In a few weeks, Gujarat goes to the polls in what is bound to be one of the most important elections in India since the Emergency elections of 1977. According to many pundits, and the prestigious India Today (IT) opinion poll, the BJP is not only expected to win, but to triumph with 55 percent of the vote and two-thirds majority of seats. Some analysts are talking about the distinct possibility of the next national election being between “leaders” like Narendra Modi and Sonia Gandhi. A Hindu chauvinist and an Italian Christian – it speaks volumes about our democracy but not much about our future.

To go against such conventional, nee universal wisdom, may be appropriate for stock investments but is more than foolish, perhaps plain stupid, when it comes to election forecasting. But I want to provide two major reasons why it is unlikely that the doomsday scenario will come true. The first point is with regard to the fundamentals of any democracy, the second with regard to "fear of polling".

It is rare to find democracies going for extremes. The thrashing that Mr. Pen obtained in France for the presidential elections is the most recent reminder. The reason for this universal phenomenon, which the Indian analysts are not incorporating into their forecasts, is that it is the same red blood that runs through all veins. People around the world, except the rich left elite (who have too much of goods and not enough of guilt), want to improve their material well being. The poor for survival reasons, and the middle class because they want to better the life of themselves, and their children. The proclivity to vote is also the highest among the middle class - the Patels. (The poor also vote but in democratic India there is an identification problem - did the poor cast their vote or was it purchased?).

This middle class is mightily upset at the BJP, and Mr. Modi. In the pogrom of Feb-April, the Muslims lost lives and the Patels lost their businesses. Is there any investor who will choose Gujarat as a destination for her fortune? This applies to all investors - Gujarati, Indian and foreign - and for foreigners, Gujarat is not even on the radar screen. But this state grew at 3.5 percent per capita per annum for the last two decades, and 4.4 percent per annum in the nineties. The national average for both periods has hovered around 3.0 percent. This growth has meant that absolute poverty is below 15 percent in both rural and urban areas of Gujarat. Further, there has been a large decline in poverty, or addition to the lower middle class, or about 20 percent in the last decade.

Modi's near certain election win (as we are told) will mean that red blood has ceased to flow in Gujarat. That the middle class is willing forsake certain material progress for the questionable luxury of protecting "Hinduism" from imagined threats. Is this possible? Theoretically and historically, whether in Gujarat, India or the world, no. Nor does it seem possible according to the IT opinion poll.

In a euphoric commentary accompanying the poll "results", Swapan Dasgupta writes in India Today " A dramatic conclusion is inescapable - Gujarat is reveling in its honeymoon with Modi. Four weeks to polling, the outcome seems foregone". In 1989, Rajiv Gandhi's election was also touted by the same initial opinion polls as a foregone conclusion. The imagined number then was 300 plus seats; he eventually obtained less than 200. At that time, I had offered the hypothesis that Rajiv Gandhi would lose based on the likelihood that the people interviewed were lying i.e. they were stating that they would vote for Mr. Gandhi, when they intended to do just the opposite. The reasons for lying can be many. Today, the all important reason for lying is fear - fear of voting against the "assumed" Hindu majority, fear of being found out that one was not a true "Hindu".

There seems to be considerable evidence in the IT opinion poll about lying - and history repeating itself. There are two signposts. First is the standard question pertaining to "what issues will influence your vote?" A plurality, 24 percent, state that it will be prices, and the second highest reason (only 16 percent) is "the riots of March 2002" . It is curious that prices retain their stranglehold on the minds of the Indian (Gujrati) public - or is it the fear of deflation that voters are responding to? Inflation has been at record lows in India now for some three years (excluding government controlled oil prices in the region of 2 - 3 percent in each of the last three years).

This nonsense about prices being the most important factor is the first give-away that all may not be right with the opinion poll. The second crucial indicator is the standard opinion poll question "will you change your mind between now and polling day?" Normally, no more than 5 to 7 percent reply "yes" and there is little reason for pollsters to believe that this percentage is not distributed evenly among all the parties i.e. it makes little difference to the prediction. However, a surprising 24 percent in the Gujarat opinion poll indicate that they are likely to change their mind - and 7 percent state that they will definitely change their mind!

What is the probability that these uncertain voters (a quarter of the population) will change their mind in favor of the BJP's Modi ? What is the extra bit of information they will obtain about him between now and election day? Have the Patels not seen enough damage to their existence, and to their future welfare? The answer has to be that very few, almost zero, of the voters who are going to change their minds are going to do so in favour of the BJP. This means that the Congress party (it does not really matter who the opposition is, as long as it is human) is likely to gain these 7 percent. Which means that the foregone conclusion is now a neck to neck race - 49 percent (Congress) vs. 48 percent (BJP).

What about the large 17 percent who will "probably" change their mind? It is unlikely that all of these votes will go against the BJP, but two-thirds is a likely prospect. Which means that the Congress increases its vote share by an extra four percent or so, or about 53 percent versus about 44 percent for the BJP. Which is the exact opposite of the forecast in the opinion poll - 55 percent for the BJP, 42 percent for the Congress.

There is no possibility of "hope" in the above analysis. Personally, I believe a Modi win will be a major dent on the secular Indian psyche, something that is both undesirable and should be inconceivable. The reason for the forecast that Modi will lose in Gujarat has nothing to do with hope and everything to do with the likelihood that pre-election opinion polls will contain mostly fear based answers - fear that will lead the voters to lie about voting against Mr. Modi.