The Hindu : National / Elections 2004 : Swings, splits, vote shares and seats

Online edition of India's National Newspaper
Saturday, May 01, 2004

National

National - Elections 2004

Swings, splits, vote shares and seats

The four-cornered contest developing in Uttar Pradesh is not likely to yield a decisive victory for any party, says Yogendra Yadav.

There are two ways of analysing the electoral battleground of Uttar Pradesh. One could look at it purely psephologically, in terms of swings and splits and what effects that might produce in converting votes into seats. Or, one could look at it sociologically in terms of the arithmetic of caste and community that plays a crucial role in determining the electoral outcome. It would however be a mistake to see these two as rival interpretations. The psephological approach is not a rival reading of why people vote the way they do. It is meaningless to talk about the swing factor as an alternative to things like the incumbency factor or the caste factor or different issues. Swings and splits are no more than names for the effect produced by the rise or decline in popularity and that of electoral coalitions or their absence.

It is important to understand this, for the British election system that we follow has a very surprising element. There is no direct relationship between the votes secured by a party and the seats won by it. A psephological understanding helps us understand and anticipate this relationship that often defies political commonsense. But such an understanding must be supplemented by a substantive understanding of why people voted the way they did. That is what we need to do if we wish to understand the current electoral context in Uttar Pradesh. Let us begin with the psephological and then turn to the sociological and political to uncover some of the complexities of this State.

A quick look at the results of the last four Lok Sabha elections in the State could help learn the first lesson in psephology: the largest party enjoys a huge advantage in our system, particularly if it is a multi-cornered contest. Beginning with the Ayodhya wave in 1991, the BJP dominated the Lok Sabha elections in U.P., winning about 50 seats or more out of the 80 seats that now fall in the State (excluding the 5 seats that have gone to Uttaranchal). But that was not because the party was very popular in the State. The BJP's share of the popular votes during this period was always around one-third. The highest it touched was 37 per cent in 1998. A similar vote share in a two-party competition could mean a series of electoral defeats for a party in any other State. But in UP, it gave the BJP nearly two-thirds of the seats.

The key to this success of the BJP lies in the fragmentation of the party competition in U.P. of the 1990s. Following 1991, the Congress sunk to a level of around 10 per cent of the vote share. But it was not replaced by any single force. Most of the constituencies of U.P. experienced multi-cornered contests, as different parties took away different slices of the Congress' rainbow coalition. While the BJP secured only 32 to 37 per cent votes, it was always nearly 10 percentage points ahead of its rivals in the State. That was enough to give it a very large number of seats every time. This can be expressed in shorthand by saying that the Index of Opposition Unity (IOU) was very low against the BJP. The IOU measures the proportion of the `opposition' (opposition to any dominant party) votes that are secured by the leading opposition candidate. The remaining votes get `wasted' in our system, for they go to candidates who do not pose a challenge to the dominant party. A low IOU of 35 per cent or so indicates that a majority of the non-BJP votes got wasted in Uttar Pradesh, thus giving the BJP its major chunk of seats.

This situation changed in the Lok Sabha elections of 1999. The main reason was not greater unity among the non-BJP forces (though the Congress gained from an alliance with Ajit Singh's Rashtriya Lok Dal). It was simply a loss of popularity for the BJP. The BJP's vote share plunged to 30 per cent (only 27 per cent in the area that now constitutes U.P.), reducing its gap from the rivals down to just a couple of percentage points. The SP and the BSP gained in votes, but the main advantage in terms of seats went to the SP. The BJP's votes were spread evenly, while the SP's votes were concentrated in some pockets. That is what enabled the SP to win one more seat based on a vote share that was smaller than that of the BJP. The logic of our electoral system now started working to the disadvantage of the BJP.

Since then the BJP has shrunk further. Its disastrous performance in the State Government led to its worst showing in the State Assembly elections in 2002. On its own, the BJP managed to secure only 25 per cent votes. Its allies, including the RLD, brought it another 2 per cent. Once again, the SP won a much higher share of seats based on a similar share of votes as the BJP. The RLD has since left the BJP and joined the SP in the Government and the pre-election alliance. The BJP has tried to make up for this by bringing in Kalyan Singh whose Rashtriya Kranti Dal had acted as a significant spoiler in the Vidhan Sabha elections. Thus the starting point of the Lok Sabha race is not very different from the Vidhan Sabha elections 2003.

That is what worries the BJP. If the Vidhan Sabha election results are simply replicated in the current Lok Sabha elections, the Samajwadi Party would emerge on the top with as many as 39 seats. The BJP may have wiped out some of that advantage with its new alliances, but these gains could be just formal statistics. There are only two ways in which the BJP can improve upon its tally of 1999. One simple way is for the party to gain more votes and thus have a swing in its favour. A mere 2 percentage point swing in its favour could have lifted its tally to 36, thus giving it a gain of 11 seats. That looked like a real possibility in the early phases of the election campaign. Now that does not look like a real possibility and even the BJP does not boast of it. The only other way for the BJP to gain seats is to hope and ensure that its biggest rival does not gain. A mere two percentage points swing in favour of the SP could give it as many as 40 seats and make it a major national player. That would hurt the BJP directly in terms of seats. The BJP would like to avoid that at any cost. Much of the BJP's current electoral strategy is designed to ensure that there is a fragmentation of votes.

This is where the Congress' rise can help the BJP. The Congress is below the threshold of seats in U.P. Starting from about a 14 per cent vote share, the Congress does not stand to gain much seats if it gets a swing of votes in its favour. A swing up to 4 percentage points would give it no additional seat, a 6 percentage point swing would yield only four additional seats. However, the Congress could start converting its votes into seats once it crosses the 20 per cent barrier. The BJP's ideal scenario would be that the Congress stops around 20 per cent, while the other rivals also fall to a similar level, thus giving the BJP a decisive edge. That is of course a lot of optimism. A four-fold contest of the kind that is developing in the State is not likely to yield any decisive victory for any party. The BJP could draw some consolation from that as well.

One note of caution is in order here. The macro picture presented here could be upset by two kinds of micro developments. The BJP may not gain very much from the opposition's fragmentation if the non-BJP voters identify the winnable candidate and rally behind him or her. This happens rarely, as the ordinary voters are not always a good judge, nor is their level of motivation so high to defeat one party. The opposite could also happen. The BJP may not gain many votes, but may use its famous micro-management skills to convert many possible defeats into victories, just as it did in Rajasthan Vidhan Sabha elections. That cannot be ruled out, given the organisational strength and capability of the BJP. But it cannot alter the picture presented here in a fundamental way.