Indian election result, 2014

Does the dramatic result of the Indian election re-write the fundamental rules of Indian politics? I don’t think so.

One disadvantage of waiting all day to write something about the Indian election is that others beat me to many of the points I was thinking of making. Or is that an advantage? Manuel offered some important perspective on the votes-seats translation in a comment to the earlier thread. Adam Ziegfeld, writing at the Monkey Cage, has made the correct–in my view–point that this election was not as “historic” as many think.

Inevitably, some of what I write here is going to duplicate some points, but maybe offer slightly more detail or different emphasis along the way. Who knows, maybe I’ll even say something original.

All results are preliminary, so I am going by what is on the Electoral Commission website as of around 6:30 AM, Delhi time, 17 May (about 6:00 p.m. Friday my time).

The Bharatiya Janata Party (BJP) is sitting on 282 seats. In further detail, that is 278 won, 4 leading. This is important, because 272 is a majority, and the Commission says it has “won” enough to clear the majority threshold. Congress, the lead party in the incumbent coalition government and the party that has ruled India far more than any other, has 44 seats. This is a stunningly bad result. In fact, it is only seven seats more than the third largest party, the All India Anna Dravida Munnetra Kazhagam (AIADMK), which is a party based solely in the southeastern state of Tamil Nadu.

The media narrative is all about a landslide and a “Modi wave”–the latter referring to BJP PM candidate, Narendra Modi, the current Chief Minister of Gujarat. In terms of seats, one can hardly argue with the outcome being characterized as a landslide. We have a majority party that has 7.6 times the seats of its closest rival. (Mull that over a bit!) And no doubt Modi was a central factor in the campaign and result. However, the media coverage that I have seen does not even mention the voting breakdown (though Ziegfeld, in the post linked above, makes it central to his argument).

The BJP has won 31.1% of the vote (based on preliminary results). Yes, that is 51.9% of the seats on not even a third of the votes. Congress trails far less in the votes than in the seats, with 19.3%. Still a terrible result, but it drives home just how disproportional the electoral system is. In recent elections, disproportionality has not been as high as in many single-seat district plurality systems, but in this election, the result is highly disproportional. That is mainly due to the fact that there was no one clear alternative to the BJP. Congress was discredited by ten years in power marked by recent slowdowns in growth and numerous corruption allegations. Its pre-electoral coalition, the United Progressive Alliance, partially split up, meaning Congress and regional parties did not put up a united front.

On the other hand, an important thing to consider is just how much the BJP’s allies in its National Democratic Alliance (NDA) have contributed to this victory. Obviously, the BJP does not need allies in the Lok Sabha, having a majority on its own. However, it is certain that stand-down agreements it negotiated with its various partners before the election were essential to winning this majority. (I will give one critical example below, but surely there are others as well.) I would imagine there will be a coalition government for this reason. Moreover, assuming Modi and its other leaders are smart (and I think they are), the BJP will not govern as if it has a sole mandate. (It will have to restrain some unsavory radical elements within, and that will be a major challenge.)

I will not pretend to offer a systematic account of the state-level and alliance dynamics, but some things stood out for me in perusing the results.

The BJP made a huge breakthrough in Uttar Pradesh (UP). This does not surprise me, as that state (India’s largest) has multi-party politics. In recent Lok Sabha and state assembly contests the BJP has often been the second or third party in numerous districts that another party won with well under 50%. So it was well positioned to swing a lot of seats its way even if it came well short of a majority of the statewide vote. And that is what happened. In the state, the BJP won 71 of the 80 seats on 42.3% of the vote. The Samajwadi Party, which had won a majority of the state assembly in 2012 on only about 29% of the vote, came second in this election, with 22.2%. But that netted it only 5 of the state’s 80 Lok Sabha seats.

In Maharashtra, the BJP has won 23 of 48 seats on 27.3% of the vote. However, this understates its victory. We can’t overlook the alliance it has with a state-based party, the Shiv Sena. This ally won another 18 seats on 20.6%. So the NDA has 41 seats (over 85%) on 47.9% of the votes. Congress and its in-state ally, the Nationalist Congress Party, combine for only 6 seats; their combined votes were 34.1%–obviously a far better result than the Congress itself nationwide. Many of the seats the BJP has won in the state are due to its not facing competition from a Shiv Sena candidate (and, of course, the reverse is also true, in other districts).

These two state results are important, because they contain a key message for thinking about the broader implications and possible future trends: Alliances remain the key to Indian politics. If the BJP had not had an alliance in Maharashtra, it would not have won a majority in the Lok Sabha. And if the Congress were able to forge an alliance in UP, it could potentially wrest many seats back from the BJP. Making an alliance is something it tried to do on various occasions in recent years, but separate party interests always got in the way. Things might look rather different to the various parties now that they face a strong BJP. The point is that this strength is conditional to a considerable degree on parties’ strategic pre-election behavior.

As for the Modi wave, if it were so strong, then why did his party get only 31%? I just don’t see it. Sure, he was important; the BJP has a higher vote share than it has had before, and much of that is surely do to the strong personal reputation of Modi. But don’t look at this as if it was equivalent to a national presidential election. It was not. Modi benefitted from the disproportionality of the plurality electoral system for parliamentary elections and strategic alliance behavior. The existing practices of Indian politics do not seem to me to have changed fundamentally.

14 thoughts on “Indian election result, 2014

  1. These regional parties, for example the Shiv Sena and Samajwadi, are these parties “feudal” in nature or are they modern ideologically-based parties?


  2. You’ve left unmentioned the fact that the BJP doesn’t have a majority in the upper house (Rajya Sabha) – far from it, in a house of 242 members, they have a mere 46 seats compared with the Congress’s 68. This means that the BJP will need its allies to override an upper house veto on anything other than a money bill.


    • My reading of the constitution is that the bicameral settlement procedure is a joint session. Because the Lok Sabha has more than twice the membership of the Rajya Sabha, the government shouldn’t have too much trouble passing its legislation. But you’re right that the BJP alone currently has only 46 of 245 upper house seats, so they’d need a little help.

      Let’s see: (282+46)/(543+245)=41.6%. They’d be 67 seats short of a bicameral majority if everyone showed up.


  3. Matthew, I guess the next question is why the BJP was able to negotiate all of those stand-down agreements, and what they had to promise to get them done.


  4. JD, you are right, there are a lot of things left unmentioned in the post. That’s why there’s a comment section! The lack of majority in the Rajya Sabha was an obstacle to the outgoing UPA II government, and could be to this new government as well.

    Further reason why the comment section is good to have: Mike has done the calculation on the joint session that I had been wanting to do.

    But let’s modify Mike’s calculations to take account of the alliances: The total number of LS seats for the NDA looks to be around 330 (BJP+Shivsena+TDP, and a few others). I believe they have over 60 in the RS. That puts them a lot closer to a majority, but probably still in need of allies. In the days since the election there was a Hindustan Times article (which I can’t locate now) about the AIADMK (Tamil Nadu) and BJD (Odisha) being likely to extend support in the RS.

    And, yes, I wish we knew what deals are made for the stand-down agreements or for post-poll support. The deals are rarely public (the UPA-Left confidence and supply agreement of 2004 being a rare exception). I read the English-language Indian press almost daily, and while pacts are mentioned all the time, rarely is there discussion of what has been agreed, aside from the stand-down agreements themselves, in the case of the NDA and UPA pre-electoral pacts.


  5. The TDP is a party in Andhra Pradesh, the state that is about to be split. The TDP won 15, and the BJP 2, of the seats in the new Seemandhra state; it looks like TDP will join the cabinet at the center.

    Regarding Pete’s question about the regional parties, I do not know what “feudal” would mean in this context. However, aside from the Left bloc parties, which continued their slide in this election, these are basically all family affairs (as is Congress in terms of its top leadership) and generally non-ideological.

    It is also important to separate the regional parties that are in alliance with one of the national parties (e.g. Shivsena, TDP) and those that never join alliances (e.g. the two UP parties, Samajwadi and BSP), and a residual category of parties that are sometimes in one alliance, sometimes in the other, and sometimes in none (e.g. the Trinamool Congress and AIADMK, which did really well running alone in WB and TN, respectively, in this election).


  6. What would the result had been under proportional representation? Is it good that BJP won’t be too dependent on finicky coalition partners that block reform?


    • We could run the results through some proportional formulas, but that could hardly simulate what the result would have been like if they had really been using PR, which would have changed the game completely. Just for a start, the alliances which now so define Indian politics would have had little raison d’etre and so would not exist full-scale. For somewhat comparable country which actually does use PR, I would look at Indonesia – but if we wanted to speculate on that basis, the (main) problem is that regional parties are actually banned there.


      • Those who have access to academic libraries should check out the article by Adam Ziegield (author of the Monkey Cage post) that attempts to answer the question of how PR would work in India.

        Adam Ziegfeld, “Are Higher-Magnitude Districts Always Better for Small Parties?”, Electoral Studies 32 (2013): 63–77.

        However, my recollection is that he does not attempt to adjust for alliances. It is likely that many parties that currently stand down in some states and districts would not do so under PR (dependent on magnitude, of course), and hence it is impossible to know the vote shares. Even so, it is a really valuable article.


  7. It is ominous to have such a disproportional relationship between actual overall votes and overall seats in the Indian electoral system for parliamentary elections. For one, it absolutely is undemocratic that a party with less than a third (33.33%) of the votes becomes the majority party merely on the accumulation of cleverly positioned seats that are plenty in number, while representing only a few people. For another, it portends the possibility yet of a zealot or bigot coming to power by playing seats intelligently. This is indeed very tragic. Not only for what has happened, but also because the media is ignoring it totally and completely as if to say that the party that won actually represents the choice of the people. People should realize what’s happening and wake up to prevent the recurrence of such a situation in future. The reason as to why they should is beautifully enshrined in Adam Smith’s Statement about the need to work towards one’s self interest (not selfish interest) for any real benefits to society of which one is a part. So the Election Commission of India or the President of India must take it upon themselves with immediacy the task of correcting this disproportional relationship between actual overall votes and overall seats. There are ways to achieve it. One way (as is seen in the US electoral college) is to raise the seats-to-votes ratio in the less densely populated states while lowering the same ratio in the more densely populated states. This ensures the total number of seats in the Lok Sabha still stands at 543, while it requires political parties to work much much harder for getting the seat-majority of 272 For example, if UP had proffered only 60 seats as opposed to the 80 it did, then any party wanting to win would have had to work extra hard on other states for the additional seats required to reach 272. Another way of removing this disproportional relationship would be to reassign the seats-to-votes ratio to states in such a way that the magic seat number of 272 seats for a majority would be reached only with a majority of seats in at least half the number of states. Yet another way would be to require the majority party to both get 272 seats as well as at least 50% of the popular vote.


    • Electoral reform cannot and should not be handed down from on high by the presidency or the electoral commission. Nor should it be directed to artificially inflating the value of some votes at the expense of others. Allow inequality in voting and I assure you that governments will find ways to make sure the inflated votes are their own and the deflated votes belong to the opposition.

      The problem in India is not that the government has a parliamentary majority, but that the parliamentary majority represents nothing like a majority of the electorate. The corrective measure would be a system that more closely reflects the popular vote, not one that moves away from the popular vote.

      A simple PR system South Africa’s would achieve what you want, without this extraordinary business of abandoning equality in voting.


  8. At that point, wouldn’t it just be easier to have proportional representation rather than any sort of complex gerrymandering scheme? Because altering the votes to ensure that single member districts align with the national popular votes is gerrymandering.


  9. Pingback: Delhi assembly: Massive AAP win | Fruits and Votes

  10. Pingback: Bihar 2015: Indian democracy still works | Fruits and Votes

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.