Between November 2013 and May 2014, nearly a billion Indians would have been asked to choose the party that will govern them for the next five years. Five of these elections would have been at the state level and one at the national level.
It is reasonable, therefore, to ask if people are being presented with a fair choice, not in terms of the political parties that they have to choose from but in the very way these choices are set up in any election system.
Consider this: whereas in every walk of life other than their own, political, intellectual and religious leaders urge citizens to be sensible and choose combinations, when it comes to their own turf they urge citizens to do the opposite and choose only them.
Thus, political leaders will ask you to accept diverse intellectual ideas and be tolerant of other faiths; intellectuals will ask you to accept political and religious plurality; and religious leaders will tell you that democracy is a good thing because it encourages diversity.
But this sarva sambhava approach does not apply to them. You are with them or against them; you are a friend or an enemy; you are clever if you are an acolyte and a fool otherwise.
Non-convex preferences
Economists, who deal with goods and services, have regarded such choice options as a problem for a long time. They have always said that rational choice requires people to choose combinations of goods rather than just one or the other.
For example, given a certain amount of income, it makes sense for people to choose a combination of whiskey and soda rather than just whiskey. Or, heaven forbid that the day should come, just soda.
Yet, in religion and politics, you can choose only one or the other. Akbar had recognised the problem when he tried out his Deen-e-Elahi religion which offered a combination of Hindu and Islamic religious practices. It didn’t catch on.
Political parties have never been open to such eclecticism. They have always offered what economists call non-convex or extreme outcomes rather than in-between ones which are called convex outcomes.
This despite the fact that it has been conclusively shown that either/or (that is, non-convex) outcomes are bad for society in far too many ways to be recounted here.
Key question
So here is a question that I want better minds than mine to answer: if non-convexity always leads to market failure in economics (that is, bad economic outcomes) why is it assumed that it doesn’t lead to bad outcomes in politics as well even though the evidence is to the contrary?
The answer is that we have been conditioned to make only non-convex choices in religion and politics. That is why Deen-e-Elahi failed and that is why elections which do not offer voters combinations of candidates and political parties to choose from result in an increasing number of sub-optimal outcomes, whether in India or elsewhere.
True, combinations do emerge after the elections in the form of coalitions. But these coalitions are not what the voter actually wanted; they are coalitions built by those who have been elected.
Thus, I may prefer A to B and elect her. But that does not mean I have conferred the right on her to partner B after the election. That defeats the entire purpose of my vote.
One may argue that pre-election alliances – NDA, UPA and what not – take care of this problem. When I vote for one of the alliance members, I know what combination it entails.
Constituency issues
This is necessary (and good) but is it sufficient?
Not quite, because it is really at the constituency level that non-convexity poses the biggest problem because of what is called seat adjustment.
This in fact forces the voter into what is in reality party discipline by denying those who may prefer candidate X to vote for candidate Y because X’s party A has entered into an agreement with Y’s party, B.
This makes a complete nonsense of voter choice because this is partly how the issue of “None of the Above” (NOTA) has arisen. In any case, NOTA doesn’t solve the problem.
For instance, suppose every candidate is rejected by the voters. Even then, each candidate will receive his or her own vote. This means everyone will be tied with one vote.
This brings up the key issue, however counter-intuitive it may seem: how can we eliminate non-convexity at the constituency level? How can we devise a system that offers voters combinations at the constituency level?
The answer requires that each constituency be represented by at least two legislators simultaneously because, when you look at it, there is no logical reason for having just one. How to devise such a system is the challenge.
Indeed, this is desirable for another reason as well. In 1947 when India’s population was 350 million we had about the same number of seats in the Lok Sabha as now when the population is 1.25 billion.
In contrast, the UK with a population of just 63 million has 650 members. Game, set and match.
-------------------------------------------------------------------------------------------------------------------------------------------------------
T C A Srinivasa-Raghavan will respond to selected questions and comments about this column on Monday, December 2, 2013.
Disclaimer: These are personal views of the writer. They do not necessarily reflect the opinion of www.business-standard.com or the Business Standard newspaper
