Analysing historical trends in female candidature and performance in India’s general elections
Now that the dust from the announcement of election results last Thursday has all but settled, I figured it might be worth digging into the numbers to understand how women did in this round of Lok Sabha elections.
Some of this might be old news for you — several news outlets have touted how this is the highest number of female members of parliament ever elected to the lower house. At 78, female members now comprise 14.6% of the Lok Sabha, up from 65 in the 2014 elections.
However, these figures conceal other trends in success rates, candidate characteristics and spatial distributions. Here is an interesting Twitter thread on female candidates from earlier this month that inspired me to dig into the data myself.
This post is largely comprised of line charts and scatter plots over time because I was more interested in noting changes over time over studying facts from this election itself. I also dislike pie charts so you are more likely to see bar charts in places where you would conventionally expect a pie.
I sourced my data from the supremely useful Trivedi Center for Political Data at Ashoka University.
1. The number of female candidates and winners has been consistently rising
This is the figure you have seen already — 78 women were elected to the Lok Sabha out of 713 who contested seats across India (out of a total of 8023 candidates). Out of these 713, about 563 lost their deposit. This indicates that 150 female candidates commanded over 16.67% of the vote share in their respective races.
2. The proportion of female candidates has also been rising
While the number of candidates has been increasing, the proportion of female candidates within the overall pool has also been rising. However, the situation remains far from any semblance of parity. In the 2019 elections, only 8.9% of candidates were women, up from 7.6% in 2014.
3. On average, female candidates have a higher success rate than male candidates
Defining success rate as the percentage of candidates who win a seat out of those who contest, it would seem as if female candidates tend to perform consistently better than their male counterparts. However, the success rate differential has been declining over the years, as the number of female candidates continues to rise with a slow change in number of winners.
However, the vast majority of candidates in Lok Sabha elections end up losing their election deposits, i.e. they garner less than 1/6th of the total valid votes cast in their constituency.
In the 2019 elections, only 1129 out of 8023 candidates did not forfeit their deposits across 543 constituencies. This implies that most electoral races in the country only have two or three candidates with a real shot at winning.
After dropping underperforming candidates from the dataset, the trend is much harder to eyeball. While female candidates still have a higher success rate on average, the differential is far reduced.
This change is driven by the fact that women lose their deposits less frequently on average (Figure A), indicating that male candidates who have no shot at winning are more likely to contest than similar female candidates. This trend makes sense — on average, female candidates tend to be more competent because of the higher barriers to entry for women in politics. Given the high costs, it is unlikely that a female candidate would contest an election if she doesn't think she has a shot at winning. Thus, stronger female candidates self-select into contesting elections. This is further corroborated by the fact that between non-deposit losing male and female candidates, there seems to be little discernible difference in success rate.
4. The gap between cumulative female and male vote share has been declining
Self-explanatory, in line with previous trends.
5. Female participation is not evenly distributed
180 constituencies did not see a single female candidate contesting. That is 33% of Lok Sabha seats that are only contested by men. Furthermore, an overwhelming majority of the remaining constituencies only had 3 or fewer female candidates. For context, the average number of candidates contesting each seat is 14.4. As a voter in a constituency, this equates to having only 2 female candidates to pick from for every 11 male candidates.
6. North Eastern and Southern states seem to have relatively fewer female MPs
This trend also maps onto the percentage of female candidates in these states. The smaller North Eastern states, which have 1–2 Lok Sabha seats each, see no female candidates for these seats either (Figure B).
In Odisha and West Bengal, BJD and TMC contributed significantly to the states’ good performance on this metric due to their policy of handing out 1/3rd of party tickets to female candidates.
7. Most female MPs still come from national parties, though state parties are catching up
Most female candidates contest as independents, however the vast majority of them are unable to get more than 1/6th the vote. After accounting for the forfeiture of deposits, the distribution of candidates across party types is fairly similar to the distribution of final winners.
State and national parties perform equally in terms of party-wise female representation in the Lok Sabha (number of female MPs/total number of MPs). In terms of candidature, national parties outdo state-based parties — 12.5% of national party candidates were female versus 10% of state-based ones.
8. Out of parties that secured more than 10 seats, the Federal Front of state-based parties outperforms in female participation
While the national parties BJP and INC are fairly comparable in female participation, 17 out of the 56 MPs from the Federal Front parties, BJD, TMC, and YSR Congress, are female. BJP allies on the other hand have only 2 female MPs out of 34.
While these figures seem promising in aggregate, we still have a long way to go to achieve parity in female political participation. This election recorded the highest ever levels of female candidature, wins and voter turnout but it the question of whether improvements in these descriptive indicators will amount to substantive change for women’s rights in the public sphere. Simply looking at aggregate numbers does not help us answer this question — female candidates who go on to espouse regressive outlooks do little to improve the status quo, and may actively hurt women’s causes.
This entire piece was written quite hastily so please leave a comment if you think any of the numbers/figures seem wrong or if you have any other questions or concerns!
Figure A
Figure B
