OPINION | What UP tie-ups portend for BJP in Lok Sabha polls
Uttar Pradesh (UP) was key to the commanding performance of the Bharatiya Janata Party (BJP) in the 2014 national election. The BJP won 71 Lok Sabha seats (73 when one considers its coalition), meaning that more than 25% of all the seats the BJP won in 2014 were in UP.
A similarly commanding performance by the BJP in the 2017 UP state election has obliged former foes Samajwadi Party (SP) and Bahujan Samaj Party (BSP) to join forces to take on the party. Clearly, a good performance in UP will be important for BJP to return to power in 2019. What can we surmise about how this new SP-BSP coalition will affect the BJP’s performance in UP in the upcoming national election?
To grapple with this question, we need to discern how the new SP-BSP combine is likely to perform. It is not a simple matter of adding the vote shares of the SP and BSP. Coalitions can have “miscoordination effects” that cause their vote share to be less than the sum of vote shares of the individual parties that make it up or “coordination effects” that cause its vote share to be more than the sum of its parts. To investigate this issue further, it is instructive to take a closer look at the history of the SP and BSP.
The SP and the BSP have formed a coalition before to stop the BJP. In 1991, the BJP swept to power, in both state and national elections, in UP on promises of building a Ram Mandir in Ayodhya (as well as a firing that led to the deaths of kar sewaks in Ayodhya in October 1990).
After the Babri Masjid in Ayodhya was demolished by kar sewaks in 1992, fearing an ascendant BJP, the SP and BSP joined forces and defeated the BJP in the 1993 state elections.
Interestingly, an analysis of the SP-BSP coalition in 1993 shows significant evidence of coordination effects. The BSP and the SP (then in the Janata Party) had similar results in the 1991 state and national elections, with BSP winning 10% and 11% vote share and SP winning 13% and 11% vote share in the 1991 state and national elections, respectively. The SP-BSP coalition netted 29% vote share in the 1993 state elections, 6-7 percentage points higher than the sum of the individual vote shares of the BSP and SP in 1991.
Coordination effects typically occur because large coalitions create strategic incentives for voters to support them. In 1991, an anti-BJP voter may have selected any number of parties, but he or she knew to vote for the SP-BSP coalition if the BJP was to be defeated in 1993.
Indeed, the BJP vote share was remarkably stable over this period, with an average vote share of 32-33% in each of the elections in 1991 and 1993. It was the Congress that sustained heavy losses, dropping from 18-19% vote share in the elections in 1991 to 15% vote share in 1993, as voters strategically left it for the SP-BSP coalition. These coordination effects are visible even today.
As Roshan Kishore has shown in an analysis in HT, the SP-BSP coalition gained a higher vote share in the recent bypolls of Gorakhpur, Phulpur, and Kairana than the sum of the individual SP and BSP vote shares in these constituencies in the 2014 national election.
Nonetheless, there are major reasons why the SP and BSP may face miscoordination effects. For one thing, the SP-BSP alliance was short-lived, and fell apart in 1995, with BSP leader Mayawati accusing then-SP leader Mulayam Singh Yadav of trying to kill her. Furthermore, the party cadres for the two parties have been competitors for over two decades and they may not be able to work together.
Most importantly, just because parties form a coalition doesn’t mean voters will follow suit. The core voters of the BSP (Jatavs) and the SP (Yadavs) do not like each other very much. Will a Jatav voter really support an SP candidate in her constituency? These kinds of miscoordination can benefit the BJP, as frustrated voters leave the SP-BSP combine to support the BJP.
The biggest difference today is that the BJP, and its National Democratic Alliance (NDA), is a significantly more popular party in UP than it was in the early 1990s — securing 43% vote share in the 2014 national election and 41% vote share in the 2017 state election. This suggests that, just like the 1990s, the BJPs support base is stable. Also, there is less room for the SP-BSP coalition to have big coordination effects, as the Congress is a far weaker party than it was in the 1990s.
So how much will the SP-BSP alliance affect the BJP’s seat share in UP, and how does it vary with levels of coordination/miscoordination by the alliance? In figure, I model the projected number of seats for the BJP using 2014 vote shares for SP and BSP (and small partner Rashtriya Lok Dal) while varying the level of coordination/miscoordination. It is clear that the coalition has consequences for BJP’s seat share.
Even if the SP-BSP coalition lost 8 percentage points vis-a-vis the BJP due to miscoordination, the BJP coalition would receive 55 seats — a big drop from the 73 its coalition received last time. If there is no miscoordination or coordination effect, the BJP coalition receives 36 seats.
The coalition requires a very large coordination effect of 6.3% to get the BJP coalition to drop to 25 seats — suggesting the demise of the BJP in UP might be greatly exaggerated.
Barring a surge in support for Narendra Modi and the BJP, the party’s tally in UP is likely to come down. But, as these analyses show, the BJP’s performance is fundamentally a function of coordination/miscoordination effects in the SP-BSP coalition. We will only get a sense of these effects when the campaign begins in earnest.