India’s election system won’t work in the US

During my BTech, I took an elective named Appropriate Technology, which was offered by the Centre for Technology Alternatives in Rural Areas, or CTARA, a bastion of the Gandhian faction. The course only made me sceptical about the concept of appropriate technology, because my suspicion was that the term was just a euphemism for romanticizing low tech jugaad solutions in the garb of being appropriate for the particular rural setting in which it is used. Prof. Date, who taught the course, saying that “farmers can sing while using it” as a point in favour of some contraption did nothing to allay my suspicion. My argument was, yes, we should gauge the appropriateness of the technology for a particular setting, but sometimes the most appropriate technology is not low-tech, but hi-tech that leapfrogs over the path other countries have taken.

I’ll admit though that the EVM is a very good example of the appropriate technology that the AppTech course was claiming to favour. It doesn’t fetishize low-tech and does not adopt hi-tech for the sake of it. Adopts just enough technology to solve the specific constraints that the Indian system faces – stop rigging via booth-capturing, be sturdy enough to work in dusty places without a steady supply of electricity, etc. It doesn’t try to solve problems that don’t need to be solved. It didn’t succumb to the temptation of connecting to the network to make it easier to tabulate votes. The higher tech a machine is, the less secure it is, so the the fact that it is low-tech is a feature, not a bug. In fact, I was uncomfortable with the idea of adding a paper trail to the voting machines, not because I was opposed to paper trails, but the general principle is that adding moving parts, features and inputs or outputs to anything increases the chances of failure or compromise. In the event, the VVPAT enhancement seems to have been done well, so there is no longer a need to object to it.

Now, every time the Americans hold an election, or for that matter every time we hold elections, many misguided people point to the way Americans hold their election and hold it up as an example to emulate. This is deeply stupid for two reasons:

  1. The American system for conducting elections is objectively terrible. It has probably the worst system among democracies, and if it were any worse, it would no longer remain a democracy.
  2. The American system works under a different set of constraints and requirements from the Indian one.

This is why, while it is misguided to say that we must emulate the Americans, it is also misguided to say that the Americans should just outsource their elections to the Election Commission. The Indian system won’t work in the US for the same reason we can’t copy the Americans. Our constraints are very different from theirs.

For example, Indians have good reason to envy the fact that Americans have a much greater ability to vote by mail. Postal ballots do exist in India, but only a small fraction of citizens are eligible to cast their franchise by that method. An expansion of postal ballots in India, however, would prove disastrous. There would be widespread vote-buying and intimidation of voters.

Similarly, American systems seem to provide a better ability for a person to figure out whether his vote has been counted or not. He can send his vote by mail, check online to see if it has been received, and if not, go to the polling station and vote. But before we think of adopting a better audit trail, we must realize that there is a trade-off between the secrecy of the ballot and an audit trail. The United States faces a different trade-off from us. Secrecy is a lot more important in India than in the US. In the US, people are quite open about their voting preferences. More importantly, they can rely on rules that require officials who count the vote to maintain secrecy. That choice would be inadvisable here.

So, while we should certainly envy some of the features of the American systems, we should be careful before translating that envy into imitation. Designs involve trade-offs, and we face a different set.

The most important reason, however, for why the Indian Election Commission would face its Waterloo in the United States is cultural. As a rule, I find that Indians design for control while Americans design for convenience. This is true of not just elections. It is the reason why the US is so reluctant to require PINs while making card purchases, and why we had to learn of the benefits of offering easy returns from the Americans. Of course, the reason for the differing cultural choices is partly the differing trade-offs that our nations face – the US can accept a higher risk of fraud because it has a better legal system that is able to catch more of the fraud that occurs. It is equally true that cultural differences take on lives of their own and exist independent of the underlying reasons that gave rise to them. So an American and an Indian, when faced with a similar set of trade-offs, will make differing design choices, the former favouring convenience while the latter, control. An Indian would find it weird that some Americans can register to vote right on election day, for example.

This is of course a generalization, and there are exceptions – for example, Indian voters are registered by EC officials visiting their homes and taking down their names, which is surely more convenient for those who do get registered that way (not so much for those who get left out). But the generalization is valid enough. In case of elections, Americans should probably learn to be a little more rigid, but in general we would do well to learn better how to design for convenience.

Opinion Polls

Opinion polls have a reputation for inaccuracy in India, but how inaccurate are they in reality? Has a meta-study been conducted on them? The reason I ask is that as I see it, there are two main steps where the polls can go wrong. Either they get the vote percentages wrong, or they make an error in translating the votes into seats.  My ill-informed guess is that the first step is easier to get right than the second. It is possible that they get the vote percentage right and screw up in their model that will translate it into seats, or a small error in the first leads to a large error in the second. It should be possible to test this hypothesis by looking at the data for the last few elections. I am willing to bet (though not a large amount) that the non-shady organizations that conduct the polls get the vote percentage right. (I am also willing to accept the possibility that there are in fact no non-shady pollsters.)

The reason for this speculation is that it occurs to me that if I am right, this time the polls should be very very inaccurate, because the delimitation must have completely screwed up whatever models they have to translate votes into seats.

Replacing FPTP

Karan Thapar discusses former Chief Election Commisioner Lyngdoh’s solution to the problem of “hate speech”.  Lyngdoh believes that hate speech is a consequence of the first past the post system which will often lead to candidates getting elected even with 20-30% of the vote.  So, all a candidate has to do is to appeal to a hardline base, which often means that he can profitably utilize hate speech towards that end. The solution, according to Lyngdoh, is in two parts. The first part is to utilize run-off voting where the top two candidates slug it out for a second round. The second part is to use proportional voting to fill part of the legislature.

I have discussed this when I reviewed Arun Shourie’s book for Pragati, but both these solutions will worsen the problem. In the FPTP system, you have to win the first time. Yes, you can win with 20-30% of the vote, but only if no one else gets more. What is stopping you from appealing to a broader section of the population right now?  In a run-off voting system, you have an incentive to run a two-stage election strategy. In the first stage, your campaign is extremist, focusing on your base. In the second stage, you move to the centre to take advantage of the median voter – something that happens with American Presidential elections. In the FPTP system, you have little incentive to shoot for the second or third place. But in a run-off system, you have an incentive to try to secure 10-15% of the vote, so that you “transfer” it in the second round in return for favours.

Proportional voting has similar problems. In the FPTP system, there is little incentive to appeal to a religion, caste or section that is only 5% strong, but distributed across multiple constitencies. In a proportional system, a party that is focused on just that 5% will still get 5% of the seats.

To be honest, I actually like the proportional system. If you combine it with a directly elected President (making parliamentary majorities irrelevant) the system has some advantages – for one thing, it will provide better representation to the middle class that is now spread across multiple constituencies. But let’s not look at it to solve problems it won’t solve.

Rerun – Popular Will and Divine Will

As you ponder over the results of the elections in the five states, it is time to rerun an old post from over a year back: Popular Will and Divine Will

Essentially, I believe that the first fundamental lacuna of India’s democratic system is that a government’s performance at governance has nothing to do with its performance in the elections. Everyone can explain an election after the results are declared, but no one can predict it in advance. I believe that in India, a statement like “If you do X, your probability of returning to power in the next elections is Y” cannot be made for any values of X or Y. This applies to all X, whether X stands for “good” policies  or populist policies. Neither kind of X will have any kind of cause-and-effect relation on election results. 

The problem is not just the electoral system. It is also because no value of X will translate into any result on the ground. A politician can hatch a scheme whereby he can promise free colour TV to all voters. He may think that voters will get TVs and vote for him, while he gets kickbacks from the manufacturer. But given the corruption in the administrative mechanism, it is pointless to try and put this scheme in action. There is no guarantee that the TVs will reach the voters, and therefore there is no way to ensure that his constituents vote for him. 

Given this reality, if I were a politician, I would basically forget about trying to get reelected and concentrate on making money.

Very few politicians have tried to break out of this cycle, and I believe that the person with the greatest chance of succeeding is Modi.

Urban Predictions

 “…the impact of this [delimitation] commission on India’s politics  will be at least as far reaching as that of the Mandal commission.”  

Now that we have seen the impact of the delimitation and increased representation for urban areas in the Karnataka election results, there will be increased talk of this phenomenon, but I just want to place on record that I had talked of this back in January 2008. I am not sure if I was the first one to talk of this, but just in case I was, let the date be noted.

I must also point out that I am not sure what the impact will be.  I have been speculating quite a bit and much of the speculation will prove to be wrong. I am more likely to be correct about generalities (“national parties will benefit”) than about the particulars (“BJP will benefit”). My reasoning is more likely to be correct than the conclusions, because a small error in reasoning is likely to result in large errors in conclusions.  But I am most confident about the statement made above – i.e. in time, the impact of urbanization on India’s politics will be as large, if not larger than the impact due to the Mandal politics introduced by V P Singh.

What’s New in Karnataka

About the elections in Karnataka, Neel asks:  “Once again. So whats new? Nothing much.”

Ah, but there is something new this time.  Over 30 years back, a constitutional amendment had frozen the map of India’s parliamentary and legislative constituencies to reflect the India of 1971. The moratorium has now ended, and the picture has moved forward to reflect the India of 2001, an India that is much more urban than it was in 1971.  I have reflected briefly on what urbanization means for India’s politics in the January 2008 Pragati 

Karnataka is the first state to go to the polls after the delimitation. Of course, both voters and politicians will take time to adjust to the new situation, but if I am right, this will be the beginning of one of the most significant changes in India’s politics, rivalling the change brought about by VP Singh in 1991.

Gujarat Predictions

Two contradictory predictions about Gujarat:

  1. Modi will win comfortably, because he was never in trouble. It was media bias that made journalists see rebellions where there were mere rumblings.
  2. The exit polls are overestimating Modi’s votes for the same reason they underestimated Mayavati’s votes  – those who voted against him are less likely to say that they did.

Which is your pick?