Recent elections that have produced coalition governments in Australia and the UK have prompted claims that Durverger’s Law, a plurality rule electoral system tends to favor a two-parties, is failing. I think claims of Duverger’s death have been exaggerated.

First, as Gary Cox observed in Making Votes Count, Duverger’s Law holds at the district level, not the national level. That a plurality system could produce two parties at the district level, but many more at the national level is perfectly consistent with Durverger’s Law.

Second, its not clear that Australia’s election was that much of an outlier. Examining the district-level returns (where we expect Duverger’s Law to hold) shows that on average there were 2.6 effective parties in its 2009 election in each single-member electoral district. This is only a bit higher than Australia’s long-term average of 2.3 and not that much higher than the US’s long-term average of 2.4.

I think its a pit premature to put a toe tag on Durverger’s Law.

The Economist’s analysis sounds right to me. Vote rigging is pretty common in African elections, but that is not surprising in nascent/quasi-democracies. What seems more important is that Africans, by and large, see democracy as the only legitimate form of government. Even though the practice of democracy often falls short on the continent, it is more common for people to criticise the leaders of their country than their form of government. This was pretty much what I heard at the local government conference I attended in Zambia this week as well in Chatsworth last week. At the end of the day, I agree with The Economist’s optimism.

Three cheers for the Mikocheni Report for starting an election lexicon for Tanzania. Her explanation of “fisadi” made me laugh out loud. Best definition of corruption ever. Can’t wait for the next installment.

First time. Two round system: if no one gets at least 50%, there will be a runoff in July.

Somaliland votes.

Here’s my analysis at Election Guide.

I tend to believe that more transparency in campaign finance is better than less. If people are buying politicians, I want to know who they are and how much they are giving. Strangely, the case for more transparency is not quite as clear in partial democracies, especially those with weak rule of law. The National Democratic Institute has just released a report on Tanzania that explains why this is the case.

Like many partial democracies, Tanzania has a hegemonic party, the Chama Cha Mapinduzi (CCM), that is unlikely to lose an election anytime soon. The CCM recently passed a campaign finance law requiring donors to identify themselves and how much they are donating. On the surface this seems like a good idea, so why do opposition parties not like the law? They believe that the law is designed so that the ruling party can figure out who is donating to the opposition and/or discouraging such donations. Are these fears reasonable? I would say yes as the CCM has a habit of harassing and intimidating those who oppose them. Lesson: don’t assume that what’s good in a consolidated democracy is good in a weak one.