Responses: On Popular Will

Ritwik responds to  my previous post. He says that my argument rests on two incorrect assumptions. His first point happens to be the same one Chetan made in the comments – i.e. politicians have a much greater incentive to discern the popular will than pollsters, so how can I extrapolate from the pundits’ inability to call election results to the claim that politicians are similarly unable? This is a valid question. I have replied to this point when I replied to Chetan, but I will expand.

His second “assumption” leaves me puzzled. He says that most policies are made by bureaucrats in any case, so the fact that they are not guided by popular will is moot. But I was referring to a “normative” statement. I was referring to the belief that in a democracy, policies should be guided by popular will. I must say that i myself am not an absolute democrat, and no two people will agree on what powers the government should have in the first place, how many of those should be delegated to political decision-making, how many to civil servants and how many should be the province of the constitution, enforced by the judiciary. My point is that to the extent that you envision some role for democratic decision-making, the failure of popular will to get transmitted to the top is a bad thing.

Now let me come back to the first point. Yes, the opinion polls have problems with their sample sizes,  statistical models, etc. But my belief is that those problems reflect real world problems that even politicians face when they make their estimates.

Yes, the sample sizes are too small. But the appropriate sample size is a function of the variance of the population.  In a homogeonous population, you don’t need that large a sample. The real problem is that the voting decisions are too heterogenous, too dependent on local caste configurations and have too little to do with actual policies of the government they are voting for. In such a situation,  the feedback that top level leaders get from the party workers is useless because it cannot be used to make governance decisions.

7 thoughts on “Responses: On Popular Will

  1. opinion polls, any form of poll which has a decent statistical base assumes that people are firm in their choices. I believe that people either lie to pollsters on their choices, or secondly are indifferent between the parties, and dont make up their mind until actually casting the vote.

    why are exit polls wrong? well either the sample is size is wrong. or else secondly, it could me due to malpractices. it is a simplistic assumption, but then malpractices are rampant in india.

    i have only flimsy excuses for why exit polls are wrong.

  2. Having lived under Governor Raj and also under elected governments, I can say without the slightest doubt that for ordinary people an elected government is many times better than Governor Raj/bureaucracy raj. The only reason elected governments deliver better than extremely smart & educated bureaucrats is that elected governments are much more in touch with the needs of the people. They receive feedback and actually care about it to fortify their future prospects.

    British historians have described the Indian polity as a client-protector system. This has been refuted by many Indian historians but I think it is a good description. The politicians get their feedback through a vast network of client-protector subgroups under them. They provide services to their clients and in return keep their support and maintain their networks. Hence we see the ‘darbars’ and ‘baksheesh’ systems that are still prevalent in most parts of India. It also explains why corruption is tolerated and even preferred by a lot of people even if they are relatively poor because most of them have been co-opted into one of these networks.

  3. Bureaucrats make less and less of important policy decisions these days. Most decisions are in fact ad hoc, many made to meet the demands of ministers, ruling party MLAs, Opposition MLAs, moneyed interests that support the ruling party and the minions of the ministers/ruling party MLAs in that order. Even if the decisions are ad hoc, they do deliver better governance than any planned clever initiative thought up by bureaucrats. Bureaucrats have increasingly become restricted to the prescribed roles of implementation and administration.

    In many ways, the Indian system is becoming a lot like that of America’s except for the lack of an overarching elite, though we can already see the development of such an elite.

    Whether Indian Ad Hoc is better than Chinese Planned, is something that only time will tell.

  4. I think one issue that is neglected in studies of opinion polling is the mindset (or prejudice) of the setter of the questions. After all, by a suitable line of questioning, it is possible to force the respondent to give the answer that the pollster wants to hear. A superb example was given in that brilliant TV series Yes, Minister.

  5. Ravikiran,

    Political parties have their workers right down to the level of polling booths. These workers don’t makes estimates, they know almost exactly how many votes in their polling booth are going to which party. They have the electoral rolls at their disposal and they actually count the votes before they are polled, because they personally know each and every voter. From polling booth level to ward level etc and finally up to the level of a constituency, these numbers are tabulated neatly (for, against, undecided, absent etc). This is how political parties get their internal assessments.

    You are, of course, right in saying that this feedback doesn’t help the national leaders or parties in deciding policies in any meaningful way. But that’s because voting takes place on extremely local issues (except in cases of a “wave”), not because politicians don’t know which way the wind is blowing. Regardless of such feedback, they have to continue making optimistic noises in media in order to (a) keep their flock together, and (b) to influence the genuinely undecided, right till the voting day.

    I agree with your conclusion (democracy being broken) though, because (IMO) it is not affected by the politicians’ ability or the lack of it to call an election correctly.

  6. Their are many voters who are not even aware of the party’s manifesto, heck people do not even know what a manifesto means. Last time I went to vote I saw a female carrying a kid complaining to the official sitting at the polling booth that her kid pressed some button before she could do and she wants to vote again. With this kind of voting, I am not surprised that the complex statistical models are failing.

Comments are closed.