3 June 2019

Look Beyond Quotas for Equality

BY PRANAY KOTASTHANE AND NITIN PAI

Affirmative action aimed at ending discrimination has a long and complex history in India. A new chapter was added to this story on May 10 when the Supreme Court upheld a Karnataka law, saying quotas for promotion of scheduled caste and scheduled tribe candidates in public employment was constitutional and did not require demonstrating ‘backwardness’ of the community.

In January, the Union cabinet approved a Constitution Amendment Bill to provide 10% reservation to economically backward sections. And a month later, the Karnataka government amended rules to give priority to Kannadigas in C&D group jobs in private companies.

In short, even groups opposed to quotas want the same benefit extended to them. For long, this has been the only solution to address inequity in India. So these recent developments provide a good opportunity to reflect on the question: can we imagine better ways to achieve social equity goals?


The starting point for this conversation is that though some form of affirmative action is a necessity in our deeply iniquitous society but does that solution always have to be quotas? This might seem almost heretical, given how normalised reservation has become but there can be better options.

Consider this thought experiment. There are no predetermined quotas for any posts. Positions are filled only based on a composite score of all applicants. The composite score is a combination of two measures. The first is an inequity score — calculated to compensate for the relative disadvantage faced by an applicant.

The second measure strictly represents an applicant’s ability to be effective for the position they are applying for. Selection is on the basis of the composite score. No seats are reserved and yet the score allows for addressing multidimensional inequity much better than current methods.

How does this work in real life?

The inequity score can be used to indicate relative disadvantage along several dimensions: individual, social and geographic. Different factors can be assigned different weightages. For instance, given the salience of caste in the Indian social context, the greater the disadvantage a community faces, the higher the weightage.

In addition, we can incorporate other parameters into the inequity score — parents’ level of education, income levels, rural upbringing, or even childhood nutritional deficiencies. Currently, our system of quota-based allocations does not account for non-caste disadvantages that have a disproportionate impact on life outcomes.

A national commission for equity can be formed to propose and review parameters and their weightages within a cooperative federal framework. It doesn’t have to be one-size-fits-all solution. States can assign their own factors and weightages according to the local conditions.

The second measure — an effectiveness score — can then be kept completely independent of equity considerations. It can take the form of a test, an interview or any other indicator to assess candidates’ ability to perform the job they have applied for. Information about the inequity scores can be masked from evaluators of the effectiveness score.

By filling positions based on a sum of the two scores, it becomes possible to be more comprehensive in addressing social inequities while also creating stronger incentives for an individual pursuit of excellence.

What makes this better than the status quo?

Quotas bring with them the charge of unfair equality and this leads to further acrimony between groups. Different cutoffs for different quota categories further strengthens the charge of unfairness. This is a dangerous outcome, as other groups then seek to create new categories for themselves. In fact, a 2017 study published in Nature Human Behaviour claims that people actually prefer unequal societies over unfair ones.

The abstract of the paper reads: drawing upon laboratory studies, cross-cultural research, and experiments with babies and young children, we argue that humans naturally favour fair distributions, not equal ones, and that when fairness and equality clash, people prefer fair inequality over unfair equality (Starmans et al).

The proposed alternative aligns well with this human preference of fairness over equality.

This method can also accelerate dissolution of caste boundaries better than the current system. There is little evidence to show that more reservations have reduced the relevance of caste to our society. In fact, it has only resulted in more individuals identifying themselves with their castes.

Inter-caste marriages, which can break caste identity perpetuation, continue to be abysmally low. If an inequity score can take into account factors other than caste, its salience to our politics is likely to relatively decline. To be sure, caste-based political mobilisation to secure a better inequity score will still be possible.

Satish Deshpande and Yogendra Yadav in 2006 proposed a similar model as an alternative to reservations in higher education. They write that an evidenced-based model addressing multiple sources of group and individual disadvantages helps to de-essentialise identity markers such as caste or religion; that is, it provides a rational explanation why specific castes or communities are entitled to compensatory discrimination and undermines attitudes that treat such entitlements as a “birth right”.

Because the inequity score takes into account multiple parameters, it is a much better indicator of relative disadvantages faced by individuals. Caste, language and religion are crude measures that required the innovation of concepts such as “creamy layer” to assuage concerns of unfairness.

These crude measures were perhaps the only available indicators seven decades ago, in the early years of the republic. But today, it is at least theoretically possible to construct better measures that seek to provide equal opportunity at individual level and for annihilating caste at the societal level.

Pranay Kotasthane and Nitin Pai are with the Takshashila Institution

No comments: