Reservations and sub-classification of SC/STs: A mixed bag for substantive equality

On 1 August 2024, in State of Punjab v Davinder Singh, a seven-judge bench of the Supreme Court upheld the constitutional validity of sub-classification in the Scheduled Caste and Scheduled Tribe categories. In doing so, the majority overruled the five-judge bench decision in E.V. Chinnaiah v State of Andhra Pradesh. 

I am not going to summarise the reasoning adopted in the six opinions (which includes Justice Bela Trivedi’s dissent). If you’re looking for a summary, an excellent one can be found here. In this article, I will reflect on a broader doctrinal issue contained in the opinion and how the Court proposed states go about the process of sub-classification. 

The opinion by Chief Justice D.Y. Chandrachud (on behalf of himself and Justice Manoj Misra with all judges in the majority expressing agreement) wades into tricky waters on reservation jurisprudence. For most part, it provides much-needed clarity. One of the enduring legacies of this opinion will be its take on the relationship between reservations in Article 16 and ‘efficiency’ in Article 335. 

This article also engages with specific aspects of sub-classification in the judgement. The basis and the evidence required for sub-classification seem particularly onerous. They raise concerns over how categorisation would work in practice and whether sub-classification based on this approach would even survive judicial review. 

Further, I also reflect on the impact of creamy layer exclusion in SC/ST reservation and its consequences from the perspective of ‘presence’ in public employment. Finally, I look at how the approaches prescribed by the Court seem to have assumed that a group-based approach is the only method to distribute quotas.

Efficiency and Article 16 reservations

The majority in Davinder Singh has undone the unreasonable reading of Article 335 in M. Nagaraj v Union of India (2006). Article 335 provides for the claims of SC/STs in public employment to be taken into consideration “consistently with the maintenance of efficiency of administration.” In Nagaraj, while concerned broadly with reservations in promotions for SC/STs through the insertion of Articles 16(4-A) and 16(4-B), the Supreme Court ruled that the impugned articles were not violative of the basic structure since they did not do away with, inter alia, the requirement to maintain overall administrative efficiency. 

It went on to hold that, in order for specific reservation provisions to be constitutional, the state will have to show that the provisions do not hamper the efficiency of administration. By doing so, the Court had elevated the efficiency concern in Article 335 to serve as a check on the Constitution-amending powers of Parliament. 

Given the express reference to ‘efficiency’ in the text of Article 335, the position in Nagaraj has been difficult to dislodge. But Chief Justice Chandrachud’s opinion in Davinder Singh reaches the explicit conclusion that the concern of efficiency in Article 335 is not a check on the power to provide reservations under Article 16. His interpretation provides a meaningful way out of this conundrum that is in consonance with principles of substantive equality. 

Noting that since Article 335 refers only to SC/STs, he holds that ‘efficiency’ as a general limitation on Article 16 reservations would defy constitutional logic. Chief Justice Chandrachud’s opinion clarifies that ‘efficiency’ could mean that the requirement of a qualifying mark cannot be done away with completely, but nothing more. According to the CJI, the proviso to Article 335 establishes the view that relaxation in qualifying marks does not mean a reduction in efficiency. The opinion provides the much needed restatement that Article 335 was meant to be an affirmative provision which acknowledged the claims of SC/STs in public employment rather than work as a check on the power to provide reservation. 

Basis for sub-classification

All the judges in the majority agree with Justice Chandrachud’s opinion on the evidence required for sub-classification measures to survive judicial review. Davinder Singh affirms Jarnail Singh v Lachhmi Narain Gupta’s (2018) overruling of Nagaraj on the point that evidence of backwardness is not required while providing for reservations for SC/STs as whole. However, if a state wants to sub-classify SC/STs, it will have to provide ‘quantifiable data’ that indicates a substantial difference in social backwardness between the proposed sub-groups and the rest. 

Further, the state will also have to produce data to establish that there is inadequate representation using the metric of ‘effective representation’ within the different grades of state services. Government jobs are grouped under different grades depending on qualification requirements, service conditions, remuneration and perks, etc. For instance, Grade-D services would include gardeners, security guards and drivers; Grade-C might include constables, stenographers and clerks. 

According to the majority in Davinder Singh, the state can’t go as wide as public employment as a whole; neither is it required to go as narrow as the ‘cadre’, which refers to the specific kind of job, say that of a peon (as required by Nagaraj). 

As per Davinder Singh, the inadequacy requirement is to be established in the broader sub-group of services within which cadre falls. Therefore, if the cadre of peons fell within Grade-C, the inadequacy has to be demonstrated within Grade-C services of the state. The majority acknowledges the phenomenon that reservations tend to fill out the positions in the lower grade services of the state but not the higher ones. 

This move away from the cadre-centric approach of Nagaraj to assess inadequacy is a welcome one. To demonstrate cadre-based inadequacy sat oddly with the aim of ensuring the appropriate share in public employment across levels. The cadre was too narrow and demanding a metric for the state to produce data. 

More generally, the centrality and intensity of data-dependent judicial review of reservations has received inadequate attention. The requirements and standards of this approach remain unclear in reservation jurisprudence while seeking to appear objective. Further, reservations seem to be the only site of judicial review where these exacting yet unclear standards of data requirements seem to be in play.

Creamy layer 

Perhaps the most troubling aspect of the majority judgement is Justice B.R. Gavai’s observation further entrenching the position that the creamy layer needs to be excluded even within SC/ST reservations. The five-judge bench positions in Nagaraj and Jarnail Singh find strong approval in Justice Gavai’s opinion, which also has the agreement of Justices Vikram Nath, Pankaj Mithal, and S.C. Sharma. 

With the combined opinion of CJI Chandrachud and Justice Manoj Misra remaining silent on the issue (and the overall dissent of Justice Bela Trivedi), the necessity of creamy layer exclusion in SC/ST reservations is established by a 4:3 majority. However, it is evident that the issue of creamy layer was not before the Court and must be read as obiter dicta. 

The creamy layer exclusion of SC/STs is problematic as a matter of social fact. It misunderstands both the nature of discrimination faced by the SC/STs and also what reservations can do towards alleviating that discrimination. 

The social discrimination which is at the core of the lived reality of members of Scheduled Castes and Scheduled Tribes cannot be done away with only through educational attainment and public employment. One only needs to imagine a wealthy homosexual person or wealthy woman to understand that deeply entrenched aspects of social discrimination remain unresolved by economic mobility. 

In the wake of the judgement, we are already seeing some resistance to such exclusion from the political class—it might be that democratic pressures on the ground will correct what the Court seems to have got wrong. But for now, it looks like mandatory creamy layer exclusion of SC/STs is here to stay as a part of constitutional doctrine.

Group-based distribution of quotas within SC/STs

There is a fairly detailed discussion in the judgement that indicates constitutionally acceptable methods to sub-classify. I do not get into the details here, but I’m interested in the fact that all options considered by the Court adopt a group-based distribution of quotas within SC/STs. I believe that the Court missed an important opportunity to consider whether a group-based approach is the only way to approach distribution. 

The constitutive groups of SC/STs, while having certain connecting threads of discrimination and marginalisation, have differing levels of social backwardness among themselves. The same problem is likely to afflict any sub-groups created by sub-classification. While the SC/STs are identified as a group for reservations, it must be possible for us to imagine an individual-based approach for distribution of quotas. Individuals within the SC/ST group suffer marginalisation and discrimination along common and different axes. 

Different models have been suggested for assessing marginalisation and deprivation of individuals. The deprivation points system in the JNU Admission Policy (page 5) and Anand Teltumbde’s model are examples of concrete ways to think about this. These allow for factoring intersectional claims while deciding the priority of accessing reservations. Factors like gender, economic status, the family’s education and employment status, the applicant’s own educational background, including the district where such education took place, can be utilised in developing a priority scale. 

It might be far more effective to undertake an exercise that allows for an overall assessment of an individual’s marginalisation and prioritises the access to quotas along those lines. To illustrate, a poor Dalit woman is likely to be worse off than a rich Dalit man belonging to the same caste. While both the candidates would be eligible for reservations, a poor Dalit woman suffering across three axes of marginalisation—class, caste, and gender—should be given priority over the rich Dalit man. 

This would address the concern that it is the relatively better-off within the group who are best-placed to repeatedly benefit from reservations. At the same time, it would also prevent the complete exclusion of the relatively better-off as is being attempted through creamy layer exclusion. It would only mean that the relatively better-off have a lower priority in accessing reservations. 

Concluding remarks

One of the aspects of the judgement that requires closer scrutiny is CJI Chandrachud’s attempt to reconcile the tensions in the Supreme Court’s jurisprudence on the relationship between reservations and merit. The Chief explicitly acknowledges the tension and it is evident that he is aware that the positions taken by the nine-judge bench in Indra Sawhney v Union of India (1992) pose a significant challenge. 

On the one hand, the majority in Indra Sawhney accepts and establishes the position from State of Kerala v NM Thomas (1975) that reservations are a facet of equality and not an exception. In that context, the majority does not view reservation as being antithetical to merit. However, when it comes to reservation in promotion or certain kinds of specialised jobs, the majority in Indra Sawhney does not allow for reservations. 

To resolve this dichotomy, the Chief Justice’s opinion relies on parliamentary action that was taken to overcome the prohibition on reservation in promotions in Indra Sawhney. In that sense, the resolution of this inconsistent approach in the Court’s discourse is resolved only by constitutional amendments. He calls the amendments an “emphatic repudiation of the binary of reservation and merit.” 

But the Chief’s attempt, captured in paragraphs 72 to 75 of the opinion, is far less convincing than his reasoning on efficiency. Admittedly, for the seven-judge bench in Davinder Singh, the stare decisis consequences of the doctrinal tensions of the nine-judge bench decision in Indra Sawhney could seem a bridge too far. There is a need for a more detailed assessment of this attempt to reconcile the Court’s stance in Indra Sawhney with an endorsement of law-making to overcome it. 

While there are important doctrinal clarifications in the judgement, a majority of judges finding in favour of creamy layer exclusion casts a dark shadow on the project of substantive equality. Unless the Court starts resolving foundational questions on what reservations are meant to do in different spheres for different beneficiary groups, it will remain an area in search of doctrinal coherence.

Anup Surendranath is Professor of Law at National Law University, Delhi. Research assistance by Pulkit Goyal; V-Year B.A., LL.B (Hons), NLU Delhi.