The proposition that India’s post-liberalisation economic growth has loosened the bond between caste and economic outcomes — allowing merit, education, and market participation to displace ascribed social identity as the primary determinant of occupational and income trajectory — remains one of the most contested claims in contemporary Indian social science. Proponents of this view point to the visible emergence of successful entrepreneurs, professionals, and corporate executives from Scheduled Caste and Other Backward Class backgrounds as evidence that caste ceilings are yielding to economic mobility. Critics counter that these exemplary cases represent the thin upper tail of a distribution that remains fundamentally shaped by inherited social capital, caste-based network effects in job referral and marriage markets, and the persistent segregation of low-caste communities into low-productivity occupational niches.This paper takes up the debate empirically using two complementary nationally representative datasets: the Periodic Labour Force Survey 2022-23 microdata, which provides detailed household-level employment and income data for the contemporaneous cross-section; and the India Human Development Survey-II (IHDS-II) panel, which links adult children’s occupations to their parents’ occupations and thus permits direct measurement of intergenerational occupational mobility. The combination of cross-sectional inequality measurement through Lorenz curves and Gini decomposition with intergenerational mobility analysis through transition matrices and relative mobility indices provides a more complete picture of the caste-income nexus than either dataset could yield alone.The geographic focus on urban Tamil Nadu and Andhra Pradesh is deliberate. These states have implemented the most comprehensive reservation and targeted development programmes in India — Tamil Nadu’s 69 percent reservation in state educational institutions and government employment is the highest in the country — and constitute natural test cases for the effectiveness of affirmative action in reducing caste-based income inequality and promoting upward mobility over an intergenerational timescale. The urban focus is motivated by the expectation that urban labour markets, with greater economic complexity and reduced social surveillance compared to rural settings, should exhibit the most favourable conditions for caste-independent mobility; findings of persistent caste stratification in this context therefore represent a conservative estimate of caste’s economic effects nationally.This study contributes to a gap in the quantitative sociology of caste in South India by providing state-specific intergenerational transition matrices disaggregated by caste group, a Lorenz-curve-based income inequality decomposition comparing within-caste and between-caste components, and a regression-based decomposition of the earnings gap attributable to observable human capital differences versus residual caste effects. The findings have direct implications for the design of affirmative action policy in the post-EWS reservation era, where the constitutional basis and effectiveness of caste-based reservation is subject to renewed political and judicial scrutiny.