Caste in confusion

The real issue is not just about whether reservations for Dalits and Other Backward Castes (OBCs) in private sector jobs will hurt the competitiveness of the economy as India Inc fears. Neither is it about whether it is fair for the government to cast(e) its responsibility on to the shoulders of someone else. Nor is it about whether the Prime Minister (PM) was exhorting industry to do his job, or whether he was just giving them the velvet glove treatment since other ministers in his government have very clearly been talking of legislation to this effect.

The real issue is that the entire debate is taking place in a vacuum. No one, including the PM, has any idea of the scope of what he’s talking of. Immediately after the PM’s CII speech, Hindustan Lever gleefully pointed out that 55 per cent of its staffers were from the reserved categories (19 per cent were SC/ST and the rest OBCs)—trust a company of Brit origin to keep such data! Other companies, however, don’t keep such data, which is the way it should be.

But were the companies to ask the PM how much of their workforce should they reserve, Manmohan Singh would be hard-pressed to give an answer. The Mandal Commission report says 52 per cent of the population is OBC. The Census does not report data by caste, but the National Sample Survey does, and in 1999, it said 32 per cent of the population was OBC! Since that’s a difference of around 200 million people we’re talking of—the total work force is around 400 million—the PM clearly needs to do some homework himself before he passes on the buck to someone else, either gently or threateningly.

While the PM’s looking at the numbers, another interesting set of numbers that the NSS throws up is that of the share of different social groups in both employment and unemployment—since the two are mirror images of one another, the results are naturally similar. In 1999, the NSS says, 31 per cent of the total employed were SC/ST, 10 per cent were Muslim and 33 per cent were OBC. Juxtapose this with the shares of these groups in the country’s population and, surprise, you find the shares are roughly in keeping with what the PM’s asking for!

SC/STs comprise 28-29 per cent of the population and their share in the workforce is higher at 31 per cent. In the case of OBCs, the employment share and population share are nearly equal. The only group for whom the share in population is greater than the share in employment (12-13 per cent versus 10 per cent) is the Muslims. The PM, though, is not asking for reservation for Muslims and, needless to say, the BJP is not asking for this, either.

So, is the case for reservations/affirmative action in jobs over, except for Muslims? While the broad NSS data would suggest this is the case, it’s possible to argue, and so it will be argued, that the SC/STs and OBCs are employed in low-paying jobs while upper-caste Hindus are cornering the high-paying jobs. That will require slicing the employment-by-social-group data and examining wage levels. Surjit Bhalla, who’s the only person I know to have studied the NSS data in this regard, is currently doing that and presumably, his analysis should be out soon.

Bhalla’s analysis of education using NSS data, in Business Standard on April 15, shows the same result of Muslims being the worst off. It also shows what the Dalits have been arguing for a long time, that they are worse off compared to OBCs. The SC/ST share of those enrolled in colleges across the country is 15.1 per cent as compared to their population share of 28.3 per cent (that’s a ratio of about 53 per cent); for OBCs, it is 23.3 per cent versus 32 per cent (72 per cent); and for Muslims it is 5.8 per cent versus 12.2 per cent (around 48 per cent). All of which would suggest the only case for reservation is in education. And since the shares of these groups in college enrolment are roughly similar to their shares in high school, the real case is for reservations in schools, not colleges.

The ultimate in discussion in a vacuum, of course, is that we don’t even know what’s happening in the jobs market. The National Rural Employment Guarantee Scheme, and much of the UPA’s propaganda, for instance, are predicated on employment growth having fallen. Yet, the UPA’s own Mid-Term Appraisal points out that between 1994 and 2000, employment rose by 1.02 per cent per annum as compared to labour force growth of 1.08 per cent; between 2000 and 2003, while the labour force grew 2.85 per cent, employment grew by 2.93 per cent!

While purists will argue this is based on “thin” samples by the NSS, the problem is that data from NSS’ “thick” sample are still not out though the survey was completed in June last year—results of the last NSS “thick” round, in 1999, were out within a few months of the survey being completed. It would be interesting if the results showed the two big UPA policies (employment guarantee and job reservations) were started when the data showed there was no need for them!