A critique of Pavithra Suryanarayan‘s paper on ‘Why the poor vote for right-wing?’

By Ambika Pandey

This article is a response to Pavithra Suryanarayan’s article in The Print titled ‘Why do the poor in India vote for Right-wing parties like the BJP?’ dated 18 August 2018 based on her study titled ‘When do the poor vote for the right-wing and why: Hierarchy and vote choice in the Indian states’. Pavithra Suryanarayan is an Assistant Professor of International Political Economy at John Hopkins University who argues that the reservation policies recommended by the Mandal Commission resulted in a surge of high-caste Brahmin votes for the right wing. Her paper finds that “places with greater Brahman Dominance in education in 1931 were associated with a larger increase in right-wing vote share in state elections held after 1990, a relationship that did not exist in elections held prior to the threat of affirmative action.” The chief claims presented by Ms. Suryanarayan, the sources used and/or omitted, and the methodology utilized are analyzed as a part of this paper.

The Right vs The Left

To begin with, perhaps the most striking fallacy in the paper is its inflexible definition of the right wing. The author writes “I categorise BJP as right-wing because during the national and state elections of 1990–1995, the party explicitly appealed to the interests of wealthy business, landowning, and professional classes in India.” She further elaborates on her statement, stating “by the 1990s its (BJP’s) party activists were more likely than any major national party to favour trade and investment liberalisation, private-sector expansion, and to oppose state intervention in the economy.”

However, in a country like ours, with 1841 political parties that switch between “left wing” and “right wing” policies depending on their position in the larger political picture, a comparison of the right versus the left is redundant. A party which supports a policy when in government, may oppose it when in the opposition.

To understand this pattern more succinctly with reference to the author’s claims, let us take the example of economic policy. In the history of independent India, the Congress party has been the chief proponent of privatisation. In 1991, P.V. Narsimha Rao, then Prime Minister, and Finance Minister Manmohan Singh became the pioneers of liberalisation in India. But the Congress’s ideology is socialism, at least in theory, and therefore the party is excluded from the author’s categorisation of the right wing. Following the same rationale, the author has labelled the right wing as anti-poor, concluding that “the poor’s support of Right-wing parties is puzzling as they are choosing parties that work against their economic interests.” Although Atal Bihari Vajpayee carried forward and expanded upon Narsimha Rao’s liberalisation policies, his government also championed the pro-poor cause. The Pradhan Mantri Gram Sadak Yojna set an example for future schemes with its progressive approach towards diminishing the rural-urban divide. Roads linking a village to a city opened a plethora of new opportunities for the rural Indian — a farmer could now transform his livelihood into a business and a daily wage worker could access a metropolitan city for better resources. Another step towards alleviating poverty was the implementation of the Sarva Shiksha Abhyan in 2001 — it ensured free and compulsory education for every child within the age group of 6–14. The result was a surge in the Gross Enrolment Ratio from 81.6 in 2000–2001 to 94.9 in 2004–2005 at the primary school level.[1] Therefore, welfare schemes and a pro-poor agenda can be attributed to both the BJP and the Congress. So is there an explicit distinction between the economic policies of the “right”, “centre”, or the “left” in India? The answer is no.

The author has extended her rigid division of ideology to voters as well. But similar to the politicians, the voters show no fixed pattern of allegiance to the right or the left. It will be safe to estimate that many sections of the population, if not the majority, are socially conservative. Masses of the rural areas possibly fit better into this assumption. Voters with such a background are no advocates of postmodern liberalism, yet they support pro-poor schemes such as land distribution for farmers and lower rent. Hence on the one hand, their conservatism makes them inclined to the right, yet the economic reforms they support indicate a preference for pro-poor policies. Therefore, different situations call for the different ideological preferences of the voter.

The Number Game

Now that it is established that voter identity cannot be termed as right wing or left wing, it is imperative to verify the data that is employed to depict “Brahman” vote. The author has used archaic data from the last cast based census of 1931 to account for the caste composition of various constituencies across India. All other factors which have influenced population dynamics in the past 87 years have been completely ignored. One of such significant factors is inter-state migration. The 2011 census shows that there are 454 million migrants in the country — that makes 139 million (13.9 crore) internal migrants between 2001 and 2011[2] . While we may assume that many Brahman dominated areas as indicated in the 1931 census still retain a Brahman population, the extent till which these statistics can accurately account for the current population composition is debatable (owing to the large number of internal migrations). Besides migration, the fact that all voters are not Hindu voters has also been pushed into oblivion. It is a big possibility that the vote-share coming from voters of other religions such as Sikhism or Buddhism, or for that matter, voters from non religious backgrounds, may have contributed to the rise of the BJP.

The research paper makes clear the need for the author to have a better understanding of dates and timelines. Her main claim throughout remains that “places with greater Brahman Dominance in education in 1931 were associated with a larger increase in right-wing vote share in state elections held after 1990, a relationship that did not exist in elections held prior to the threat of affirmative action.”

As the author herself has declared, she uses 1990 as an inflection point and particularly centres her arguments around the early 90s. Yet, it seems that when she does not find data that fits into her frame of argument, she uses the results of the National Election Survey of 2004, which took place 14 years after the introduction of reservations. The following figure has been taken from this analysis:

Studying this data further, it is evident that a considerable proportion of voters for the right wing come from a non-Brahman background. And although the author claims that Brahman vote-share in “high brahman dominance” areas is higher, the same is the case for all other categories of castes as well. Other upper castes, STs, SCs and OBCs are also likely to vote more for the right wing in areas with a historically higher Brahman dominance. Furthermore, there has been an attempt to conceal information from the reader — the statistics for other parties have not been displayed. Which other parties were these castes voting for, and in what proportion? If their votes did not go to the BJP, then were they in favour of a party like the Shiv Sena or perhaps the Congress? The research paper leaves no hint for the readers to pick on.

The Jan Sangh and the Inception of the Bharatiya Janata Party (BJP)

The fluctuating timeline used in Suryanarayan’s statistics also finds a parallel in her comprehension of the BJP’s history. Where the Bharatiya Jan Sangh and the BJP should be studied as separate political entities that did not overlap each other’s lifespans, she has used them interchangeably. She states that the BJP had “traditionally targeted conservative hindu voters by focusing on cultural issues like …. opposition to the Hindu Code Bill”. Let’s get the facts straight here — the Hindu Code Bills were a series of Laws passed in the 1950s, but the BJP was only formed in 1980. Even if we are to include the Jan Sangh in our analysis, (since the formation of the BJP was initiated by former leaders of the Jan Sangh), any rational person reading Suryanarayn’s paper can argue that there is no evidence of a backlash to reservations impacting the Jan Sangh’s performance till its dissolution in 1977 (ST and SC reservations have been in India since 1951). The party could never mobilise large-scale support as the Congress did. So why was the Brahman population passive to the idea of reservations back then? And how did it suddenly become the main contributor to the BJP’s popularity after OBC reservations? The author’s line of argument does not make much sense.

Advani’s Hardline Hindutva and the Kashmir Exodus

Perhaps one of the most significant reasons for the BJP’s increase in vote-share after 1990 is its Hindutva appeal. In 1984, the party secured only 2 seats in the Lok Sabha. Two years later, L. K. Advani took over as party president and became a fervent advocate of hardline Hindutva. During the early 1980s, the Vishwa Hindu Parishad had initiated a campaign for the construction of a Ram Temple in Ayodhya, and under Advani, the BJP officially extended its support to this cause, making it a part of its political campaign. The result was a remarkable rise in the BJP’s popularity — the party secured 86 seats in the 1989 general elections. At this point, Suryanarayan believes that Brahman vote-share for the right wing rose because their community is more likely to benefit from the hierarchical nature of caste divisions perpetuated by reservations. But in fact, traditional Brahmins associate strongly with being Hindu, and therefore are more likely to subscribe to the idea of Hindutva.

Apart from the Ram Mandir movement, a pressing factor which could have impacted the BJP’s election results of 1991 is the Kashmir Insurgency. While Kashmiri Hindus had been victims of violence and oppression since the separation of India and Pakistan, the conflict was intensified when the Jammu and Kashmir Liberation Front (JKLF) assassinated Pandit Tika Lal Taploo, a prominent BJP Leader, in front of several eyewitnesses.[3] In January of 1990, two Srinagar based newspapers Aftab and Al-safa published a message threatening all Hindus to leave the Kashmir Valley[4]. And finally, on the fateful night of 19th January 1990, mosques broadcasted messages against Kashmiri Hindus and the idea of India. As a result, approximately 400,000 Hindus living in Kashmir[5], mostly Kashmiri Pandits, were forced to abandon their homes. Therefore in this period, violence against the Hindu community had reached its peak and is bound to have garnered sympathy for the BJP from Hindu voters.

So if evidence points to the conclusion that the BJP would have been most likely to receive Hindu votes, who did the Muslims, Dalits, and other minority communities vote for?

The Rise of Regional Parties

As result of the politics that ensued in the 1990s, especially after the report by the Mandal Commission the rise of parties such as the BSP and the SP, was among the major changes in the voting dynamic in the period. The BSP was primaily supported by Dalits, while the SP had united the Yadav and Muslim vote. In fact, on 30th October 1990, at the peak of the Ram Mandir movement, Mulayam Singh Yadav, then Chief Minister of UP, had ordered the police forces to open fire on kar sevaks, which was seen as an attempt to gain significant support from the Muslim community.

Taking a look at the data below, it can be observed that even though the BJP’s vote-share soared after 1990, it gradually receded to smaller figures compared to that of new political entrants like the BSP and the SP, who seemed to be the real gainers (in terms of rise of vote share percentage) from the political environment succeeding the Mandal Commission.

Party wise vote-share for Uttar Pradesh

(Samajwadi Party was formed in 1992 after the fragmentation of the Janata Dal; the 1989 and 1991 statistics are representative of the JD’s vote-share)

The BJP’s vote-share remained relatively the same from 1991 till 1996, after which it began to fall, while the BSP’s and SP’s vote-shares went a notch higher every subsequent election.[6][7][8][9][10] This comparison is essential in understanding that it was not solely the BJP that attained popularity after the implementation of the Mandal Commission report — caste-based parties had more to gain from the division of castes and sub-castes. The OBC categories which may have previously aligned with the BJP, now had a party that catered exclusively to their identity, and the result was an increasing consolidation of power in the hands of regional parties like the BSP and SP.

In conclusion, Pavithra Suryanarayan seems to have overestimated the impact of the backlash induced by OBC reservations on the rise of the BJP. Major political developments including the Ram Janma-bhoomi movement, the Kashmir exodus and the inception of cast-based parties like the BSP and the SP exert a considerable influence on the political dynamics of the 90s. The author’s use of the 1931 census as the foundation of her data analysis, along with her aforementioned statistical inconsistencies, raises serious doubts on the credibility of her paper.

Not only has she based her paper on the flawed concept of right and left wing in the context of Indian politics, but she has also chosen to represent social phenomena such as ethnic fragmentation purely through numbers without explaining the political and social events that lead to its fall or rise. Numbers without context are ultimately unwanted noise; they must be complimented with analysis and a frame of reference.

Most importantly, the author needs to be cognizant of the fact that in post independent India, the only caste-based categories that can be validated by law are the SC/ST/OBC categories. To further categorise voters into “Brahmans” and a detailed list of “other upper castes” only perpetuates divide in society. Journalists, researchers, and academicians must have a proper understanding of data and present it accurately for what it is, rather than to extrapolate far fetched conclusions that are far from the truth.


Author’s Biases

Ambika Pandey is a Research Intern at the Indian Youth Economic Association. She considers herself a nationalist and leans towards the BJP