By Seema Chishti
September 21, 2018
As rising aspirations of India’s demographic dividend shape social and political discourse, comes a sobering new study: Looking at education and income, there is little inter-generational mobility (upward mobility from parent to child) in India. Muslims are the biggest losers in the intergenerational mobility index. Scheduled Castes and Scheduled Tribes have done better on the index while upper castes and OBCs have remained where they were.
Authors of the study, based on 5,600 rural sub-districts and 2,300 cities and towns, are Sam Asher, of the World Bank; Paul Novosad of Dartmouth College and Charlie Rafkin of the Massachusetts Institute of Technology (MIT).
Their study, which looks at social groups and intergenerational mobility concludes that, overall, there is little inter-generational mobility and it has not changed even after economic liberalisation. It, however, concludes that some mobility made possible for Dalits and Scheduled Tribes has been offset by severely declining mobility for Muslims.
The study also says that mobility levels for African Americans in US are better than those for Muslims in India but the movement of Dalits and Scheduled Tribes is comparable to that of African-Americans. It finds that South India is overall more inclusive, as is urban India and that education acts as a booster of prospects: “On average, children are most successful at exiting the bottom of the distribution in places that are southern, urban, or have higher average education levels.”
Inter-generational mobility is about change of status across generations, and change of this economic status is a useful measure for describing “changes in access to opportunity over the long run”. But the study also recognises the perils of looking at just economic data or educational indicators as proxies of status mobility.
It focuses on “the expected outcome of a child born into the bottom half of the parent outcome distribution (upward interval mobility, referred to as upward mobility); and the expected outcome of a child born into the top half of the parent distribution (downward interval mobility).”
The paper, out this month, says as a summary: “From before liberalization to the present, we find that intergenerational mobility for India as a whole has remained constant on average, but with considerably cross-group heterogeneity. While our results reinforce earlier findings of rising opportunity among Scheduled Castes and Scheduled Tribes that have been noted by other authors, we find that these changes are almost exactly offset in the population as a whole by declining upward mobility among Muslims.”
Co-author Paul Novostad says in a detailed explanation that they chose “a measure of intergenerational mobility; if you are born to a parent in the bottom half of the education distribution, what is your expected education rank?”
Novostad and his co-authors assert that they have done their best to use administrative data as carefully to compensate from tax and other income data in developing countries which not being very reliable. They say that “while India’s growth has made almost everyone a lot better off, it has not changed the rate of churn at all, if you started at the bottom, you’re just as likely to finish at the bottom as you would have been in 1950.”
In between social groups though, the paper concludes that while Scheduled Castes and tribes are more mobile now, prospects for combined forward castes and OBCs have remained the same.
On Muslims, defined as the “least upwardly mobile group in India”, the paper makes the damning claim that the situation is worse than that of African Americans. In the US, while those born at the bottom half of an education distribution on average tends to get to the 34th percentile, Muslims can expect to only get to the 28th, something they term as “really low.”
The paper also makes a case for geography being important, with urban areas being better even for those who are at the bottom of the pyramid: “We also find that urban areas are significantly more mobile than rural areas — for example, the mobility gap between urban and rural locations is about equal to today’s gap between higher caste Hindus and SCs. Using granular geographic data, we find that village assets like roads and schools are associated with more upward interval mobility; on the other hand, SC/ST segregation is associated with lower levels of upward interval mobility.”