A lot of quick explanations have been thrown around to explain the results of the recent Maharashtra assembly elections: the continued effect of the Modi wave, the anti-incumbency effect, the snapping of alliances, the engineering of defections before the campaign. But to measure the impact of these assertions, it is necessary to back them with data.

While the Modi wave cannot be measured due to the sweeping character of the assertion, we can provide data on some other indicators of the Bharatiya Janata Party’s success and of other parties’ fortunes and misfortunes.

We can also look as the state of representation of women and minorities to see whether these elections mark a departure or a continuation from previous trends.

1.     Candidates’ information and deposits

 Elections in India have attracted larger numbers of candidates over the years. A large part of this rise is due to the increase in the number of independent candidates and of micro-parties. Despite the rise, however, the number of effective candidates and of parties represented in the assembly has been stable over time.

Another way to measure the effectiveness of candidates is to look at the number of candidates who forfeit their deposits ‒ meaning those who have gained less than one-sixth of the total votes polled in their constituency.



These elections confirmed the trend of the increase in the total number of candidates, a trend that was also visible in the late 1980s and the 1990s. It reiterated that the higher the number of candidates, the higher the ratio of candidates losing their deposit: around 90% in the 2014 elections.

Most of the candidates forfeiting their deposits are independents or micro-party candidates. If we break down the data among the main parties, we observe some variations. This election, 53% of the Congress candidates and 57% of the Nationalist Congress Party candidates lost their deposits, which is an indicator of strong rejection from voters. Similarly, 45,6% of Shiv Sena candidates lost their deposits. Only 18.6% of the BJP forfeited their deposits, which shows that even losing BJP candidates did fairly well in terms of vote share, compared to other losing candidates.



If we look at the socio-demographics of the deposit losers among main parties, we observe that women candidates tend to lose more their deposit than men, particularly in the Congress and the Shiv Sena. We also observe that candidates with criminal charges tend to keep their deposit more than “clean” candidates, which reinforces the idea that tainted candidates make for stronger contestants.



2. Incumbents and seat retention.

The second striking feature of Indian politics is the high turnover of MLAs at every election. It is known that voters in India tend to reject their government, a phenomenon that is called anti-incumbency. Government incumbency or anti-incumbency can actually hide greater variations at the candidate level. Even great victories can hide many individual defeats, as the net result of a party’s tally is the difference between gains and losses of seats between two elections.

This phenomenon has a significant political impact: the expected political life expectancy of MLAs can affect the way they behave once elected. The need to retrieve the investment required to win a seat can be more pressingly felt when the candidate knows that the probability of serving only one term is quite high. How did the incumbent MLAs fare in these elections? How many recontested and how many survived?

The first observation is that most MLAs elected in 2009 ran again in 2014, discarding the notion in this case that the parties in power try to avoid anti-incumbency by fielding new candidates.



The second observation is that incumbent’s performance follows their party’s performance, which isn’t surprising. Only 28% of the sitting Congress MLAs managed to get re-elected, way more than half of the party’s total tally of 42. In comparison, 83% of sitting BJP MLAs were re-elected.

Another way to look at incumbents and party performance is to measure the capacity of parties to retain seats from one election to another. As noted earlier, a party’s gains can hide important losses of seats previously held.



This table reveals that the BJP retained 39 seats it previously held and succeeded in winning seats from all other major parties, including its former ally, the Shiv Sena (13 seats). The Congress could wrest only four seats previously held by the BJP and two from the NCP. It lost most of its seats to BJP candidates (34 out of 51). The Shiv Sena won seats from all major parties except the BJP (only one seat). The good performances of the NCP and the Congress were essentially in constituencies they had previously held, indicating the strong hold that these two parties have over some of the seats they previously won, despite their overall poor performance. This table shows who is exactly contesting against whom. In the present case, the BJP and Congress had more direct confrontations than the regional parties.

3.  Turncoats

As in every election, much has been made of defectors, or politicians who shift party allegiances on the eve of an election. The reasons why a politician would do so are many and obvious. Rival parties may try to lure strong local politicians in areas where they do not have a significant presence or don’t have a strong chance of winning. Sitting MLAs may seek to avoid the anti-incumbency sentiment likely to hurt their government by joining an upcoming party. How important has this factor been in the recent Maharashtra election? How have these defectors – or turncoats - fared? Did it benefit those who left the ranks of Congress and NCP before the boat sank?

Comparing candidates’ lists between 2009 and 2014, it’s immediately clear that the numbers are not impressive. Out of the 1,791 candidates fielded by the main parties (the four main players and those that stood a chance of winning at least a seat), only 68 contested under a different party ticket, out of which 17 were incumbent MLAs. Out of these 68 turncoats, 31 had shifted to the BJP and the Shiv Sena, and 28 had shifted towards the NCP and the Congress. This means that defecting is not necessarily a one-way street.

Seventeen of these 68 turncoats have been elected, that is to say 25%. Not surprisingly, those who made the move towards the BJP or the Shiv Sena (13 of them, including eight incumbent MLAs) were more successful than those who joined the ranks of the former majority (only one).

If we look at movement between the four main parties’ incumbent MLAs, the numbers get even smaller. Three former Congress and five ex-NCP contested under different party tickets. The five ex-NCP and two ex-Congress went to the BJP. Three ex-NCP won and all the Congress defectors lost.

Defectors can help to make a difference in case of tight races or majorities, which clearly was not the case in the present election. The phenomenon of turncoats or defectors remains quite marginal, even though the publicity of defection of sitting MLAs can help a party to have the wind blowing in its direction.

4.     Women’s representation

 Despite a slight decrease in the number of women candidates, Maharashtra voters have elected 20 women MLAs to the new assembly, the highest number ever in Maharashtra. (They were 19 in the 1980 assembly). This a significant increase from 2009 (11 women MLAs). However, the ratio of women legislators remains very low, at 6.94%. Twelve women legislators were elected on a BJP ticket, five from the Congress and three from the NCP. None of the 12 women Shiv Sena candidates (out of a total of 282 candidates) were elected.



Interestingly, women tend to win with higher margins than men, although the difference in numbers between candidates of the two sexes makes it difficult to draw many conclusions from this. There were only two seats that saw women as both winner and runner-up.

The representation of women candidates in Maharashtra elections has been more or less constant since 1991, within the range of 4%-5% of the candidates. The ratio of elected women candidates always stood lower, in the 2%-4% range. This year saw an inversion of these two curves for the first time – the ratio of elected women MLAs being higher than the ratio of women candidates. This is partly due to the BJP’s higher than usual selection of women candidates. But the figures remain on the overall quite low and are not yet indicative of significant change.

 5. Muslims

These elections have finally been marked by the entry of the Hyderabad-based Muslim party All India Majlis-e-Ittehad-ul Muslimeen, which won two seats the first time it contested elections in Maharashtra. However, the performance of AIMIM hides a bleaker picture of the representation of Muslims in the state.



The number of Muslim candidates and MLAs are on two opposite trend line. Even though the number of Muslim candidates has increase over the years (375 this year against 357 in 2009), the number of Muslims elected to the assembly has gradually decline, by one unit every election since 1999 (from 12 in 1999 to nine in 2014). While Muslims constituted 8.51% of the total number of candidates in 2014, they ended up forming 3.13% of the MLAs, which is very low in a state that in which Muslim comprise 13.4% the population, according to the latest census.

One reason for this under-representation is the geographical concentration of Muslim candidates within Muslim-dominated seats. Seven seats with high Muslim populations had eight Muslim candidates or more (though there were 20 Muslim candidates in Nanded South). A situation like this could lead to splitting votes between Muslim candidates. In the end, only two of these eight seats elected a Muslim.

The entry of AIMIM into Maharashtra politics indicates the growing mutual disaffection between Muslim voters and traditional parties. Though recent surveys show that the Congress received the majority of Muslim votes in Maharashtra (52% according to CSDS figures), it scored relatively low in Muslim-dominated constituencies. The party fielded only 17 Muslim candidates, out of which 4 were elected.

Conclusion

Some of the practices that are usually evoked to explain electoral outcomes – engineering defections, candidate’s incumbency – actually play a marginal role in the results. Longer trends – such as the low representation of women and Muslim candidates and legislators, were confirmed, even though there was a bit of progress on the matter of women’s representation. It is unusual for the BJP to lead the race in this regard.

 With inputs from Akshay Barik, Tanuj Bhojwani, Sonal Bobde, Ratul Chowdhury, Kaustubh Khare, Venkat Prasath and Rajkamal Singh.