Wednesday 11 October 2017

An Early Prediction For India's 2019 Election

In financial markets, two types of analysis are common. Fundamental Analysis goes into all the factors that affect markets, and attempts to predict them, and thereby their impact on the prices of assets. But Technical Analysis completely ignores fundamentals. Operating under the idea that "price discounts everything", Technical Analysis looks purely at the movements of prices, and attempts to discern patterns from time series data.

I've decided to do some Technical Analysis of the Indian Lok Sabha (parliamentary) elections. I'm going to look at vote shares and seat shares, without worrying about the fundamentals that impact them. We will see that this approach has some merit too.

The Indian parliamentary elections of 2019 are likely to be the biggest in the world, as usual

First off, I noticed a curious fact, i.e., that the sum of the seats of the two main parties, the Congress and the BJP, has been virtually the same in the last two elections, suggesting that the gains of one come mainly at the expense of the other and not any third party.

[In 2009, the Congress had 206 seats to the BJP's 116, for a total of 322. In 2014, the BJP had 282 seats to the Congress's 44, making a total of 326. The strength of the house is 543.]

I then plotted the correlation between the BJP's vote share and its number of seats between 2009 and 2014, interpolating for all the percentages in-between. A screenshot of the spreadsheet is below.

A straightforward analysis with some stark results

In interpreting this spreadsheet, I made two fundamental assumptions.

1. The BJP's vote share of 31% in 2014 is unsustainably high, since it represents a swing of 12% in a single election. It will have to drop in 2019.

2. However, the BJP's vote share is unlikely to drop below its 2009 level of 19%. That represents its core base.

Hence, we have to look at all the vote share percentages from 20% to 30% (inclusive) to assess the likely outcomes.

And two facts are immediately obvious from this analysis:

1. The BJP will not succeed in replicating its absolute majority in 2019. Even a 1% drop in vote share will take its seats tally to 268, which is below the 272 required for an absolute majority. Hence a coalition government is virtually guaranteed even if the BJP does return to power. It will have to be as part of the larger NDA coalition.

2. It is highly unlikely that the Congress-led UPA will return to power. The BJP's vote share will have to fall from 31% all the way down to 22% or less, for the UPA to have a chance at returning to power. In no case will the Congress get an absolute majority on its own.

In spite of these fairly certain outcomes, the power dynamics after the results are out are likely to be very interesting. A lot depends on the personal characteristics of Modi himself. His flexibility, humility and ability to reach out to different groups in a spirit of compromise will be on test. So far, he has shown discomfort in dealing from anything but a position of absolute power.

Those days now look numbered.

No comments: