The clause that killed PR
This article does not concern itself withw the pros and cons of proportional representation (PR). I write in my capacity as a student of mathematics, not as a commissioner on the Elections and Boundaries Commission (EBC).
Let me begin by posing a question and I want you to answer it without agonising too much about what the answer should be. Just go with your common sense notion about what you think PR is meant to accomplish.
Party A gets 6000 votes; party B gets 4900; party C gets 4700 and party D gets 4400. How should four aldermen be allocated? I’ve asked dozens of people this question and they’ve all said the same thing—that each party should get one alderman. The reality is that, based on the formula used for the local government elections (LGE) 2013, party A will get all four aldermen, the others none. One may legitimately ask, how come a system that was supposed to give every voter a voice gives nothing to almost three-quarters of them and everything to just over a quarter?
Here’s one taken from the LGE: In one municipality, Party A got 37788, party B got 18331 and party C got 17750 votes. One could sensibly conclude that A should get two aldermen with B and C getting one each. But, no, A got all four with just over half the votes.
What could have caused this apparently unfair allocation to take place? In the original version of the Municipal Corporations Amendment Bill, this would not have happened and the allocation would have been in keeping with the common-sense notion of what PR was all about, that you would get aldermen in proportion to the number of votes you got.
However, some time during the debate in Parliament, probably at the committee stage, some wise guy or gal added a clause that mandated that a party would qualify for consideration for aldermen only if it got at least 25 per cent of the valid votes cast in the municipality. It appears that this was done by persons who did not have a clear understanding of the mathematics and, hence, the implications of such a clause. (I guess that’s what happens when you put lawyers, social/political scientists and politicians to do a mathematician’s job.)
Without doing any analysis, any person with a modicum of mathematical savvy would immediately realise that in a three-party fight, almost half the voters could get no voice, contrary to the stated objective that “every vote will count in every district”.
In a four-party fight, almost three-quarters could have no say. The two examples above bear this out. In the first one, the 25 per cent cutoff is 5000 and neither B, C nor D reached it so they got nothing. In the second example, the cutoff was 18537 (in addition to the three parties, others got a small number of votes) and neither B nor C reached it so A got everything. There are many other examples from the 2013 LGE.
I experimented by doing the allocation of aldermen with and without the clause. (It takes only about five minutes once the voting figures are known.) With the clause, the only municipalities in which the allocation was “proportional” were Chaguanas and Sangre Grande and this only because the three major parties all acquired at least 25 per cent of the votes. Without the clause, everything would have worked out just fine with the allocation being proportional in all 14 municipalities.
I’ve heard it said that the clause was added to prevent parties with a very small percentage of votes from getting an alderman. Sounds like a good idea. Yet no one could provide me with an example of how this could happen because it just can’t. The mathematics will ensure that it doesn’t. So the clause was added to solve a problem that didn’t exist but, in so doing, scuttled the good intentions of PR. We ended up with an allocation that was only slightly better than what would have obtained with a first past the post system. But, as they say, the longest journey begins with the first step, which we have now taken.
The good news is that the problem is very easy to fix. There are several ways to fine-tune the allocation formula. What is certain is that the 25 per cent threshold must go. Indeed, no threshold is necessary. Those who believe otherwise are invited to provide an example which shows how not having one can create an allocation that is less fair than what currently obtains.
As regular readers would know, I often advise the West Indies Cricket Board that they should hire a competent mathematician to oversee our team’s interests in limited-overs competitions. Not having one has caused us grief in many a match. Perhaps the same advice is pertinent to Parliament. It could save thousands of voters from the grief of not getting an alderman to represent them when they have earned the right to one.
• Dr Noel Kalicharan is senior lecturer in computer science at UWI, St Augustine