Identity and public policy
with Rohith Unnikrishnan
Some political parties have demanded that caste be enumerated in the Census. Actually, the demand amounts to one of counting the Other Backward Classes, for the Scheduled Castes (SCs) and Scheduled Tribes are already counted. The demand has been accompanied by the argument that the efficacy of public policy for enhancing well-being across the population is tied to the enumeration of Indians by their caste. One approach to assessing this argument would be to compare development outcomes in States where political parties have adopted caste-based mobilisation with those in States where political programmes for ending deprivation have taken the social democratic route, without resorting to identity politics. Tamil Nadu would be an example of the former, while Kerala would be an example of the latter. A comparison of the developmental experience of these States would therefore be instructive.
Findings along three variables
As the availability of data across social groups is limited, we focused on three variables. These were adult literacy, infant mortality and consumption. Each of these indicators is related to one of the three components of the United Nations Development Progamme’s Human Development Index. Having chosen the development indicators of interest, there are two ways in which we may assess the difference made to the condition of the least well-off in a population, in this case the SCs, by differently driven social programmes. For any indicator, we may focus on either the impact made in terms of its distribution among groups or the absolute level achieved by a cohort.
In a comparison of Tamil Nadu and Kerala, for consumption — a proxy for income — the gap between the general population and the SCs is greater in Kerala than it is in Tamil Nadu but smaller when it comes to the other two indicators. However, when it comes to the absolute level achieved, the SCs of Kerala are better-off than the SCs of Tamil Nadu on all the three indicators. Interestingly, they are also better off than the general population of India, i.e., they have superior consumption, literacy and infant mortality outcomes. This is striking. At the same time, the exercise also revealed a hazard when focusing on relative standing alone. We found that for more than one indicator, the distance between the SCs and the general population is far lower for the country as a whole than it is in at least one of the two States considered, even though the State concerned registered a superior level for the same indicator. This leaves us veering towards the maximin principle in evaluation, according to which that policy is preferred which maximises the position of the worst off in a society. Now, Kerala will be chosen as better performing, for the most disadvantaged have higher indicators there. Though we could do with more analysis and the use of controls to arrive at a definite conclusion, this evidence at least suggests that identity-based public policy may not be as effective as one based on an identity-less or universalist approach that is the hallmark of a social democracy.
Though we are already in a position to say something about the potential of information on the caste status of individuals in the elimination of deprivation, we take our investigation to another field before concluding. It has been known for decades that gender inequality exists in India. Knowledge of low literacy and high infant mortality among females has, however, done little to spur counteracting public policy that will ensure women’s empowerment. Returning to the two States of our study, Kerala greatly disappoints when it comes to women’s empowerment, and lags behind Tamil Nadu on labour force participation, the proportion of female legislators and judges, and crimes against women. Counting the number of women through a census has proved insufficient to eliminate the deprivation and inequality they face. Politics and not the availability of information drives public policy.