Maldives 2013, or how not to elect a president

The Maldives is running something of a clinic on how not to elect president. The first attempt, on 7 September, seemed to be off to a reasonably good start. With an ex-president, Mohammed Rasheed, in the lead with 45.5%, and the runner-up more than twenty percentage points behind, it seemed both like a foregone conclusion and another of those cases that makes one wonder if the two-round absolute-majority rule used in Maldives really is necessary. (To be fair, the 2008 election showed just when a runoff is a good idea: Rasheed won on the second round, when long-time president President Gayoom managed only 40% in the first round, and Nasheed had 24.9%.)

Due to some evidence of irregularities, the election was annulled by the Supreme Court and and a re-vote was ordered for 19 October. This, too, was aborted–before it even got going. Now a re-re-vote is planned for 9 November.

The term of the incumbent, Mohamed Waheed Hassan, ends on 11 November. This is cutting things just a little close, especially if no one wins over 50%. In that case, the runoff would be on 16 November.

In the first round, first attempt, Hassan won only 5% of the vote, thereby offering another entry in the contest for worst showing by an incumbent; unlike some other cases discussed in the linked thread, Hassan is not an elected president. He came to power following a coup that overthrew Rasheed in February, 2012.

13 thoughts on “Maldives 2013, or how not to elect a president

  1. Isn’t it pretty obvious from the facts that you prevented that the people who overthrew Rasheed in the coup are still in charge, and the one thing they will not do is let Rasheed get back into power?

    The electoral difficulties probably follow from that, and not the two round presidential electoral system, which after all lots of other much larger countries such as France and Brazil have been using without difficulty.

  2. I do not know enough about the Maldives, but that certainly is a plausible interpretation.

    As for two-round majority systems, they serve well enough for more established democracies. And they may very well be preferable for semi-presidential systems, where it is advantageous to build an actual majority rather than allow a plurality to suffice. I have written before–both in this “Electing Presidents” block and in my academic writings–about reasons to be skeptical of the system for pure presidential democracies in countries with no established record of competitive party politics.

    I don’t think plurality-elected presidents are necessarily a bad thing, except when the plurality is too small and/or too narrow.

    If more powerful state forces are not willing to see power transferred to certain individuals, then it may indeed be moot.

  3. I can hardly express how much I’d like to see some presidential or semi-presidential countries adopt AV or some other ranked-ballot system…

    • It doesn’t seem to have made much difference in the Maldives’ near-neighbor, Sri Lanka.

      I forget which of the variants Sri Lanka has. It is not AV, but it is a system of ranked-choice ballots.

  4. Sri Lanka has a form of the contingent vote or what in the UK is called the ‘supplementary vote’. Voters fill in their first three preferences and if none has a majority, all but the top two are eliminated. As voters can’t even rank more than three of the candidates, all the while not truly knowing who will be eliminated after the first ’round’, I would say it’s a rather second-rate substitute to AV…

    • Right. Somewhere in the AV/IRV block at this blog there is a discussion of these different forms of rules that could be called “instant runoff”, including a reference to some academic work comparing them.

      If I am not mistaken, contingent vote and supplementary vote are not actually the same (number of preferences?), but share the very bad feature of eliminating all but the top two in one round of counting, and then redistributing.

      I think the academic consensus among those who have studied these various rules is indeed that this “eliminate third and lower, and redistribute” is worse than either two-round majority or FPTP (and, of course, much worse than the sequential-elimination procedure of AV) on various criteria, such as likelihood of selecting a Condorcet winner and risk of selecting a Condorcet loser.

      In Sri Lanka, it has not mattered, because the winner has always had over 50% or close to it.

  5. In fact it is arguably worse than FPTP. It not only combines all the disadvantages of FPTP and preferential voting. It manages to add entirely new disadvantages neither system has.

  6. The contingent vote allows you to express unlimited preferences. The supplementary vote, used I think only for local offices in England, restricts you to first and second references only. Both have only a single elimination round where all but the two leading candidates are eliminated.

    The chance of tactical voting effecting the result is probably greater under either system than under FPTP, but I’ll leave that issue to someone else innumerate. I do not know the history in Sri Lanka, but In England it was very much about advantaging the Conservative/Labour duopoly.

  7. Minneapolis does the opposite of the traditional contingent vote (where voters may rank all the candidates, but all but the top two on primary votes are excluded). There, the count is done in traditional AV style, but voters are restricted to expressing three preferences. I believe they do this because they elect multiple offices on the same ballot paper and cannot fit more than three preferences per office on one double-sided sheet (as rather than writing numbers for preferences, voters darken circles for first preference, second preference and third preference).

    It certainly can deliver better results than contingent vote (such as where the third candidate on first preferences is the overwhelming second choice), but I’d hardly recommend it in favor of true AV.

  8. PNG restricts you to 3 preferences under a system they call Limited Preferential Voting.

    LPV is justified with the argument that a full preferential system would simply be too user-unfriendly in a population where illiteracy remains common. I’m not sure the electors of Minneapolis would be eager to equate their degree of political literacy as the same as that in PNG.

  9. Does the Limited Preferential Vote strengthen the party system in PNG? Would PNG be better off with the STV?

  10. A weird piece of history. The contingent vote’s longest use was in the states of Alabama and Queensland. What is the advantage the elites of sub-tropical, under-developed, racially-divided states see in contingent voting?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s