The National Democratic Alliance, led by the Bharatiya Janata Party, suffered a crushing defeat in the 2015 Bihar State election. In the 2014 national election, the NDA won 172 out of 243 Assembly constituency (AC) segments. But in the 2015 Bihar election, just 18 months later, the NDA won only 58 ACs. It is useful to think about how this election informs our understanding of the electorate.

Shifting party coalitions

Much has changed in Bihar’s political landscape since 2010, when the Janata Dal (United) was in alliance with the BJP under the NDA banner and went on to sweep the State election, winning 206 out of 243 seats. Since then, Bihar has seen a major reconfiguration of party alliances. Just before the 2014 national election, the BJP and JD (U) split over the choice of Narendra Modi as prime ministerial candidate, with the BJP selecting new coalition partners and the JD(U) running alone.

The NDA won 31 of 40 parliamentary constituencies, the JD(U) won two. Chief Minister Nitish Kumar of JD(U) resigned, taking responsibility for his party’s poor showing even though his party kept control of the State government. Fearing the rise of the BJP and its allies, once bitter foes, Nitish Kumar and Lalu Prasad Yadav (Rashtriya Janata Dal) (RJD), joined forces.

The Grand Alliance (JD(U), RJD and Congress) swept to power in 2015. Its success will ultimately depend upon whether such a disparate coalition is able to effectively govern in Bihar.

Nonetheless, a pre-electoral coalition of parties that greatly dislike each other is surprising to most observers of electoral politics outside India, although these ‘grand’ electoral coalitions have been present in national politics in India for decades.

Two features of the Indian political system make such disparate electoral coalitions feasible. First, most parties are not deeply ideological, although they may have discernible social or caste bases. In the West, a ‘leftist’ socialist party is unlikely to ally with a ‘rightist’ free market party. Second, most parties are controlled by a few individuals at the top, or a single charismatic leader. This means that forming coalitions across parties only requires the buy-in of a small number of people. Both of these conditions were present in the formation of the Grand Alliance (and NDA) in Bihar.

The two models of vikas

From late September to early November, with colleagues Bhanu Joshi and Ashish Ranjan, I travelled to Bihar to observe the electoral campaign. Early on, we realised a debate over models of economic development, or vikas , was emerging. Nitish Kumar is widely credited as turning around Bihar’s economy and strengthening law and order. When he became chief minister in 2005, he massively expanded State spending. The Reserve Bank of India estimates that Bihar was the only State by fiscal year 2013-14 where the sum of development expenditure, social sector expenditure, and capital outlays exceeded 40 per cent of the gross State domestic product (GSDP), whereas the average among non-special category States in India stood at a little less than 24 per cent.

In contrast, Prime Minister Narendra Modi’s ‘Gujarat model’ is associated with a smaller government that prioritises private investment and job creation. In our interviews, we found Modi’s model quite popular with many young voters who seemed frustrated with the lack of jobs and new industries in Bihar.

It is often wrongly assumed that voters cannot comprehend the impact of economic policy choices. However, Bihar has a very mobile population due to labour migration across India and abroad, and people are able to observe very different models of development and their consequences. While many States in India have urbanised at a rapid rate in recent decades, Bihar remains the second least urbanised State in India. Joblessness, low private investment, and weak urbanisation are just as observable to voters as State benefits.

For all the talk of the role of caste and religion in politics, economic policy is rapidly emerging as the most important axis along which parties and politicians are differentiating themselves. In the 2014 national election, for instance, Modi came to power on the promise of replicating the Gujarat model across India in contrast to the previous Congress-led government, which prioritised centrally-sponsored benefits like those under the Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA).

The campaign counts

Scholars distinguish between two types of party appeals to voters, those that target a party’s committed core voters and those that target floating voters not committed to any party. When one party has the numbers in its favour, the optimal strategy is to appeal to its core base of voters. In volatile, undecided elections with many floating voters, as in the Bihar 2015 election, the optimal strategy is to convince enough floating voters to win the election.

Based on field observations, we opined that promises of vikas were most attractive to floating voters. Had the NDA argued its side of the vikas debate, it could have won this election. Too often, however, it went off script, talking about the appropriateness of eating beef and rethinking caste reservation. The NDA’s wayward campaign strategy effectively handed the election to the Grand Alliance.

We noticed a perceptible shift away from NDA towards the Grand Alliance during the electoral campaign. Surveys from the Centre for the Study of Developing Societies picked up an 8 percentage point shift away from NDA to the Grand Alliance over the campaign period. Such large, decisive shifts in the electorate toward one party during the electoral campaign are quite unlike in the West, where the impact of electoral campaigns is typically thought to be small and fleeting.

In the West, most voters are believed to base voting decisions on underlying ideological preferences and existing ideological connections to parties, making the outcomes of elections far more predictable. In India, voters use the electoral campaign to assess a party’s promises, the capacity to deliver on those promises, and the strength of the party organisation. This yields large shifts during the campaign. It also suggests that analysts of Indian elections should pay more attention to how parties conduct their campaigns, rather than simply focusing on structural factors such as caste and religion.

The Bihar election provides an important snapshot of larger trends in the Indian political system. Electoral volatility has now become a fundamental characteristic of the Indian political system, with shifting party coalitions and changes in electoral support over the campaign. The Indian electorate is increasingly deciding between economic policies that will shape India’s future, and much more effort is needed to understand how voters make these decisions in such a volatile environment.

The writer is a Fellow at the Centre for Policy Research, New Delhi. This article by special arrangement with the Centre for Advanced Study of India, University of Pennsylvania

comment COMMENT NOW