This article analyses the underrepresentation of Muslims in the public and private sector, state-wise and over more than 10 years. We use the 66th and 68th rounds of the National Sample Survey, called ‘Employment and Unemployment,’ whose data was collected from July 2009 to June 2010 and July 2011 to June 2012. We also use the Periodic Labour Force Surveys (PLFS) of July 2018-June 2019 and July 2019-2020. In each case, the samples are very large — in 2009-10, 122,359 people, in 2011-12, 125,931, in 2018-19, 99,988 and in 2019-20, 100,991.
These surveys examine the composition of the public and private sectors, religion-wise. For the private sector, the category “proprietors” refers to the fact that the persons enumerated are not part of the salaried milieu but “self-employed” (a formula we keep) because if few of them are CEOs, most of them do work that pays less — shopkeepers and artisans for instance.
To make the interpretation more robust, some weightage has been applied. Weight is basically a survey design variable that says approximately how many households in the population a surveyed household represents. The weightage associated with one household defines the number of households it represents in the population.
The first two series, the 2009-10 and 2011-12 surveys, have only one public sector variable, while the 2018-19 and 2019-20 surveys account for the bureaucracy and state-owned enterprises. For the sake of comparison between the first two rounds and the later ones, we have bracketed together “bureaucracy” and “state-owned enterprises” and compared these data with the “public sector” category.
There are significant variations between the first two rounds and the last two, even if we see more variations in frequency counts graphs than in weighted graphs — this is because frequency counts represent sample population and weighted data represent total populations. But to minimise the variations, we have calculated average figures for the first two rounds on the one hand and for the second two rounds on the other. As a result, two datasets have been generated by combining four data sets.
The results are telling. While Muslims form 14.2 per cent of the Indian population, their proportion of the public sector employment has stagnated at below 7 per cent (from 6.75 to 6.87 per cent) between 2009-12 and 2018-2020. In contrast, Hindus who constitute 80.2 per cent of India’s population comprise about 86 per cent of the public sector. On the other hand, Muslims are over-represented among the self-employed — 16.5 per cent in 2009-12 and 15.5 per cent in 2018-20 per cent, an erosion we see in most states.
In India’s largest state, Uttar Pradesh, where Muslims represent 19.26 per cent of the population, their percentage in the public sector had gone up from 5 per cent in 2010 to 11.5 per cent in 2012, when the Samajwadi Party was in office. But it has dropped to 7 per cent in 2019 and 6.5 per cent in 2020. Muslims are over-represented among the self-employed, simply because they have no other choice but to be on their own. Between 2010 and 2020, their percentage among the “self-employed” remained around 24 per cent. This pattern repeats in Madhya Pradesh as well, where Muslims (6.6 per cent of the population) represented 3 per cent of the public sector’s employees in 2009-12 and 4.5 per cent in 2018-20. They represented 10.7 per cent of the self-employed in 2009-12 and 11.7 of this category in 2018-20.
In Rajasthan, Muslims, about 9 per cent of the population as per the 2011 Census, constituted 4.1 to 4.3 per cent of the employees in the public sector. They were over-represented among the self-employed despite some erosion in the later years — 13 per cent in 2009-12 to 10.2 per cent in 2018-20. In Delhi, the trends are the same: Muslims, 12.9 per cent of the population, represented only 4 to 5 per cent of the government employees but 14.5 per cent of the self-employed in 2009-2012 and 13.41 per cent of this category in 2018-20. They are over-represented among the self-employed but experienced some erosion. In Maharashtra too, the percentage of Muslims in the public sector, 4.8 per cent in 2009-12 and 5.2 in 2018-20 was much below their share of the state’s population —11.5 per cent. They remained over-represented among the self-employed –16.9 per cent in 2009-12 and 16.4 per cent in 2018-20. Similar trends are found in Karnataka, where Muslims constitute about 12.9 per cent of the state but only 6.2 per cent of public sector employees in 2009/12 and 5.2 per cent of the public sector in 2009-12/2018-20. They were also over-represented among the self-employed, despite some erosion — 20.3 per cent in 2009-12 and 19.1 per cent in 2018-20.
In some states, the erosion of the share of Muslims among the self-employed is even more pronounced. As a result, they are under-represented not only in the public sector but also among the self-employed. In Gujarat, where Muslims are 10 per cent of the society, their share in public sector employment has come down from 7 per cent to a minuscule 1.5 per cent. Amongst entrepreneurs, their share has dropped from 12.5 per cent in 2010 to 9 per cent in 2020.
In Assam, where Muslims account for 34.2 per cent of the population as per the 2011 Census, they have always been significantly under-represented — 17.4 employees in the public sector and 19.32 per cent in 2018-20. They were around 30 per cent of the self-employed in the period under review.
The declining representation of Muslims, not only in the public sector but also among the “self-employed” suggests that they are now increasingly overrepresented among the unemployed. The share of Muslims deemed as jobless in the NSS surveys under review jumped from 2.62 per cent in 2009-10 to 7.16 per cent in 2018-19.
Jaffrelot is senior research fellow at CERI-Sciences Po/CNRS, Paris, professor of Indian Politics and Sociology at King’s India Institute, London, and non-resident scholar at the Carnegie Endowment for International Peace, Saini is a Data Analyst & Researcher on Indian politics