The Karnataka elections have resulted in politicians, hungry for power, trying each combination of the various post-poll permutations possible after an election.
It is entertaining to watch from afar – but it also makes one wonder if the results betray a breakdown in the transmission efficiency of electoral democracy as it is practiced in India.
In a two-party system, an election where the number of seats won don’t match with the popular vote share is generally attributed to ‘gerrymandering’. This is a process where those in power redraw the boundaries of each constituency such that they end with greater number of seats despite winning a lower share of the popular vote. The American voting process, particularly in the last decade or so, has been hopelessly gerrymandered. That touchstone of pop-culture for bringing arcane things to late night comedy, Last Week Tonight with John Oliver, dedicated an entire episode to this phenomenon last year.
Two important tactics that American politicians who are redistricting the electoral map resort to are cracking and packing. Cracking is diffusing voter strength of one’s opposing party throughout the state so that they don’t have sufficient votes to win specific seats despite having a large number of voters state-wide. And packing is when the opposite happens, that is, concentrating the voters of the opposing party into one constituency so that the additional votes don’t get a new elected representative but are wasted in electing someone who’d have won anyway.
India is not a two party system. And politicians do not get to directly redraw the maps of constituencies either. But it’s interesting nevertheless to see that in Karnataka the party which won a greater vote share at the state level, the Congress at 37.7%, did not end up winning the greatest number of seats. Instead the BJP, which won 36.3% of the vote, won the maximum number of seats.
In the language of the Americans, that’s cracked. Except this cracking happened organically. Whereas the BJP managed to find a better balance in terms of neither being too diffused nor being confined to a small pocket.
One way of understanding the degree of cracking or diffusion of vote base is to look at how many constituencies each party came in either first or second places instead of just looking at the winner. A party that comes third, one can assume, had no chance of winning while the party that came in second was in the race. It’s no surprise that, measured thus, the Congress has a greater footprint.
A larger footprint and vote share, however, shouldn’t mean the Congress should automatically have won greater number of seats. Individual constituencies are right to think of themselves as the base unit of political discourse, just as the states of India are right to think of themselves as a unit of political discourse when it comes to federal policymaking.
Another interesting data point in the Karnataka results is that median margin of victory for BJP was 15,433; for the Congress it’s 10,398. Again, it confirms the intuition that BJP’s votes are more concentrated while the Congress’s are diffused.
The natural question to wonder about here is, would the result have been very different had the elections been held under the alternate voting (AV) method instead of first-past-the-post (FPTP).
There isn’t enough data to calculate that exactly. But if one assumes the second and third preferences of people are equally likely among the two other parties of the top three contestants, it’s only natural that the party with a greater footprint would have won greater number of seats. Is that a good enough reason therefore to think it’s alright for the Congress to keep the BJP away? Maybe. Maybe, not.
But whatever that outcome would have been under alternate voting, it’d have been a closer representation of people’s will than the current result; almost by definition. Given that there is a moral argument to be made there and political gain to be enjoyed, it’s odd that the Congress does not bat for alternate voting.
Lines for voting during the recent Karnataka elections. Credit: Reuters
In the current scenario though, a more pressing question is: do we have a fair system when it comes to recording the votes of elected representatives on the floor of the assembly.
Specifically, if an MLA is elected under one party’s symbol should he or she be allowed to vote against the instructions of that party’s whip? The vote could be on legislations or, as in the drama unfolding in Karnataka, on confidence votes that will install and uninstall governments. Should an elected representative of the people be subject to rules of a political party? One can understand why parties want such a law: governing becomes very difficult if the party in power has no idea if it’s going to stay in power or if its legislative agenda will get passed. But why should the rest of us agree entirely to that proposition?
Voters in Karnataka, whatever the pundits want us to believe, did not vote for a chief minister. They voted for an MLA. And more importantly, the Legislative Assembly exists to make laws at the state level. That someone ends up becoming a chief minister is incidental. The question is, can we design a system that’s fair for the intended purpose of such a representative democracy without resorting to direct democracy.
An easy answer is a system where each MLA is given ‘N’ votes to begin with, where N is proportional to the number of seats that the MLA’s party won in the Assembly. And from that point forward, the MLA is allowed to vote or withhold votes and use it strategically. If we have a hung Assembly, as we do in Karnataka, we can imagine this system will expend a lot of votes in the beginning to pick a winner; thereby having very little to govern with and thus not passing any huge pieces of controversial legislation. On the contrary if a party won with a massive majority, we can imagine them effecting changes as they please, which is the case now even if they don’t have a massive majority. Most times, it’s somewhere in-between. Which means individual MLAs can vote their conscience if they truly believed in something with all their remaining votes at one go, regardless of the party they belonged to. Particularly if it’s to stop something the government is attempting to pass and their constituents are deeply opposed to it.
Of course the obvious flaw in such a system is that the ruling establishment could raise the stakes on some obviously inflammatory thing as an opening gambit, make the minority lose all their ‘N’ votes for all MLAs who’d vote against the proposed plan, and then introduce the actual legislation. But even under the current system the majority or the party in power behaves exactly this way. In this suggested method, at least it’s a slight improvement from the existing system which will ultimately break down when gamed to the extreme. But it merely breaks down to what’s now the status-quo!
An obvious next iteration of this is to bring that N and the power to bringing legislations up for voting to people themselves. Why should MLAs or MPs do this and not people themselves when blockchain technologies allow for a safe, secure and verifiable voting is something democracies should be asking themselves. For such a direct democracy to work though, its size and diversity has to be a lot smaller than India. Which again is a question the Indian Union will have to ask itself.
