of contemporary Indian society. It is this paradox that explains why the EWS quota was widely accepted by Indians, including those from quota communities, without significant resistance. The fundamental question remains: how do dominant-caste rich communities end up with reservation benefits nearly proportional to their population share, while OBCs—who form more than half the population—are confined to just 27%?
Importance of Caste Census
The resistance to a nationwide caste census stems largely from the apprehension among dominant castes that such an exercise would reveal the actual social structure—who holds numerical strength, property, power, and privilege across castes and communities. The findings of this survey make these disparities visible, highlighting varying degrees of backwardness among different groups.
For subaltern castes, however, a caste census is not perceived as a threat to national unity or social cohesion—it is seen as a means of empowerment and resilience. In this context, caste is no longer viewed solely as a marker of discrimination but increasingly as a strategic resource: an instrument for accessing political power and advancing cultural and economic mobility, thereby shaping participation in modern democratic processes and development.
The Telangana caste survey tells us that a caste census is not an untouchable idea for the nation —rather, it would socialise democracy and the nation.
This is the first of two articles analysing the findings of the Telangana Caste Census of 2024. The next article which will offer an extended analysis will be published the week beginning May 18—Editor.
Bhangya Bhukya is a professor of history at the Hyderabad Central University.
