Home >Opinion >Online-views >Maharashtra elections: Who will benefit from splintered alliances?

In the Indian electoral landscape, alliances are largely static. It is unlikely that a pair or trio of alliance partners will dramatically alter their seat-sharing arrangement between two consecutive elections. For example, when the Bharatiya Janata Party (BJP) and the Shiv Sena first came together in 1990, the BJP contested 105 seats in the Maharashtra assembly elections, with the Sena contesting the remaining 183. This equation has changed over time, albeit slowly (, with the BJP’s share creeping up to 119 in the 2009 assembly elections.

Now, following Narendra Modi’s spectacular victory in the Lok Sabha elections, the BJP believes that it has much greater potential in Maharashtra than before, and wants a bigger share of seats in the assembly polls. The Sena, weakened no doubt, was unwilling to budge, and any non-linear difference in the number of seats to be given to the BJP was ruled out.

It is a similar story with the rival Congress-Nationalist Congress Party (NCP) alliance. Ever since they first came together for a pre-poll alliance in 2004 (in 1999 they were post-poll partners), the Congress has been the senior partner in Maharashtra, and the number of seats contested by these two parties has followed that arrangement. However, given the historic low that the Congress hit in the Lok Sabha elections earlier this year, the NCP believed that it deserved a greater share of seats than what the Congress was willing to grant. Also, in early September, Election Metrics pointed out (following a regression on data from the 1999 and 2004 elections) that NCP voters possibly contributed more to the NCP-Congress alliance than Congress voters (

So the question arises yet again: How do you recalibrate an alliance? One way is to look at past performance in terms of seats contested. For example, in the 2009 elections, the BJP got 34% of the votes in the 119 seats it contested, while the Shiv Sena got only 29% of the votes in the 160 it contested. While this might have been a bargaining point for the BJP to try and get a bigger share of seats this year, there are several mitigating arguments, including the choice of seats that the BJP contested last time round, and the influence of the Shiv Sena in those seats, which might have contributed to the BJP’s impressive performance.

Thus, even if the Sena were to make allowances for the fact that the BJP deserved more, the evidence from the previous elections would be insufficient to warrant any non-linear change in shares, which is what the BJP wanted.

In this context, the best way to recalibrate an alliance is to temporarily break it. If the BJP and the Sena were to go it alone in these elections, then the data from these elections (unburdened by any transfer of votes) would provide sufficient backing for a recalibrated seat-sharing agreement the next time around. It is a similar case with the NCP and the Congress. Their current alliance is based on data from the 1999 elections, when the Congress was much stronger than the NCP and ended up as the senior partner in the alliance. Fighting these elections separately would give the NCP an opportunity to recalibrate this relationship.

Thus, it is clear that the BJP and the NCP, the junior partners in the two alliances in Maharashtra, were looking for more, and decided to break their respective alliances. The question, however, arises as to why the two alliances broke down with only two days remaining for filing of nominations, and why the announcements of the breaking of both alliances happened on the same day.

The answer lies in game theory.

The situation can be set up as a classic Prisoner’s Dilemma, with the players being the BJP and the NCP, both of whom wanted a recalibration, and thus a temporary break from their respective alliances. Both of them had two options—to cooperate (break their alliance) or defect (keep their alliance intact). The payoff from their respective choices can be seen in the table.

Notice that for both the BJP and the NCP, staying within their respective alliances is the dominant solution. For example, if the NCP decides to break up with the Congress, the BJP would rather stay with the Shiv Sena and come to power with a thumping majority. It is also important to note that if the NCP decides to stick with the Congress, the optimal solution for the BJP is to stick with the Shiv Sena and come to power as a junior partner rather than getting wiped out. The case is similar for the NCP. Irrespective of what the BJP’s private decision is, it is optimal for the NCP to stick with the Congress.

However, if you look at the top left and bottom right cells in the table, it is clear that the bottom right is preferable for both the BJP and the NCP compared with the top left. Thus this set of payoffs represents the classic Prisoner’s Dilemma, where in the absence of collusion, the optimal solution is for both parties to defect (stay within their respective alliances).

Given this set-up, it cannot be ruled out that the two junior partners might have wanted to break their respective alliances in the past, but were unable to do so since they were unable to coordinate among themselves and thus ended up defecting.

This time, though, it is possible that both parties felt more confident of breaking up their alliances (provided the opposing alliance also broke up), and thus kept making regular public statements to the effect that they were seeking a break-up of their respective alliances.

Each such public statement nudged the collective decision of the BJP and the NCP away from (defect, defect) to (cooperate, cooperate).

It is not clear if the BJP and the NCP actually held talks where they agreed to cooperate, but that doesn’t matter. The announcement of both the decisions just two days from the nomination deadline meant that there was no time for either alliance to renegotiate, and thus it was almost impossible for either party to defect after they had respectively announced that they would cooperate.

And thus we have the situation where the four parties are all going up against one another. It is not a good time to be an election forecaster, but the results of the elections will indeed throw up some interesting data for analysis.

Subscribe to Mint Newsletters
* Enter a valid email
* Thank you for subscribing to our newsletter.

Never miss a story! Stay connected and informed with Mint. Download our App Now!!

Edit Profile
My ReadsRedeem a Gift CardLogout