LONDON: African regimes are excessively long-lasting, but two have just ended. In Ghana change happened through an election won by the opposition; in Guinea through a coup. The Ghana result has rightly been welcomed by the international community, mainly because in Africa it is so rare for oppositions to win. They succeed only when ruling parties lack the power to cheat.

The ruling party in Ghana was relatively powerless because the incumbent president had hit his term limit and so was not standing; power relations in Africa are so personalised that this weakened the party’s grip. Another problem was that, atypically, the government did not have big revenues from natural resources. (Oil has been discovered but has not yet come on stream.) Regimes with resource revenues keep taxes low so as not to provoke scrutiny, and then buy support through a patronage system.

In Kenya in 2002 the same combination – no incumbent standing and no resource revenues – enabled the opposition to win. In both the preceding and subsequent elections, the incumbent stood and duly won – implying that term limits and transparency of resource revenues both matter for clean elections.

Guinea was at the other end of the spectrum: no term limit and big resource revenues. When the president died after 24 years in office, the likely prospect was that Guinea would follow Togo where, after the demise of the president-for-life after 38 years in office, his son graciously assumed power in a token election. No wonder that junior army officers sought to pre-empt the evolution of Guinea from dictatorship to absolute monarchy by seizing the moment. The gut-reaction condemnation of the coup by the international community is sanctimonious condescension. The coup in Guinea also followed a pattern: the absence of term limits and prolonged periods of rule increase the chances of a coup.

Clean elections matter for integrity, but do they matter for development? Does electoral accountability actually discipline a government to run the economy better? With Lisa Chauvet I have been analysing 30 years of data on elections in the developing world. To capture how the economy was run we used two measures: a commercial rating for investors, and a rating done by the World Bank. A reasonable objection to each is that while they reflect what international investors and World Bank staff want, there is no reason to expect that they represent what voters want. Indeed, some NGOs have argued for years that they are precisely the policies that citizens do not want.

During the battles over policy conditionality, many governments certainly put up a lot of resistance to them. Yet, lo and behold, when citizens have been given the power of the vote to force governments to attend to their concerns, what they demand is better policies and governance as measured by these two ratings: the discipline of elections brings a significant improvement.

However, once we distinguish elections according to the quality of their conduct, we find that badly conducted elections fail to improve policy. Governments that steal elections are free to continue the old game of plunder.

So, the struggle for clean elections such as Ghana has just had matters not only for integrity, but for prosperity. Unfortunately, it is a struggle that is a long way from being won. The coup in Guinea was not the clean election the country needed, but was it so very bad? It seems to have been welcomed by many citizens.

Coups have a mixed history. The Senegal election of 2000 was one of those rare occasions when an incumbent president conceded defeat. The reason was that the army threatened a coup if he stole it: so credible was the threat that the president actually conceded before all the votes were counted. So the very forces that sanctimony comfortably condemns can sometimes be the most effective ally of democracy.

Of course, coups can take a society from bad to worse. Such was the coup that overthrew Haile Selassie, the absolute monarch of Ethiopia. By the time of his overthrow the emperor was senile: almost any change would have been an improvement, except the one that actually happened. So coups such as the one that has pre-empted absolute monarchy in Guinea are unguided missiles. The challenge is not to suppress them but to harness them as a force for good.—Dawn/The Guardian News Service